The Maxima Book Paulo Ney de Souza Richard J. Fateman Joel Moses Cliff Yapp 19th September 2004 1 Credits Paulo Ney de Souza Jay Belanger Richard Fateman Joel Moses Cliff Yapp Contents Preface 5 I The Maxima Program and Standard Packages 6 1 Introduction 7 1.1 What is Maxima? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2 A Brief History of Macsyma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2 Available Interfaces to Maxima 10 2.1 The Terminal Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.2 The Emacs Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.2.1 Installing the Maxima Emacs Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.2.2 Maxima-mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.2.3 Enhanced Terminal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.2.4 Emaxima Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3 Xmaxima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.4 TEXmacs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.5 Other Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3 The Basics - What you need to know to operate in Maxima 29 3.1 The Very Beginning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.1.1 Our first Maxima Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.1.2 To Evaluate or Not to Evaluate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1.3 The Concept of Environment - The ev Command . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1.4 Clearing values from the system - the kill command . . . . . . . . . . . . . . . . . . . . . . 38 3.2 Common Operators in Maxima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 3.2.1 Assignment Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4 Trig through Calculus 42 4.1 Trigonometric Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.2 Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 4.3 Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 4.3.1 The assume Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 4.3.2 Definite Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 4.3.3 changevar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 4.3.4 Behind the Black Box - Using Specific Approaches . . . . . . . . . . . . . . . . . . . . . . . 48 4.3.5 Other Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 2 CONTENTS 3 5 Advanced Mathematics - ODEs and Beyond 50 5.1 Ordinary Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.1.1 Defining Ordinary Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.1.2 Solving Ordinary Differential Equations: ode2 . . . . . . . . . . . . . . . . . . . . . . . . . 51 5.1.3 Solving Ordinary Differential Equations: desolve . . . . . . . . . . . . . . . . . . . . . . . 57 6 Matrix Operations and Vectors 60 7 Introduction to Maxima's Programming Language 61 7.1 Some Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 7.2 Unconventional Conditionals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 7.3 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 7.4 Arbitrary Numbers of Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 7.5 Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 7.6 Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 7.7 Serious Business . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 7.8 Hardcopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 7.9 Return to Arrays and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 7.10 More Useful Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 7.11 Part Hacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 7.12 User Representation of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 8 Graphics and Forms of Output 70 8.1 Options on the Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 8.1.1 1D vs. 2D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 8.1.2 TeX Strings as Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 8.1.3 Writing a Session to a File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 8.2 Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 8.2.1 2D function plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 8.2.2 3D Function Plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 8.3 Plot Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 9 Maxims for the Maxima User 80 10 Help Systems and Debugging 82 11 Troubleshooting 83 12 Advanced Examples 84 II External, Additional, and Contributed Packages 87 13 The Concept of Packages - Expanding Maxima's Abilities 88 14 Algebra 89 15 Calculus 90 15.1 asympa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 15.2 pdiff - Positional Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 15.3 qual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 16 Combinatorics 101 17 Differential Equations 102 CONTENTS 4 18 Graphics 103 19 Integequations 104 20 Integration 105 21 Macro 106 22 Matrix 107 23 Numeric 108 24 Physics 109 24.1 dimen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 24.2 dimension - Advanced Dimensional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 24.3 physconst - Definitions for Physical Constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 25 Simplification 119 26 Special Functions 120 27 Sym 121 28 Tensor 122 29 Trigonometry 123 30 Utils 124 31 Vector 125 III Installing, Resources, Misc. 126 32 Installing Maxima 127 32.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 32.2 Source Based Installation on Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 32.2.1 Configure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 32.2.2 Make . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 32.3 Source Based Installation on Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 32.4 Source Based Installation on MacOSX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 List of Figures 129 Index 130 Bibliography 132 CONTENTS 5 Many hands and minds have contributed to Macsyma, in developing it as a research program, and tuning it for use by others. We have enjoyed constructing Macsyma and we hope that you will enjoy using it. We hope you will consider contributing your carefully polished and documented application programs to libraries at your local installation and other sites. Examples 1. First Example 29 2. Quitting Maxima 30 3. End of Entry Characters 31 4. Line Labels 31 5. Labeling an Example Equation 32 6. Evaluation Toggle 33 7. Basic Use of ev Command 33 8. ev's Expand Option 34 9. Float Example ?? Part I The Maxima Program and Standard Packages 6 CHAPTER 1 Introduction 1.1 What is Maxima? Maxima (pronounced mxime1) is a large computer program designed for the manipulation of algebraic expressions. You can use Maxima for manipulation of algebraic expressions involving constants, variables, and functions. It can differentiate, integrate, take limits, solve equations, factor polynomials, expand functions in power series, solve differential equations in closed form, and perform many other operations. It also has a programming language that you can use to extend Maxima's capabilities. The Dangers of Computer Algebra With all this marvelous capability, however, you must bear in mind the limitations inherent in any such tool. Those considering the use of computers to do mathematics, particularly students, must be warned that these systems are no substitute for hands on work with equations and struggling with concepts. These systems do not build your mathematical intuition, nor will they strengthen your core skills. This will matter a great deal down the road, especially to those of you who wish to break new ground in theoretical mathematics and science. Do not use a computer as a substitute for your basic education. By the same token, however, proficiency with computers and computer based mathematics is crucial for attacking the many problems which literally cannot be solved by pencil and paper methods. In many cases problems which would take years by hand can be reduced to seconds by powerful computers. Also, in the course of a long derivation, it is sometimes useful for those who have already mastered the fundamentals to do work in these systems as a guard against careless errors, or a faster means than a table of deriving some particular result. Also, in case of an error, fixing the resulting error can often be much quicker and simpler courtesy of a mathematical notebook, which can be reevaluated with the correct parameters in place. But just as a computer can guard against human error, the human must not trust the computer unquestioningly. All of these systems have limits, and when those limits are reached it is quite possible for bizarre errors to result, or in some cases answers which are actually wrong, to say nothing of the fact that the people who programmed these systems were human, and make mistakes. To illustrate the limits of computer algebra systems, we take the following example: when given the integral Integrate 1/sqrt(2-2*cos(x)) from x=-pi/2 to pi/2, Mathematica 4.1 gives, with no warnings, \!\(2\ Log[4] - 2\ Log[Cos[\[Pi]\/8]] + 2\ Log[Sin[\[Pi]\/8]]\) which N[%] evalutates numerically to give 1.00984. Maxima 5.6 returns the integral unevaluated, the commercial Macsyma says the integral is divergent, and Maple 7 says infinity. (Cite Maxima Email list here.) Had the person who wished to learn the result 1The acronym Maxima is the corruption of the main project name MACSYMA, which stands for Project MAC's SYmbolic MAnipulation System. MAC itself is an acronym, usually cited as meaning Man and Computer or Machine Aided Cognition. The Laboratory for Computer Science at the Massachusetts Institute of Technology was known as Project MAC during the initial development of MACSYMA. The name MACSYMA is now trademarked by Macsyma Inc. 7 CHAPTER 1. INTRODUCTION 8 blindly trusted most of the systems in question, he might have been misled. So remember to think about the results you are given. The computer is not always necessarily right, and even if it gives a correct answer that answer is not necessarily complete. 1.2 A Brief History of Macsyma The birthplace of Macsyma, where much of the original coding took place, was Project MAC at MIT in the late 1960s and earlier 1970s. Project MAC was an MIT research unit, which was folded into the current Laboratory for Computer Science. Research support for Macsyma included the Advanced Research Projects Agency(ARPA), Department of Defense, the US Department of Energy, and other government and private sources. The original idea, first voiced by Marvin Minsky, was to automate the kinds of manipulations done by mathematicians, as a step toward understanding the power of computers to exhibit a kind of intelligent behavior. [?] The undertaking grew out of a previous effort at MITRE Corp called Mathlab, work of Carl Engelman and others, plus the MIT thesis work of Joel Moses on symbolic integration, and the MIT thesis work of William A. Martin. The new effort was dubbed Macsyma - Project MAC's SYmbolic MAnipulator. The original core design was done in July 1968, and coding began in July 1969. This was long before the days of personal computers and cheap memory - initial development was centered around a single computer shared with the Artificial Intelligence laboratory, a DEC PDP-6. This was replaced by newer more powerful machines over the years, and eventually the Mathlab group acquired its own DEC-PDP-10, MIT-ML running the ITS operating system. This machine became a host on the early ARPANET, predecessor to the internet, which helped it gain a wider audience. As the effort grew in scope and ability the general interest it created led to attempts to "port" the code - that is, to take the series of instructions which had been written for one machine and operating system and adapt them to run on another, different system. The earliest such effort was the running of Macsyma in a MacLisp environment on a GE/Honeywell Multics mainframe, another system at MIT. The Multics environment provided essentially unlimited address space, but for various reasons the system was not favored by programmers and the Multics implementation was never popular. The next effort came about when a group at MIT designed and implemented a machine which was based on the notion that hardware support of Lisp would make it possible to overcome problems that inhibited the solution of many interesting problems. The Lisp machine clearly had to support Macsyma, the largest Lisp program of the day, and the effort paid off with probably the best environment for Macsyma to date (although requiring something of an expert perspective). Lisp machines, as well as other special purpose hardware, tended to become slow and expensive compared to off-the-shelf machines built around merchant-semiconductor CPUs, and so the two companies that were spun off from MIT (Symbolics Inc, and LMI) both eventually disappeared. Texas Instruments built a machine called the Explorer bases on the LMI design, but also stopped production. Around 1980, the idea of porting Macsyma began to be more interesting, and the Unix based vaxima distribution, which ran on a Lisp system built at the University of California at Berkeley for VAX UNIX demonstrated that it was both possible and practical to run the software on less expensive systems. (This system, Franz Lisp, was implemented primarily in Lisp with some parts written in C.) Once the code stabilized, the new version opened up porting possibilities, ultimately producing at least six variations on the theme which included Macsyma, Maxima, Paramax/Paramacs, Punimax, Aljbar, and Vaxima. These have followed somewhat different paths, and most were destined to fade into the sunset. The two which survived obscurity, Maxima and Macsyma, we will discuss below. Punimax was actually an offshoot of Maxima - some time around 1994 Bruno Haible (author of clisp) ported maxima to clisp. Due to the legal concerns of Richard Petti, then the owner of the commercial Macsyma, the name was changed to Punimax. It has not seen much activity since the initial port, and although it is still available the ability of the main Maxima distribution to compile on Clisp makes further development of Punimax unlikely. There is a certain surprising aspect in this multiplicity of versions and platforms, given how the code seemed tied to the development environment, which included a unique operating system. Fortunately, Berkeley's building a replica of the Maclisp environment on the MIT-ML PDP-10, using tools available in almost any UNIX/C environment, helped solve this problem. Complicating the matter was the eventual demise of the PDP-10 and Maclisp systems as Common Lisp (resembling lisp-machine lisp), influenced by BBN lisp and researchers at Stanford, Carnegie Mellon University, and Xerox, began to take hold. It seemed sensible to re-target the code to make it compatible with what eventually became the ANSI Common Lisp standard. Since almost everything needed for for Macsyma can be done in ANSI CL, the trend toward standardization made many things simpler. There are a few places where the language is not standardized, in particular connecting to modules written in other languages, but much of the power of the system can be expressed within ANSI CL. It is a trend the Maxima project is planning to carry on, to maintain and expand on this CHAPTER 1. INTRODUCTION 9 flexibility which has emerged. With all these versions, in recent history there are two which have been major players, due this time more to economics than to code quality. 1982 was a watershed year in many respects for Macsyma - it marks clearly the branching of Macsyma into two distinct products, and ultimately gave rise to the events which have made Maxima both possible and desirable. MIT had decided, with the gradual spread of computers throughout the academic world, to put Macsyma on the market commercially, using as a marketing partner the firm of Arthur D. Little, Inc. This version was sold to the Symbolics Inc., which, depending on your perspective, either turned the project into a significant marketing effort to help sell their high-priced lisp machines, or was a diversionary tactic to deny their competitors (LMI) this program. At the same time MIT forced UC Berkeley (Richard Fateman) to withdraw the copies from about 50 sites of the VAX/UNIX and VAX/VMS versions of Macsyma that he had distributed with MIŤs consent, until some agreement could be reached for technology transfer. Symbolics hired some of the MIT staff to work at Symbolics in order to improve the code,which was now proprietary. The MIT-ML PDP-10 also went off the Arpanet in 1983. (Interestingly, the closing of the MIT Lisp and Macsyma efforts was a key reason Richard Stallman decided to form the Free Software Foundation.) Between the high prices, closed source code, and neglecting all platforms in favor of Lisp Machines pressure came to bear on MIT to release another version to accommodate these needs, which they did with some reluctance. The new version was distributed via the National Energy Software Center, and called DOE Macsyma. It had been re-coded in a dialect of lisp written for the VAX at MIT called NIL. There was never a complete implementation. At about the same time a VAX/UNIX version "VAXIMA" was put into the same library by Berkeley. This ran on any of hundreds of machines running the Berkeley version of VAX Unix, and through a UNIX simulator on VMS, on any VAX system. The DOE versions formed the basis of the subsequent non-Symbolics distributions. The code was made available through the National Energy Software Center, which in its attempt to recoup its costs, charged a significant fee ($1-2k?). It provided full source, but in a concession to MIT, did not allow redistribution. This prohibition seems to have been disregarded, and especially so since NESC disappeared. Perhaps it didn't recoup its costs! Among all the new activity centered around DOE Macsyma, Prof. William Schelter began maintaining a version of the code at UT Austin, calling his variation Maxima. He refreshed the NESC version with a common-lisp compatible code version. There were, from the earliest days, other computer algebra systems including Reduce, CAMAL, Mathlab-68, PM, and ALTRAN. More serious competition, however, did not arrive until Maple and Mathematica were released, Maple in 1985 (Cite list of dates) and Mathematica in 1988 (cite wolfram website). These systems were inspired by Macsyma in terms of their capabilities, but they proved to be much better at the challenge of building mind-share. DOEMacsyma, because of the nature of its users and maintainers, never responded to this challenge. Symbolics' successor Macsyma Inc, having lost market share and unable to meet its expenses, was sold in the summer of 1999 after attempts to find endowment and academic buyers failed. (Cite Richard Petti usenet post.) The purchaser withdrew Macsyma from the market and the developers and maintainers of that system dispersed. Mathematica and Maple appeared to have vanquished Macsyma. It was at this point Maxima re-entered the game. Although it was not widely known in the general academic public, W. Schelter had been maintaining and extending his copy of the code ever since 1982. He had decided to see what he could do about distributing it more widely. He attempted to contact the NESC to request permission to distribute derivative works. The duties of the NESC had been assumed in 1991 by the Energy Science and Technology Software Center, which granted him virtually unlimited license to make and distribute derivative works, with some minor export related caveats. It was a significant breakthrough. While Schelter's code had been available for downloading for years, this activity became legal with the release from DOE granted in Oct. 1998, and Maxima began to attract more attention. When the Macsyma company abruptly vanished in 1999, with no warning or explanation, it left their customer base hanging. They began looking for a solution, and some drifted toward Maxima. Dr. Schelter maintained the Maxima system until his untimely death in July, 2001. It was a hard and unexpected blow, but Schelter's obtaining the go-ahead to release the source code saved the project and possibly even the Macsyma system itself. A group of users and developers who had been brought together by the email list for Maxima decided to try and form a working open source project around the Maxima system, rather than let it fade - which is where we are today. CHAPTER 2 Available Interfaces to Maxima Maxima is at heart a command line program, and by itself it is not capable of displaying formatted mathematics beyond the ascii text level. However there are other interfaces which may be used. Maxima has the ability to export expressions using the TEX syntax, and some programs use this device to help with output formatting. (None at this time allow formatted input.) All have their strengths and weaknesses - the choice will likely depend on the skill of the user and the task at hand. We will discuss here all of the interfaces currently available, and the user can make the choice him/herself which one to use. 2.1 The Terminal Interface The terminal interface is the original interface to Maxima. While in some sense all of the interfaces to Maxima could be termed Terminal interfaces, when we refer to it here we mean the command line, no frills interface you would use in an xterm or a nongraphical terminal. It is the least capable of all the alternatives in many respects, but it is also the least demanding. 10 CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 11 How comfortable this interface is depends to some extent on what Lisp you used to compile Maxima originally. If Maxima is started by simply typing maxima on the terminal prompt, the usability depends on which lisp implimentation Maxima is running on. Clisp will have readline support, and most builds of GNU Common Lisp will as well. CMUCL and SBCL, however, will not have this support by default. As a consequence the back arrow will not function and a command history will not be present. A solution to this problem is available on Linux in the form of the program rmaxima. rmaxima is a script that uses the rlwrap utility to add advanced command line features to Maxima when native readline is not available. (Note that rlwrap is not installed by Maxima - it must already be present on the system in question.) The maxima.bat command provides a command line interface on Windows, but this is not recommended for casual users. 2.2 The Emacs Interface A really excellent Emacs mode has been written for Maxima. For those who prefer a non-graphical environment for Maxima, the Emacs mode is the most sophisticated solution available. 2.2.1. Installing the Maxima Emacs Mode The Emaxima package consists of the files maxima.el, emaxima.el, maxima-font-lock.el, emaxima.sty and emaxima.lisp. To install, place the .el files, as well as emaxima.lisp1 somewhere in the load path for Emacs. Finally, if you want to run LATEX on the resulting document, put emaxima.sty somewhere in the TEX inputs path. If you use pdflatex, you'll also need pdfcolmk.sty. Although AucTeX is not strictly necessary, you will most likely find it worth you time to install it, as many of the best features of Emaxima are LaTeX oriented. Copy the .el and .lisp files to the site-lisp directory of your Emacs installation. On a Redhat Linux system, for example, this would be /usr/share/emacs/site-lisp, /usr/local/share/emacs/site-lisp, or some variation thereof. Copy the .sty files to a directory where LaTeX can see them - I've found on Redhat Linux /usr/share/texmf/tex/latex/emaxima works fairly well. Once you have done this, run the command mktexlsr. You should now be almost ready to roll. The last step is to edit your .emacs file. In order to use the enhanced terminal mode, insert the following line: (autoload 'maxima "maxima" "Maxima interaction" t) If you wish to associate files ending in .max with this particular Emacs mode, add this line: (setq auto-mode-alist (cons '("\\.max" . maxima-mode) auto-mode-alist) This will allow you to start Maxima from within Emacs. You can do this one of two ways - either start Emacs and from within it type M-x maxima, or from the command line type emacs -f maxima to have the whole thing work in one step. If you wish to create a desktop icon to start the command line Maxima, simply place this line where they ask you what the name of your program or executable is, and it should work quite smoothly. For Maxima-mode, add the following line to .emacs: (autoload 'maxima-mode "maxima" "Maxima mode" t) The command M-x maxima-mode will start you off here. In the case of Emaxima, the line (autoload 'emaxima-mode "emaxima" "EMaxima" t) should be inserted into your .emacs file. Then typing M-x emaxima-mode will start Emaxima mode. The command M-x emaxima-mark-file-as-emaxima will put the line %-*-EMaxima-*at the beginning of the file, if it isn't there already, and will ensure that the next time the file is opened, it will be in emaxima-mode. This can be done automatically everytime a file is put in emaxima-mode by putting the line 1If Emacs cannot find emaxima.lisp, then the TEX output functions will not work, any attempts to get TEX output will only result in standard output. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 12 (add-hook 'emaxima-mode-hook 'emaxima-mark-file-as-emaxima) somewhere in your .emacs file. 2.2.2. Maxima-mode This mode is fairly basic, and is not dependant on LaTeX. It basically amounts to a text editor which allows you to send lines to Maxima. For moving around in the code, Maxima mode provides the following Maxima mode has the following completions commands: Motion Key Description M-C-a Go to the beginning of the form. M-C-e Go to the end of the form. M-C-b Go to the beginning of the sexp. M-C-f Go to the end of the sexp. Process Key Description C-cC-p Start a Maxima process. C-cC-r Send the region to the Maxima process. C-cC-b Send the buffer to the Maxima process. C-cC-c Send the line to the Maxima process. C-cC-e Send the form to the Maxima process. C-cC-k Kill the Maxima process. C-cC-p Display the Maxima buffer. Completion Key Description M-TAB Complete the Maxima symbol. Comments Key Description C-c ; Comment the region. C-c : Uncomment the region. M-; Insert a short comment. C-c * Insert a comment environment. Indentation Key Description TAB Indent line. M-C-q Indent form. Maxima help Key Description C-c C-d Get help on a (prompted for) subject. C-c C-m Read the manual. C-cC-h Get help with the symbol under point. C-cC-a Get apropos with the symbol under point. Miscellaneous Key Description M-h Mark the form. C-c) Check the region for balanced parentheses. C-c C-) Check the form for balanced parentheses. When something is sent to Maxima, a buffer running an inferior Maxima process will appear. It can also be made to appear by using the command C-c C-p. If an argument is given to a command to send information to Maxima, the CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 13 Figure 2.1: Maxima-mode CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 14 region (buffer, line, form) will first be checked to make sure the parentheses are balanced. The Maxima process can be killed, after asking for confirmation with C-cC-k. To kill without confirmation, give C-cC-k an argument. The behaviour of indent can be changed by the command M-x maxima-change-indent-style. The possibilities are: Standard Standard indentation. Perhaps smart Tries to guess an appropriate indentation, based on open parentheses, "do" loops, etc. A newline will re-indent the current line, then indent the new line an appropriate amount. The default can be set by setting the value of the variable maxima-newline-style to either 'standard or 'perhaps-smart. In all cases, M-x maxima-untab will remove a level of indentation. 2.2.3. Enhanced Terminal Mode For those just want a better terminal session, you can run a regular terminal style session in Emacs. This gives you everything the terminal interface does, plus syntax highlighting, plus more flexibility when editing your commands. If you already have a copy of Emacs open, you can start up the Maxima buffer by typing M-x maxima . If you do not have Emacs running, a shortcut is to start emacs using the following command: emacs -f maxima . In the Maxima process buffer, return will check the line for balanced parentheses, and send line as input. Control return will send the line as input without checking for balanced parentheses. The following commands are also available. M-TAB Complete the Maxima symbol as much as possible, providing a completion buffer if there is more than one possible completion. (If maxima-use-dynamic-complete is non-nil, then instead this will cycle through possible completions. C-M-TAB Complete the input line, based on previous input lines. C-c C-d Get help on a Maxima topic. C-c C-m Bring up the Maxima info manual. C-cC-k Kill the process and the buffer, after asking for confirmation. To kill without confirmation, give C-cC-k an argument. M-p Bring the previous input to the current prompt. M-n Bring the next input to the prompt. M-r Bring the previous input matching a regular expression to the prompt. M-s Bring the next input matching a regular expression to the prompt. 2.2.4. Emaxima Mode Emaxima is a major mode for Emacs that allows the user to insert Maxima sessions and code in a LATEX document. It is based on Dan Dilľs TEX/Mathematica package2, and uses a modified version of William Schelter's maxima.el. Emaxima is an extension of the LATEX mode provided by AUCTEX3, and so has the LATEX mode commands available. The resulting document can be processed by LATEX; this requires putting \usepackage{emaxima} in the preamble. This is in no sense a graphical environment, and the user will not see the benefits of any TeX formatting in real time. This mode is most useful to those who are accustomed to writing documents in LaTeX, and would like to include Maxima sessions in them. This manual itself is a good example of Emaxima in action. 2TEX/Mathematica is available from ftp://chem.bu.edu/pub/tex-mathematica-2.0. 3This can be configured so that Emaxima extends the standard TEX mode provided by Emacs, or just text mode. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 15 Figure 2.2: Enhanced Terminal Mode CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 16 Cells The basic unit of Maxima code in Emaxima is a cell. A cell consists of text between the delimiters \beginmaxima and \endmaxima A cell can be created by typing C-c C-o. (The C-o in this case stands for opening a cell.) The delimiters will then be placed in the buffer, and the point will be placed between them. When working with several cells, you can jump between them by using C-c + to go to the next cell and C-c - to go to the previous cell. Evaluating cells To evaluate the contents of a cell, the command C-cC-uc (emaxima-update-cell)4 will send the contents of the cell to a Maxima process (if there is no Maxima process running, one will be started) and return the results to the cell, separated from the input by the marker \maximaoutput To differentiate sin(x2), for example, type diff(sin(x^2),x); in a cell: \beginmaxima diff(sin(x^2),x); \endmaxima After typing C-c C-u c, it will look like \beginmaxima diff(sin(x^2),x); \maximaoutput 2 2 x COS(x ) \endmaxima To delete the output and return the cell to its original form, you can use the command C-c C-d. If the document is to be TEXed, the above cell will look like: Maxima diff(sin(x^2),x); and the cell with output will look like: Maxima diff(sin(x^2),x); Output 2 2 x COS(x ) Emaxima mode can take advantage of the fact that Maxima can give its output in LATEX form. The command C-c C-u C works the same as C-c C-u c, except now the output is in LATEX form, ready to be formatted by LATEX. In general, if C-c C-u letter returns Maxima output, then C-c C-u capital letter will return the output in TEX form. The above cell would become 4Sending the cells contents to a Maxima process and returning the results is called updating the cell, the prefix C-c C-u will be used to update cells in different ways. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 17 Figure 2.3: An Emaxima Session CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 18 \beginmaxima diff(sin(x^2),x); \maximatexoutput $$ 2\*x\*\cos x^{2} $$ \endmaxima which, when LATEXed, would become Maxima diff(sin(x^2),x); TEX Output 2 x cosx2 (Note that whenever a cell is updated, any old output is discarded and replaced with new output.) The command C-c C-u a will update all of the cells in your document, stopping at each one to ask if you indeed want it updated. Given an argument, C-u C-c C-u a, it will update all of the cells in your document without asking. The command C-c C-u A behaves similarly, except now all the output is returned in LATEX form. Initialization Cells It is possible that you want certain cells evaluated separate from the others; perhaps, for example, you want certain cells evaluated whenever you open the document. This can be done using initialization cells. An initialization cell is delimited by \beginmaxima[* Initialization Cell *] and \endmaxima The command C-c C-t will turn a cell into an initialization cell, applying C-c C-t again will turn it back into a regular cell. When LATEXed, an initialization cell will look like Maxima diff(sin(x^2),x); Initialization cells behave like regular cells, except that they can be treated as a group. To evaluate all initialization cells (without displaying the output in the document buffer), the command C-c C-u t will go to each of the initialization cells and evaluate them. If you want the output of the initialization cells to be brought back to the document buffer, stopping at each one to see it you indeed want it updated, then use the command C-c C-u i. With an argument, C-u C-c C-u i, the initialization cells will be updated without asking. The command C-c C-u I behaves just like C-c C-u i, except that the output is returned in TEX form. Referencing Other Cells Instead of Maxima code, a cell can contain a reference to another cell, and when the original cell is sent to Maxima, the reference is replaced by the referenced celľs contents (but only in the Maxima process buffer, the celľs content in the documenťs buffer is not changed). In order to do this, the original cell must be marked by having a label of the form . (The reason for the filename will become apparent later, and cell label is optional for the referencing cell.) The referenced cell must also be labeled, with the same filename but a unique cell label. To reference the other cell, the original cell need only contain the marker for the referenced cell. For example, given cell 1: CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 19 \beginmaxima diff(f(x),x); \endmaxima and cell 2: \beginmaxima f(x):=sin(x^2); \endmaxima then the result of updating cell 1 (C-c C-u c) will be: \beginmaxima diff(f(x),x); \maximaoutput 2 f(x) := SIN(x ) 2 2 x COS(x ) \endmaxima When LATEXed, the top line will contain a copy of the marker. Maxima diff(f(x),x); Output 2 f(x) := SIN(x ) 2 2 x COS(x ) A cell can contain more than one reference, and referenced cells can themselves contain references. To aid in labelling the cells, the command C-c C-x will prompt for a label name and label the cell. To aid in calling references, the command C-c C-TAB can be used for completing the the filename and cell label parts of a reference, based on the current labels. Another option is to set the Emacs variable emaxima-abbreviations-allowed to t, say, by putting the line (setq emaxima-abbreviations-allowed t) in your .emacs file. This will allow the filename and cell label parts of a reference to be abbreviated by enough of a prefix to uniquely identify it, followed by ellipses ... For example, if there are cells labelled Then the reference <...:le...> will suffice to refer to the second label above. If you want the references in a cell to be replaced by the actual code, the command C-c @ will expand all the references and put the code into a separate buffer (so it will not affect the original document). CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 20 WEB The reason for the ability to reference other cells is so that you can write what Donald Knuth calls literate programs. The idea is that the program is written in a form natural to the author rather than natural to the computer. (Another aspect of Knuth's system is that the code is carefully documented, hence the name "literate programming", but that is done naturally in Emaxima.) Knuth called his original literate programming tool WEB, since, as he puts it, "the structure of a software program may be thought of as a web that is made up of many interconnected pieces." Emaxima's ability in this respect is taken directly from TEX/Mathematica, and is ultimately based on WEB. To create a program, the "base cell" or "package cell" should contain a label of the form (no cell label), and can contain references of the form (same file name as the base cell). As a simple (and rather silly) example, suppose we want to create a program to sum the first n squares. We could start: \beginmaxima squaresum(n) := ( ); \endmaxima We would then need cells \beginmaxima, L:makelist(k,k,1,n), \endmaxima \beginmaxima L:map(square,L), \endmaxima \beginmaxima lsum(k,k,L) \endmaxima and then we would also need: \beginmaxima square(k) := k^2, \endmaxima When TEXed, the header of the cell will say that it determines the file squaresum.mu. Maxima Definition of package squaresum.max squaresum(n) := ( ); The command C-u C-c @ will put all the pieces together in the file it determines. The resulting file, in this case, will be squaresum.max and will look like: squaresum(n) := ( L:makelist(k,k,1,n), CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 21 square(k) := k^2, L:map(square,L), lsum(k,k,L) ); (Although the idea is that only the computer need look at this file.) Other types of cells When a cell is TEXed, the input and output are kept separate. To have the results look like a Maxima session, there are, in addition to the standard cells, special cells called session cells. A session cell is delimited by \beginmaximasession and \endmaximasession The command C-c C-p will create a session cell. When a session cell is updated, the output will be marked off with \maximasession, and will contain both the input and the output, with the Maxima prompts. For example, if the session cell \beginmaximasession diff(sin(x),x); int(cos(x),x); \endmaximasession were updated, the result would look like \beginmaximasession diff(sin(x),x); integrate(cos(x),x); \maximasession (C1)diff(sin(x),x); (D1) COS(x) (C2)integrate(cos(x),x); (D2) SIN(x) \endmaximasession which, when TEXed, would look like (C1)diff(sin(x),x); (D1) COS(x) (C2)integrate(cos(x),x); (D2) SIN(x) If it is updated in TEX form, it will look like \beginmaximasession diff(sin(x),x); integrate(cos(x),x); \maximatexsession \C1. diff(sin(x),x); \\ \D1. \cos x \\ \C2. integrate(cos(x),x); \\ \D2. \sin x \\ \endmaximasession CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 22 which, when TEXed, will look like (%i1) diff(sin(x),x); (%o1) cosx (%i2) integrate(cos(x),x); (%o2) sinx For particularly long output lines inside the \maximatexsession part of a session cell, the command \DD will typeset anything between the command and \\. Unfortunately, to take advantage of this, the output has to be broken up by hand. If a session cell has not been updated, or has no output for some other reason, it will not appear when the document is TEXed. There is one other type of cell, a noshow cell, which can be used to send Maxima a command, but won't appear in the TEXed output. A noshow cell can be created with C-c C-n, and will be delimited by \beginmaximanoshow and \endmaximanoshow Session cells and noshow cells cannot be initialization cells or part of packages.5 If the command to create one type of cell is called while inside another type of cell, the type of cell will be changed. So, for example, the command C-c C-p from inside the cell \beginmaxima diff(x*sin(x),x); \endmaxima will result in \beginmaximasession diff(x*sin(x),x); \endmaximasession If a standard cell is an initialization cell or a package part, its type cannot be changed. Miscellaneous Some Maxima commands can be used even outside of cells. The command C-c C-u l send the current line to a Maxima process, comment out the current line, and insert the Maxima output in the current buffer. The command C-c C-u L will do the same, but return the result in LATEX form. The command C-c C-h will provide information on a prompted for function (like Maxima's describe), and C-c C-i will give the Maxima info manual. Finally, the Maxima process can be killed with C-c C-k. Customizing EMaxima There are a few things that you can do to customize Emaxima. By default, Emaxima is an extension of AUCTEX mode. This can be changed by changing the variable emaxima-use-tex. The possible values are 'auctex, 'tex and nil. Setting emaxima-use-tex (the default) to 'auctex will make Emaxima an extension of AUCTEX, setting it to 'tex will make Emaxima an extension of Emacs's default TEX mode, and setting emaxima-use-tex to nil will make Emaxima an extension of text-mode. So, for example, putting 5That could be changed, but I don't know why iťd be useful. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 23 (setq emaxima-use-tex nil) in your .emacs file will make Emaxima default to an extension of text mode. Whether or not the dots (...) abbreviation is allowed in cell references is controlled by the elisp variable emaxima-abbreviations-al which is set to t by default. Setting this to nil will disallow the abbreviations, but will speed up package assembly. The LATEXed output can also be configured in a couple of ways. The lines that appear around cells when the document is TEXed can be turned off with the command (in the LATEX document) \maximalinesfalse They can be turned back on with the command \maximalinestrue The fonts used to display the Maxima input and output in a cell are by default cmtt10. They can be changed, seperately, by changing the TEX values of \maximainputfont and \maximaoutputfont. So, for example, to use cmtt12 as the input font, use the command \font\maximainputfont = cmtt12 The spacing in the cells can be controlled by changing the TEX variables \maximainputbaselineskip and \maximaoutputbaselinesk and so to increase the space between the lines of the output, the command \maximaoutputbaselineskip = 14pt could be used. The amount of space that appears before a cell can be changed by changing the value of \premaximaspace (by default, 0pt), and that after a cell can be changed by changing the value of \postmaximaspace (by default, 1.5 ex). Session cells can be configured similarly. Lines can be placed around a Maxima session with the command \maximasessionlinestrue and they can be turned back off with \maximasessionlinesfalse The font can be changed by changing the value of \maximasessionfont. The color of the prompts when the session is in TEX form is controlled by \maximapromptcolor, by default red, the colors of the input lines and output lines are controlled by \maximainputcolor and \maximaoutputcolor, respectively. So the command \def\maximainputcolor{green} would make the input in a TEXed session green. The session can be TEXed without the colors by using the command \maximasessionnocolor. The baselineskip is set by \maximasessionbaselineskip for normal session cells, and by \maixmatexsessionbaselineskip for TEX sessions. The amount of space that appears before a session cell can be changed by changing the value of \premaximasessionspace (by default, 0pt), and that after a cell can be changed by changing the value of \postmaximasessionspace (by default, 1.5 ex). CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 24 Emaxima mode commands Key Description C-c C-o Create a (standard) cell. C-c C-p Create a session cell. C-c C-n Create a noshow cell. C-c + Go the the next cell. C-c - Go to the previous cell. C-c C-u a Update all of the cells. With an argument, don't ask before updating. C-c C-u A Update all of the cells in TEX form. With an argument don't ask before updating. C-c C-u t Evaluate all of the initialization cells. C-c C-u i Update all of the initialization cells. With an argument, don't ask before updating. C-c C-u I Update all of the initialization cells in TEX form. With an argument, don't ask before updating. C-c C-u s Update all of the session cells in TEX form. With an argument, don't ask before updating. C-c C-k Kill the current Maxima core - this will lose all data entered into the maxima system up until this point by other cells. Commands only available in cells. Key Description C-c C-v Send the current cell to the Maxima process. C-c C-u c Update the current cell. C-c C-u C Update the current cell in TEX form. C-c C-d Delete the output from the current cell. C-c C-t Toggle whether or not the current cell is an initialization cell. C-c C-x Insert a heading for the cell indicating that iťs part of a package. C-c @ Assemble the references contained in the cell. With an argument, assemble the package that the cell defines. C-c C-TAB Complete a reference within a cell. Commands only available outside of cells. Key Description C-c C-u l Send the current line to Maxima, and replace the line with the Maxima out- put. C-c C-u L Send the current line to Maxima, and replace the line with the Maxima output in TEX form. AUCTEX commands Inserting commands Key Description C-c C-e Insert an environment. C-c C-s Insert a section. C-c ] Close an environment. C-c C-j Insert an item into a list. " Smart quote. $ Smart dollar sign. C-c @ Insert double brace. C-c C-m Insert TEX macro. M-TAB Complete TEX macro. Formatting CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 25 Key Description C-c C-q C-r Format region. C-c C-q C-s Format section. C-c C-q C-e Format environment. C-c . Mark an environment. C-c * Mark a section. Commenting Key Description C-c ; Comment a region. C-u C-c ; Uncomment a region. C-c % Comment a paragraph. C-u C-c % Uncomment a paragraph. Font selection Key Description C-c C-f C-b Bold. C-c C-f C-i Italics. C-c C-f C-r Roman. C-c C-f C-e Emphasized. C-c C-f C-t Typewriter. C-c C-f C-s Slanted. C-c C-f C-d Delete font. C-u C-c C-f Change font. Running TEX (Commands: TeX, TeX Interactive, LaTeX, LaTeX Interactive, SliTeX, View, Print, BibTeX, Index, Check, File, Spell.) Key Description C-c C-c Run a command on the master file. C-c C-r Run a command on the current region. C-c C-b Run a command on the buffer. C-c ` Go to the next error. C-c C-k Kill the TEX process. C-c C-l Center the output buffer. C-c C-^ Switch to the master file. C-c C-w Toggle debug of overful boxes. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 26 2.3 Xmaxima CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 27 Xmaxima is currently the default interface environment on Windows. It is based on Tcl/Tk, and acts largely as an enhanced command prompt. This is probably where most people will start learning Maxima, and it is not a bad place to start. In order to start this interface in Unix type systems simply type xmaxima in a terminal. From there, you should get what looks like the terminal interface in a Tk window, with an introductory html document below it. There is also a pull down menu system. For Windows users, clicking on the icon or menu entry will start this interface. 2.4 TEXmacs TEXmacs is a WYSIWYG scientific editor, and it has the ability to interface with many computer algebra systems, including Maxima. (Note: Currently this is program is only available on Unix type platforms, but porting efforts are underway and in the future it may be possible to use it on Windows.) This program takes advantage of the TEX output Maxima can produce to format iťs output. To launch Maxima inside of TEXmacs you go up to the menu and select Insert -> sessions -> maxima. The functionality from this point on is similar to other Maxima interfaces, with the considerable benefit of well formatted output. This is the most visually appealing way to run Maxima. 2.5 Other Interfaces There are several other Maxima interfaces efforts underway, but all of these are not currently recommended for general use. The Symaxx interface is unmaintained and now several years out of date. There are GTK and WxWidgets based efforts, but while both are functional they will need more development before they are ready for casual use. CHAPTER 2. AVAILABLE INTERFACES TO MAXIMA 28 Figure 2.4: TeXmacs with a Maxima Session CHAPTER 3 The Basics - What you need to know to operate in Maxima Here we will attempt to address universal concepts which you will need to know when using Maxima for a wide variety of tasks. 3.1 The Very Beginning All computer algebra systems have syntactical rules, i.e. a structured language by which the user communicates his/her commands to the system. Without being able to communicate in this language, it is impossible to accomplish anything in such as system. So we will attempt to describe herein the basics. 3.1.1. Our first Maxima Session We will start by demonstrating the ultimate basics: +, -, , and /. These symbols are virtually universal in any mathematical system, and mean exactly what you think they mean. We will demonstrate this, and at the same time introduce you to your first session in Maxima. In the interface of your choice, try the following: GCL (GNU Common Lisp) Version(2.3.8) Wed Sep 5 08:00:22 CDT 2001 Licensed under GNU Library General Public License Contains Enhancements by W. Schelter Maxima 5.5 Wed Sep 5 07:59:43 CDT 2001 (with enhancements by W. Schelter). Licensed under the GNU Public License (see file COPYING) (%i1) 2+2; (%o1) 4 (%i2) 3-1; (%o2) 2 (%i3) 3*4; (%o3) 12 (%i4) 9/3; (%o4) 3 29 CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 30 (%i5) 9/4; (%o5) 9 4 (%i6) quit(); Above is our first example of a Maxima session. We notice already several characteristics of a Maxima session: The startup message, which gives the version of Maxima being used, the date of compilation, which is the day your executable was created, the labels in front of each line, the semicolon at the end of each line, and the way we exit the session. The startup message is not important to the session, but you should take note of what version of Maxima you are using, especially if there is a known problem in an earlier version which might impact what you are trying to do. Exiting a Maxima Session You want to be able to get out of what you get into. So the first command we discuss will be the command that gets you out of Maxima, and while we are at it we will discuss how to get out of debugging mode. Debugging mode is quite useful for some things, and the reader is encouraged to look to later chapters for an in-depth look at the debugging mode, but for now we will stick to basics. As you see above, quit(); is the command which will exit Maxima. This is a bit confusing for new users, but you must type that full command. Simply typing quit or exit will not work, nor will pressing CTRL-C - if you try the latter you will be dumped into the debugging mode. If that happens, simply type :q if you are running GNU Common Lisp, or :a if running CLISP. (If in doubt use :q - most binary packages use GCL at this time.) Here's an example of what not to do, and how to get out of it if you do: (C1) quit; (D1) QUIT (C2) exit; (D2) EXIT (C3) Correctable error: Console interrupt. Signalled by MACSYMA-TOP-LEVEL. If continued: Type :r to resume execution, or :q to quit to top level. Broken at SYSTEM:TERMINAL-INTERRUPT. Type :H for Help. MAXIMA> > Correctable error: Console interrupt. Signalled by SYSTEM:UNIVERSAL-ERROR-HANDLER. If continued: Type :r to resume execution, or :q to quit to top level. Broken at SYSTEM:TERMINAL-INTERRUPT. MAXIMA> > >:q (C3) quit(); The first two lines show what happens if you forget the () or type exit. Nothing major, but you won't exit Maxima. CTRL-C causes a few more problems - if you actually read the message, you will see it tells you how to handle it. In the above example, CTRL-C was hit twice - that is not a proper way to exit Maxima either. Just remember to use :q to exit debugging and quit(); to exit Maxima, and you should always be able to escape trouble. If you find youself trying to read a long output going by quickly on a terminal, press CTRL-S to temporarily halt the output, and CTRL-Q to resume. CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 31 The End of Entry Character All expressions entered into Maxima must end with either the ; character or the $ character. The ; character is the standard character to use for this purpose. The $ symbol, while performing the same job of ending the line, suppresses the output of that line. This example illustrates these properties: (%i1) x^5+3*x^4+2*x^3+5*x^2+4*x+7; (%o1) x5 + 3 x4 + 2 x3 + 5 x2 + 4 x + 7 (%i2) x^5+3*x^4+2*x^3+5*x^2+4*x+7$ (%i3) D2; (%o3) x5 + 3 x4 + 2 x3 + 5 x2 + 4 x + 7 (%i4) x^5+3*x^4+2*x^3 +5*x^2+4*x+7; (%o4) x5 + 3 x4 + 2 x3 + 5 x2 + 4 x + 7 (%i5) x^5+3*x^4+2*x^3 +5*x^2+4*x+7; (%o5) x5 + 3 x4 + 2 x3 + 5 x2 + 4 x + 7 In (C1), we input the expression using the ; character to end the expression, and on the return line we see that D1 now contains that expression. In (C2), we input an identical expression, except that we use the $ to end the line. (D2) is assigned the contents of (C2), but does not visually display those contents. Just to verify that (D2) does in fact contain what we think it contains we ask Maxima to display iťs contents on (C3) and we see that the are in fact present. This is extremely useful if you are working on a problem which has many steps, and some of those steps would produce long outputs you don't need to actually see. In (C4), we input part of the expression, press return, finish the expression, and then use the ; character. Notice that the input did not end until we used that character and pressed return - return by itself does nothing. You see (D4) contains the same expression as (D1) shown above. To Maxima, the inputs are the same. This can be useful if you are going to input a long expression and wish to keep it straight visually, to avoid errors. You can also input spaces without adversely affecting the formula, as shown in (C5). The (C*) and (D*) Labels These labels are more than just line markers - they are actually the names in memory of the contents of the lines. This is quite useful for a number of tasks. Leťs say you wish to apply a routine, say a solve routine, to an expression for several different values. Rather than retyping the entire expression, we can use the fact that the line numbers act as markers to shorten our task considerably, as in this example: (%i1) 3*x^2+7*x+5; (%o1) 3 x2 + 7 x + 5 (%i2) solve(D1=3,x); CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 32 (%o2) x = - 1 3 ,x = -2 (%i3) solve(D1=7,x); (%o3) x = - 73 + 7 6 ,x = 73 - 7 6 (%i4) solve(D1=a,x); (%o4) x = - 12 a - 11 + 7 6 ,x = 12 a - 11 - 7 6 In this example, we desire to solve the expression 3x2 +7x+5 for x when 3x2 +7x+5 = 3, 3x2 +7x+5 = 7, and 3x2 +7x+5 = a. (a in this case is an arbitrary constant.) Rather than retype the equation many times, we merely enter it once, and then use that label to set up the similar problems more easily. The (E*) Labels In some cases, particularly when a command needs to assign generated values to variable names, the E labels will be used. These may be treated like any other maxima variable. Here is an example of E label use: Custom Labels You do not need to settle for this method of labeling - you can define your own expressions if you so choose, by using the : assign operator. Let us say, for example, that we wish to solve the problem above, but would rather call our equation FirstEquation than D1. We will show that process here, with one deliberate error for illustration of a property: (%i5) FirstEquation:3*x^2+7*x+5; (%o5) 3 x2 + 7 x + 5 (%i6) solve(FirstEquation=3,x); (%o6) x = - 1 3 ,x = -2 (%i7) solve(FirstEquation=7,x); (%o7) x = - 73 + 7 6 ,x = 73 - 7 6 (%i8) solve(FirstEquation=a,x); (%o8) x = - 12 a - 11 + 7 6 ,x = 12 a - 11 - 7 6 CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 33 (%i9) solve(firstequation=a,x); (%o9) [] You see that this process works exactly the same as before. On line (C5), you see we entered the name of FirstEquation as lower case, and the calculation failed. These names are case sensitive. This is true of all variables. Later on you will see cases in Maxima such as sin, where SIN and sin are the same, but it is not safe to assume this is always true and when in doubt, watch your cases. In general we suggest you use lower case for your maxima commands and programs - it will make them easier to read and debug. 3.1.2. To Evaluate or Not to Evaluate Operators in Maxima, such as diff for derivative, are a common feature in many Maxima expressions. The problem is, while you need to include an operator at a given point in your process, you may not want to deal with the output from it at that point in the problem. Therefore, Maxima provides the ' toggle for operators. See the example below for an example of how this works. (%i10) diff(1/sqrt(1+x^3),x); (%o10) - 3 x2 2 (x3 + 1) 3 2 (%i11) 'diff(1/sqrt(1+x^3),x); (%o11) d d x 1 x3 + 1 3.1.3. The Concept of Environment - The ev Command All mathematical operations in Maxima take place in an environment, which is to say the system is assuming it should do some things and not do other things. There will be many times you will want to change this behavior, without doing so on a global scale. Maxima provides a way to define a local environment on a per command basis, using the ev command. ev is one of the most powerful commands in Maxima, and the user will benefit greatly if they master this command early on while using Maxima. From the top We will begin with a very simple example: (%i1) ev(solve(a*x^2+b*x+c=d,x),a=3,b=4,c=5,d=6); (%o1) x = - 7 + 2 3 ,x = 7 - 2 3 (%i2) a; CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 34 (%o2) a The first line uses the ev command to solve for x without setting variables in the global environment. To make sure that our variables remain undefined, we check that a is still undefined in line (C2), and it is. Now lets examine some of the more interesting features of ev. The general syntax of the ev command is ev(exp, arg1, ..., argn). exp is an expression, like the one in the example above. You can also use a D* entry name or your own name for an expression. arg* has many possibilities, and we will try to step through them here. EXPAND(m,n) Expand is an argument which allows you to limit how Maxima expands an expression - i.e., how high a power you want it to expand. m is the maximum positive power to expand, and n is the largest negative power to expand. Here is an example: (%i1) ev((x+y)^5+(x+y)^4+(x+y)^3+(x+y)^2+(x+y)+(x+y)^-1+(x+y)^-2+(x+y)^-3+(x+y)^-4+(x+y)^- 5,EXPAND(3,3)); (%o1) 1 y3 + 3 x y2 + 3 x2 y + x3 + 1 y2 + 2 x y + x2 + (y + x)5 + (y + x)4 + 1 y + x + 1 (y + x)4 + 1 (y + x)5 + y3 + 3 x y2 + y2 + 3 x2 y + 2 x y + y + x3 + x2 + x This may be a little hard to read at first, but if you look closely you will see that every power of -3 p 3 has been expanded, otherwise the subexpression has remained in iťs original form. This is extremely useful if you want to avoid filling up your screen with large expansions that no one can read or use. Numerical Output - FLOAT and NUMER When one of the arguments of ev is FLOAT, ev will convert non-integer rational numbers to floating point. NUMER will do everything that FLOAT will, since FLOAT in invoked as part of NUMER. NUMER also handles variables defined by the user with the NUMERVAL command, which the FLOAT toggle will leave unevaluated. In order to evaluate these expressions, you can also use the float command. (%i1) a:9/4; (%o1) 9 4 (%i2) exp(a); (%o2) e 9 4 (%i3) ev(exp(a),FLOAT); (%o3) 9.487735836358526 (%i4) ev(exp(a*x),FLOAT); CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 35 (%o4) e2.25 x (%i5) numerval(b, 25); (%o5) [b] (%i6) a*b; (%o6) 9 b 4 (%i7) ev(a*b,FLOAT); (%o7) 2.25 b (%i8) ev(a*b,NUMER); (%o8) 56.25 (%i9) float(a); (%o9) 2.25 (%i10) float(b); (%o10) 25 (%i11) float(a*b); (%o11) 56.25 Specifying Local Values for Variables, Functions, etc. One of the best things about the ev command is that for one evaluation you may specify in an arg what values are to be used for the evaluation in place of variables, how to define functions, which functions to evaluate, etc. We will work through a series of examples here, probably this will be the best way to illustrate the various possibilities of this aspect of ev. (%i8) eqn1:'diff(x/(x+y)+y/(y+z)+z/(z+x),x); (%o8) d d x y z + y + z z + x + x y + x (%i9) ev(eqn1,diff); (%o9) - z (z + x)2 + 1 y + x - x (y + x)2 CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 36 (%i10) ev(eqn1,y=x+z); (%o10) d d x z + x 2 z + x + x z + 2 x + z z + x (%i11) ev(eqn1,y=x+z,diff); (%o11) 1 2 z + x z + x (2 z + x)2 + 1 z + 2 x - 2 x (z + 2 x)2 - z (z + x)2 In this example, we define eqn1 to be the derivative of a function, but use the ' character in front of the diff operator to notify Maxima that we don't want it to evaluate that derivative at this time. (More on that in the ?? section.) In the next line, we use the ev with the diff argument, which instructs ev to take all derivatives in this expression. Now, leťs say we want to define y as a function of z and x, but again avoid evaluating the derivative. We supply our definition of y as an argument to ev, and in (D3) we see that the substitution has been made. Now, leťs evaluate the derivative after the substitution has been made. We work as before, except this time we supply both the new definition of y and the diff argument, telling ev to make the substitution and then take the derivative. In this particular case, the order of the arguments does not matter. The case where it will matter is if you are making multiple substitutions - then they are handled in sequence from left to right. (need example here, one where the difference is noticeable). We can also locally define functions: (%i12) eqn4:f(x,y)*'diff(g(x,y),x); (%o12) f (x,y) d d x g(x,y) (%i13) ev(eqn4,f(x,y)=x+y,g(x,y)=x^2+y^2); (%o13) (y + x) d d x y2 + x2 (%i14) ev(eqn4,f(x,y)=x+y,g(x,y)=x^2+y^2,DIFF); (%o14) 2 x (y + x) (At the moment, ev seems to take only the first argument in the following example from solve: the manual seems to indicate it should be taking both as a list??) (%i15) eqn1:f(x,y)*'diff(g(x,y),x); (%o15) f (x,y) d d x g(x,y) CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 37 (%i16) eqn2:3*y^2+5*y+7; (%o16) 3 y2 + 5 y + 7 (%i17) ev(eqn1,g(x,y)=x^2+y^2,f(x,y)=5*x+y^3,solve(eqn2=5,y)); (%o17) 5 x - 8 27 d d x x2 + 4 9 (%i18) ev(eqn1,g(x,y)=x^2+y^2,f(x,y)=5*x+y^3,solve(eqn2=1,y),diff); (%o18) 2 x 5 x - 47 i + 5 3 216 (%i19) ev(eqn1,g(x,y)=x^2+y^2,f(x,y)=5*x+y^3,solve(eqn2=1,y),diff,FLOAT); (%o19) 2 x 5 x - 0.00462962962963 47 i + 5 3 Other arguments for ev INFEVAL - This option leads to an "infinite evaluation" mode, where ev repeatedly evaluates an expression until it stops changing. To prevent a variable, say X, from being evaluated a way in this mode, simply include X='X as an argument to ev. There are dangers with this command - it is quite possible to generate infinite evaluation loops. For example, ev(X,X=X+1,INFEVAL); will generate such a loop. Here is an example: (need example where this is useful.) How ev works The flow of the ev command works like this: 1. The environment is set up by scanning the arguments. During this step, a list is made of non-subscripted variables appearing on the left side of equations in the arguments or in the value of some arguments if the value is an equation. Both subscripted variables which do not have associated array functions and non-subscripted variables in the expression exp are replaced by their global values, except for those appearing in the generated list. 2. If any substitutions are indicated, they are carried out. 3. The resulting expression is then re-evaluated, unless one of the arguments was NO-EVAL, and simplified according to the arguments. Note that any function calls in exp will be carried out AFTER the variables in it are evaluated. 4. If one of the arguments was EVAL, the previous two steps are repeated. CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 38 3.1.4. Clearing values from the system - the kill command Many times you will define something in Maxima, only to want to remove that definition later in the computation. The way you do this in Maxima is quite simple - using the kill command. Here is an example: (%i5) A:7$ (%i6) A; (%o6) 7 (%i7) kill(A); (%o7) DONE (%i8) A; (%o8) A kill is used in many situations, and has many uses. You will see it appear throughout this manual, in different contexts. There are general arguements you can use, such as kill(all), which will essentially start you out in a new, clean environment. (Add any relevant general kill options here - save kill(rules) for rules section, etc.) 3.2 Common Operators in Maxima An operator is simply something that signals a specific operation is to be performed. There are many, many possible operators in Maxima. We will address various operators for specific jobs all throughout this manual - this section is not comprehensive. 3.2.1. Assignment Operators In mathematics, we quite often want to declare functions, assign values to numbers, and do many similarly useful things. Maxima has a variety of operators for this purpose. : The basic assignment operator. We have already seen this operator in action; it is one of the most common in maxima. (%i20) A:7; (%o20) 7 (%i21) A; (%o21) 7 := This is the operator you would use to define functions. This is a common thing to do in computer algebra, so we will illustrate both how to and how not to do this. CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 39 The right way: (%i2) y(x):=x^2; (%o2) y(x) : = x2 (%i3) y(2); (%o3) 4 Several possible wrong ways: (%i22) y:=x^2; Improper function definition: y -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) (%i23) y=x^2; (%o23) y = x2 (%i24) y(2); (%o24) y(2) (%i25) y(x)=x^2; (%o25) y(x) = x2 (%i26) y(2); (%o26) y(2) (%i27) y[x]=x^2; (%o27) yx = x2 (%i28) y[2]; (%o28) y2 (%i29) y(x):=x^2; (%o29) y(x) : = x2 (%i30) y(2); (%o30) 4 CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 40 Look over the above example - it pays to know what doesn't work. If you recognize the error or incorrect result you get, it will make for faster debugging. :: This operator is related to the : operator, but does not function in quite the same way. This is more what a programmer would refer to as a pointer. The best way to explain is to give you an example of how it behaves: (%i9) A:3$ (%i10) B:5; (%o10) 5 (%i11) C:'A; (%o11) A (%i12) C::B; (%o12) 5 (%i13) C; (%o13) A (%i14) A; (%o14) 5 You see C points to A, and A is thus assigned the value of B. ! This is the factorial operator. (%i4) 8!; (%o4) 40320 !! This is the double factorial operator. This is defined in Maxima as the product of all the consecutive odd (or even) integers from 1 (or 2) to the odd (or even) arguement. (%i6) 8!!; (%o6) 384 (%i7) 2*4*6*8; (%o7) 384 CHAPTER 3. THE BASICS - WHAT YOU NEED TO KNOW TO OPERATE IN MAXIMA 41 sqrt(x) This is your basic square root operator. (%i13) sqrt(x^2); (%o13) |x| (%i14) sqrt(1/2); (%o14) 1 2 (%i15) sqrt(9); (%o15) 3 Of course, this hardly begins to describe all the operators in the system, but what you see here are some of the more common and useful ones. CHAPTER 4 Trig through Calculus This chapter and the next will probably split into many more - Trig, Algebra, Calculus, Programming, etc,etc,etc. I just don't know at this point. These chapters will probably largely be example based. Using things such as ratsimp, trigsimp, etc. Here we will discuss Maxima's ability to handle integration, differentiation, and other related concepts. 4.1 Trigonometric Functions These operate more or less as you would expect. The following functions are defined by default: sin Sine asin Arc Sine cos Cosine acos Arc Cosine tan Tangent atan Arc Tangent csc Cosecant acsc Arc Cosecant sec Secant asec Arc Secant cot Cotangent acot Arc Cotangent sinh Hyperbolic Sine asinh Hyperbolic Arc Sine cosh Hyperbolic Cosine acosh Hyperbolic Arc Cosine tanh Hyperbolic Tangent atanh Hyperbolic Arc Tangent csch Hyperbolic Cosecant acsch Hyperbolic Arc Cosecant sech Hyperbolic Secant asech Hyperbolic Arc Secant coth Hyperbolic Cotangent acoth Hyperbolic Arc Cotangent There are a couple wrinkles worth noting - by default, Maxima will not simplify expressions which numerically are nice fractions of , so there exists a package which may be loaded to allow this called atrig1. Here is an example: (%i1) acos(1/sqrt(2)); (%o1) arccos 1 2 (%i2) load(atrig1)$ (%i3) acos(1/sqrt(2)); (%o3) 4 42 CHAPTER 4. TRIG THROUGH CALCULUS 43 Maxima is aware of the Half Angle relations, but by default will not use them. There is a variable which can be set called halfangles, and when that is set to true the Half Angle definitions will be used. (%i4) sin(a/2); (%o4) sin a 2 (%i5) halfangles:true; (%o5) true (%i6) sin(a/2); (%o6) 1 - cosa 2 You should be aware that when solving expressions involving trig functions, not all solutions will be presented. This is inevitable, since in many cases there are an infinite number - typically one will be displayed. Usually you are warned when this is happens. (%i7) solve(sin(x)=%PI/2,x); SOLVE is using arc-trig functions to get a solution. Some solutions will be lost. (%o7) x = arcsin 2 There are a few global variables you can set which will change how Maxima handles trig expressions: * TRIGINVERSES This can be set to one of three values: ALL, TRUE, or FALSE. The default is ALL ­ ALL When set to ALL, both arctfun(tfun(x)) and fun(arctfun(x)) are evaluated to x. ­ TRUE When set to TRUE, the arctfun(tfun(x)) simplification is turned off. ­ FALSE When set to FALSE, both simplifications are turned off. * TRIGSIGN Can be set to TRUE or FALSE. The default is TRUE. If TRUE, for example, sin(-x) simplifies to -sin(x). 4.2 Differentiation To differentiate an expression, use the diff command. diff(expr,var) differentiates an expression with respect to the variable var. (%i8) sin(x)*cos(x); (%o8) cosx sinx (%i9) diff(%,x); (%o9) cos2 x - sin2 x CHAPTER 4. TRIG THROUGH CALCULUS 44 To take a second order derivative, use diff(expr,var,2). (%i10) diff(sin(x)*cos(x),x,2); (%o10) -4 cosx sinx Differentiation, unlike integration, can be handled in a fairly general way by computer algebra. As a result, you will be able to take derivatives in most cases. We will show some examples here: Basic Algebraic Examples: (%i11) diff(3*x^5+x^4+7*x^3-x^2+17,x); (%o11) 15 x4 + 4 x3 + 21 x2 - 2 x (%i12) diff((x^2+1)/(x^2-1),x); (%o12) 2 x x2 - 1 - 2 x x2 + 1 (x2 - 1)2 (%i13) diff((x^2+1)^5*(x^7-5*x-2)^19,x); (%o13) 10 x x2 + 1 4 x7 - 5 x - 2 19 + 19 x2 + 1 5 7 x6 - 5 x7 - 5 x - 2 18 (%i14) diff(x^(2/3)+x^(5/7),x); (%o14) 5 7 x 2 7 + 2 3 x 1 3 Chain Rule Example: In order to handle the problem of a function which depends in an unknown way upon some variable, Maxima provides the depends command. Using it, you can derive general chain rule formulas. It should be noted these relations are understood only by the diff command - for operations such as integration you must give their dependencies explicitly in the command. (%i15) DEPENDS([U],[r,theta],[r,theta],[x,y]); (%o15) [U (r,),r(x,y),(x,y)] (%i16) diff(U,x)+diff(U,y); (%o16) d d y d d U + d d x d d U + d d y r d d r U + d d x r d d r U If we wish to take derivatives with respect to multiple variables, for example d2 dxdy , the syntax for derivatives is quite general and we can perform the operation as follows: (%i17) diff(U,x,1,y,1); (%o17) d d x d d y d2 d 2 U + d d y r d2 d r d U + d2 d x d y d d U + d d x r d d y r d2 d r2 U + d d y d2 d r d U + d2 d x d y r d d r U CHAPTER 4. TRIG THROUGH CALCULUS 45 This is the same thing as doing (%i18) diff(diff(U,x),y); (%o18) d d x d d y d2 d 2 U + d d y r d2 d r d U + d2 d x d y d d U + d d x r d d y r d2 d r2 U + d d y d2 d r d U + d2 d x d y r d d r U Trigonometric Derivatives: (%i25) diff(cos(x),x); (%o25) -sinx (%i26) diff(acos(x),x); (%o26) - 1 1 - x2 (%i27) diff(tan(x),x); (%o27) sec2 x (%i28) diff(atan(x),x); (%o28) 1 x2 + 1 (%i29) diff(sinh(x),x); (%o29) coshx (%i30) diff(asinh(x),x); (%o30) 1 x2 + 1 4.3 Integration Unlike differentiation, integration cannot be readily expressed in a general way. Maxima is quite capable when it comes to to such problems, although like all computer algebra systems it has its limits. In general, integrate is the command most users will use to perform various types of basic integrals. It is therefore a logical place to begin the introduction. Beginning with a very basic example: (%i31) integrate(a*x^n,x); Is n+1 zero or nonzero? nonzero; CHAPTER 4. TRIG THROUGH CALCULUS 46 (%o31) a xn+1 n + 1 Even in this basic case, there is a lot going on. The general form of the integrate command for indefinite integrals is integrate(f(x),x). When Maxima does not have sufficient information to evaluate an integral, it will ask the user questions. In the above example, for instance, Whether or not n+1 was zero impacted how Maxima would approach the problem. Above, it evaluated the integral after being told n+1 was not zero. If the same integral is performed again, this time informing the system that n+1 is zero, the results are different: (%i32) integrate(a*x^n,x); Is n+1 zero or nonzero? zero; (%o32) a logx 4.3.1. The assume Command As one is working on a long problem session, having to answer the same questions repeatedly quickly becomes inefficient. Fortunately, Maxima provides an assume command which lets the system proceed without having to repeatedly inquire at to the state of a variable. This command is actually useful throughout the Maxima system, not just in integration problems, but since integration is likely where most users will first encounter the need for it we will discuss it here. Remember, this is the command you will use any time you wish to instruct Maxima to assume some fact whatever you happen to be doing. You will see this command throughout this book. We will use the previous integration example as the first illustration of how this process works. (%i33) assume(n+1>0); (%o33) [n > -1] (%i34) integrate(a*x^n,x); (%o34) a xn+1 n + 1 Notice Maxima did not ask any questions, because it was able to find the information it needed in its assume database. Of couse, for one integral it is simpler to just answer the question, but now if we wish to do another integral that also depends on this knowledge: (%i35) integrate((a+b)*x^(n+1),x); (%o35) (b + a) xn+2 n + 2 Maxima already knew enough to handle the new integral. Of course, we might not want this asssumption later on, so we need a way to get rid of it. This is done with the forget command: (%i36) forget(n+1>0); CHAPTER 4. TRIG THROUGH CALCULUS 47 (%o36) [n > -1] (%i37) integrate((a+b)*x^n,x); Is n+1 zero or nonzero? zero; (%o37) (b + a) logx For multiple rule situations assume and forget will also take more than one assumption at a time, as in this example: (%i38) assume(n+1>0, m+1>0); (%o38) [n > -1,m > -1] (%i39) integrate(a*x^n+b*x^m,x); (%o39) a xn+1 n + 1 + b xm+1 m + 1 (%i40) forget(n+1>0, m+1>0); (%o40) [n > -1,m > -1] (%i41) integrate(a*x^n+b*x^m,x); Is m+1 zero or nonzero? zero; Is n+1 zero or nonzero? zero; (%o41) b logx + a logx 4.3.2. Definite Integrals The same basic integration command is used for definite integrals. Leťs take a basic example: (%i42) integrate(a+x^3,x,0,5); (%o42) 20 a + 625 4 The basic syntax is apparent: integrate(f(x),x,lowerlimit,upperlimit) CHAPTER 4. TRIG THROUGH CALCULUS 48 4.3.3. changevar Maxima provides a command changevar which can make a change of variable in an integral. It has the form changevar(exp,f(x,y),y,x) What this does is make the change of variable given by f(x,y) = 0 in all integrals occurring in exp with integration with respect to x; y is the new variable. For example: (%i43) 'integrate(exp(sqrt(5*x)),x,0,4)+'integrate(exp(sqrt(5*x+1)),x,0,5)+ 'integrate(exp(sqrt(z*x)),z,0,4); (%o43) Z 4 0 e x z dz + Z 5 0 e 5 x+1 dx + Z 4 0 e 5 x dx (%i44) changevar(%,x-y^2/5,y,x); (%o44) Z 4 0 e x z dz - 2 R 0 -2 5 y e|y| dy 5 - 2 R 0 -5 y e y2+1 dy 5 If you examine the above case, you see that the two integrals being integrated with respect to x have undergone a variable, change, while the z dependant integral has not. 4.3.4. Behind the Black Box - Using Specific Approaches Once a user begins serious work with integration in Maxima, they may find that they want to use other techniques. Maxima has several functions which allow more power and flexibility. Definite integration will be the first example: (Need example of where DEFINT fails but ROMBERG succeeds.) Discuss LDEFINT RISCH ILT INTSCE 4.3.5. Other Examples Since integration is such a major feature, we will include here a fairly extensive collection of examples of integrals. (%i45) integrate(x,x); (%o45) x2 2 (%i46) assume(a>0)$ (%i47) assume(n>0)$ (%i48) integrate(a*x^n,x); (%o48) a xn+1 n + 1 (%i49) assume(b>0)$ (%i50) integrate(a*exp(x*b),x); (%o50) a eb x b (%i51) assume(c>0)$ (%i52) integrate(a*b^(x*c),x); (%o52) a bc x logb c CHAPTER 4. TRIG THROUGH CALCULUS 49 (%i53) integrate(log(x),x); (%o53) x logx - x (%i54) integrate(a/(b^2+x^2),x); (%o54) a arctan x b b (%i55) assume(m>0)$ (%i56) integrate(x^m*(a+b*x)^5,x); (%o56) b5 xm+6 m + 6 + 5 a b4 xm+5 m + 5 + 10 a2 b3 xm+4 m + 4 + 10 a3 b2 xm+3 m + 3 + 5 a4 b xm+2 m + 2 + a5 xm+1 m + 1 (%i57) integrate(x/(a+b*x)^n,x); (%o57) b2 (n - 1) x2 + a b n x + a2 e-n log(b x+a) b2 (n2 - 3 n + 2) (%i58) integrate(x^2/(a+b*x)^n,x); (%o58) b3 n2 - 3 n + 2 x3 + a b2 n2 - n x2 + 2 a2 b n x + 2 a3 e-n log(b x+a) b3 (n3 - 6 n2 + 11 n - 6) (%i59) integrate(1/(x^2-c^2)^5,x); (%o59) - 35 log(x + c) 256 c9 + 35 log(x - c) 256 c9 + 105 x7 - 385 c2 x5 + 511 c4 x3 - 279 c6 x 384 c8 x8 - 1536 c10 x6 + 2304 c12 x4 - 1536 c14 x2 + 384 c16 (%i60) integrate(1/(a+b*x^2)^4,x); (%o60) 5 arctan b x a 16 a 7 2 b + 15 b2 x5 + 40 a b x3 + 33 a2 x 48 a3 b3 x6 + 144 a4 b2 x4 + 144 a5 b x2 + 48 a6 (%i61) integrate(sqrt(a+b*x)/(x^5),x); (%o61) - 5 b4 log 2 b x+a-2 a 2 b x+a+2 a 128 a 7 2 - 15 b4 (b x + a) 7 2 - 55 a b4 (b x + a) 5 2 + 73 a2 b4 (b x + a) 3 2 + 15 a3 b4 b x + a 192 a3 (b x + a)4 - 768 a4 (b x + a)3 + 1152 a5 (b x + a)2 - 768 a6 (b x + a) + 192 a7 CHAPTER 5 Advanced Mathematics - ODEs and Beyond 5.1 Ordinary Differential Equations 5.1.1. Defining Ordinary Differential Equations There are three standard ways to represent an ordinary differential equation, such as x2 y + 3xy = sin(x)/x, in Maxima. The simplest way is to represent the derivatives by 'diff(y,x), 'diff(y,x,2), etc. The above ordinary differential equation would then be entered as (%i1) x^2*'diff(y,x) + 3*x*y = sin(x)/x; (%o1) x2 d d x y + 3 x y = sinx x Note that the derivative 'diff(y,x) is quoted, to prevent it from being evaluated (to 0). The second way is to use the depends command to tell Maxima that y is a functions of x, making the quotes unnecessary. The above equation would then be entered as (%i2) depends(y,x); (%o2) [y(x)] (%i3) x^2*diff(y,x) + 3*x*y = sin(x)/x; (%o3) x2 d d x y + 3 x y = sinx x The third way would be to write y(x) explicitly as a function of x. The above equation would then be entered as (%i4) x^2*diff(y(x),x) + 3*x*y(x) = sin(x)/x; (%o4) x2 d d x y(x) + 3 x y(x) = sinx x Different commands for working with differential equations require different representations of the equations. For the command ode2 (see subsection 5.1.2), it is often more useful to use one of the first two representations, while for the command desolve (see subsection 5.1.3) it is required to use the third representation. 50 CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 51 5.1.2. Solving Ordinary Differential Equations: ode2 Using ode2 Maxima can solve first and second order differential equations using the ode2 command. The command ode2(eqn,depvar,indvar) will solve the differential equation given by eqn, assuming that depvar and indvar are the dependent and independent variables, respectively. (If an expression expr is given instead of an equation, it is assumed that the expression represents the equation expr=0.) (%i5) ode2(x^2*diff(y,x) + 3*x*y = sin(x)/x, y, x); (%o5) y = %C - cosx x3 If ode2 cannot solve a given equation, it returns the value FALSE. Initial and Boundary Conditions After a differential equation is solved by ode2, initial values or boundary conditions can be given to the solution. The commands for giving the conditions to the solution, however, require that the differential equation not be given explicitly as a function of the variable; i.e., diff(y,x) would have to be used rather than diff(y(x),x) to denote the derivative. For a first order differential equation, the initial condition can be given using ic1. If ode2 returns the general solution soln to a first order differential equation, the command ic1(soln, indvar=a, depvar=b) will return the particular solution which equals b when the variable equals a. (%i6) soln1:ode2(x^2*diff(y,x) + 3*x*y = sin(x)/x, y, x); (%o6) y = %C - cosx x3 (%i7) ic1(soln1, x=1, y=1); (%o7) y = cosx - cos1 - 1 x3 For a second order differential equation, conditions can be given as initial conditions, using ic2, or as boundary conditions, using bc2. If ode2 returns the general solution soln to a second order differential equation, the command ic2(soln, indvar=a, depvar=b, diff(depvar, indvar)=c) will return the particular solution which equals b and whose derivative equals c when the variable equals a. (%i8) eqn2: diff(y,x,2) + y = 4*x; (%o8) d2 d x2 y + y = 4 x (%i9) soln2: ode2(eqn2, y, x); (%o9) y = %K1 sinx + %K2 cosx + 4 x (%i10) ic2(soln2, x=0, y=1, diff(y,x)=3); (%o10) y = -sinx + cosx + 4 x Similarly, if ode2 returns the general solution soln to a second order differential equation, the command bc2(soln, indvar=a, depvar=b, indvar=c, depvar=d) will return the particular solution which equals b when the variable equals a and which equals d when the variable equals c. CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 52 (%i11) bc2(soln2, x=0, y=3, x=2, y=1); (%o11) y = (3 cos2 + 7) sinx sin2 + 3 cosx + 4 x ode2 Methods To solve a given differential equation, ode2 will attempt a series of standard methods for solving differential equations. These methods will be described below, more in-depth discussions of these techniques can be found in any standard introductory text on ordinary differential equations (such as Elementary Differential Equations and Boundary Value Problems by Boyce and DiPrima, from which most of these routines were taken). The first thing ode2 will do with a differential equation is determine whether it is first order or second order. For first order differential equations, ode2 will check to see if the equation falls into one of the following categories, in which case the equation will be solved appropriately. Linear. A first order differential equation is linear if it can be written in the form y + p(x)y = q(x). In this case, the solution is given by y = (I(x)+c)/(x), where (x) is eP(x) for some antiderivative P(x) of p(x), I(x) is an antiderivative of (x)q(x), and c is an arbitrary constant. (%i12) linode:diff(y,x) + x*y = x^2; (%o12) d d x y + x y = x2 (%i13) ode2(linode,y,x); (%o13) y = e- x2 2 2 i erf i x 2 2 + x e x2 2 + %C Separable. A first order differential equation is separable if it can be put in the form M(x) = N(y)y . In this case, an implicit solution is obtained by integrating both sides of M(x)dx = N(y)dy. (It may or may not be possible to solve for y explicitly.) (%i14) separableode:(3*x^2+4*x+2)=(2*y-1)*diff(y,x); (%o14) 3 x2 + 4 x + 2 = (2 y - 1) d d x y (%i15) ode2(separableode, y, x); (%o15) y2 - y = x3 + 2 x2 + 2 x + %C Exact. A first order differential equation is exact if it can be put in the form p(x,y)y +q(x,y) = 0, where p(x,y) = M(x,y)/y and q(x,y) = M(x,y)/x for some M(x,y). In this case, the solution will be given implicitly by M(x,y) = 0. (It may or may not be possible to solve for y explicitly.) (%i16) exactode:x^2*cos(x*y)*diff(y,x) + (sin(x*y) + x*y*cos(x*y))=0; (%o16) sin(x y) + x2 d d x y cos(x y) + x y cos(x y) = 0 (%i17) ode2(exactode,y,x); (%o17) x sin(x y) = %C CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 53 If the given differential equation can be put in the form p(x,y)y + q(x,y) = 0 but is not exact, ode2 checks to see if there is an integrating factor (x,y) which will make (x,y)p(x,y)y +(x,y)q(x,y) = 0 exact, in which case this new equation will be solved as above. (%i18) intfactorode:(2*x*y - exp(-2*y))*diff(y,x) + y =0; (%o18) 2 x y - e-2 y d d x y + y = 0 (%i19) ode2(intfactorode,y,x); (%o19) x e2 y - logy = %C Homogeneous. A first order differential equation is homogeneous if it can be put in the form y = F(y/x). In this case, the substitution v = y/x will transform the equation into the separable equation xv + v = F(v), which can be solved as above. (%i20) homode:diff(y,x) = (y/x)^2 + 2*(y/x); (%o20) d d x y = y2 x2 + 2 y x (%i21) ode2(homode,y,x); (%o21) x y + x2 y = %C Bernoulli. The equation y + p(x)y = q(x)yn, n = 0,1, is called Bernoulli's equation with index n. The transformation v = y1-n will transform Bernoulli's equation into the linear equation v + (1 - n)p(x)v = (1 - n)q(x), which can be solved as above. (%i22) berode:diff(y,x) + (2/x)*y = (1/x^2)* y^3; (%o22) d d x y + 2 y x = y3 x2 (%i23) ode2(berode, y, x); (%o23) y = 1 2 5 x5 + %C x2 General Homogeneous. A first order differential equation is said to be general homogeneous of index n if it can be written in the form y = (y/x)G(yxn). In this case, a solution is given implicitly by x = ceI(yxn), where I(u) is an antiderivative of 1/(u(n+G(u))) and c is an arbitrary constant. (It may or may not be possible to solve for y explicitly.) If the differential equation is second order, then ode2 will determine if the equation is linear or not. In the linear case, when the equation can be written y + p(x)y +q(x)y = r(x), ode2 will try to solve the equation by first solving the homogeneous part, y + p(x)y +q(x)y = 0. The general solution of the homogeneous part will be of the form y = k1y1 +k2y2 for arbitrary constants k1 and k2. If r(x) = 0, ode2 will then use variation of parameters to find a particular solution yp of the original equation. The general solution of the full equation will then be y = k1y1 + k2y2 + yp. To solve the homogeneous part, ode2 will check to see if the equation falls into one of the following categories, in which case the equation will be solved appropriately. Constant Coefficients. If the differential equation has constant coefficients, and so is of the form y +ay +by = 0, then the solution is y = k1er1x + k2er2x, where r1 and r2 are the solutions of r2 + ar + b = 0. In case r2 + ar + b = 0 has a double root, the solution of the differential equation is y = k1erx + k2xerx. In some cases where the CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 54 equation doesn't have constant coefficients, ode2 will attempt to use a simple transformation to reduce it to constant coefficents. (%i24) ccode1: diff(y,x,2) - 3*diff(y,x) + 2*y=0; (%o24) d2 d x2 y - 3 d d x y + 2 y = 0 (%i25) ccode2: diff(y,x,2) - 4*diff(y,x) + 4*y=0; (%o25) d2 d x2 y - 4 d d x y + 4 y = 0 (%i26) ode2(ccode1, y, x); (%o26) y = %K1 e2 x + %K2 ex (%i27) ode2(ccode2, y, x); (%o27) y = (%K2 x + %K1) e2 x Exact. A second order differential equation is exact if it can be written in the form [f(x)y ] +[g(x)y] = 0. Integrating this equation will reduce it to a first order differential equation, which can be solved as above. (%i28) exactode2: x^2*diff(y,x,2) + x*diff(y,x) - y =0; (%o28) x2 d2 d x2 y + x d d x y - y = 0 (%i29) ode2(exactode2, y,x); (%o29) y = %K2 x - %K1 2 x Euler. The equation x2y + axy + by = 0 is Euler's equation. The solution is given by y = k1xr1 + k2xr2 , where r1 and r2 are solutions of r(r-1)+ar+b = 0. In case r(r-1)+ar+b = 0 has a double root, the solution is given by y = k1xr +k2 ln(x)xr. (%i30) eulerode1: x^2*diff(y,x,2) + 4*x*diff(y,x) + 2*y = 0; (%o30) x2 d2 d x2 y + 4 x d d x y + 2 y = 0 (%i31) eulerode2: x^2*diff(y,x,2) + 5*x*diff(y,x) + 4*y = 0; (%o31) x2 d2 d x2 y + 5 x d d x y + 4 y = 0 (%i32) ode2(eulerode1, y, x); (%o32) y = %K1 x + %K2 x2 (%i33) ode2(eulerode2, y, x); (%o33) y = %K2 logx + %K1 x2 CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 55 Besseľs Equation. The equation x2y +xy +(x2 -2)y = 0 is called Besseľs equation of order . For = 1/2, the solution is y = k1 sin(x)/ x+k2 cos(x)/ x; for integer , the answer will be y = k1Y(x)+k2J(x), where J andY are the Bessel functions of the first and second kind. (%i34) besselode1:x^2*diff(y,x,2) + x*diff(y,x) + (x^2 - 1/4)*y=0; (%o34) x2 d2 d x2 y + x d d x y + x2 - 1 4 y = 0 (%i35) besselode2:x^2*diff(y,x,2) + x*diff(y,x) + (x^2 - 4)*y=0; (%o35) x2 d2 d x2 y + x d d x y + x2 - 4 y = 0 (%i36) ode2(besselode1, y, x); (%o36) y = %K1 sinx + %K2 cosx x (%i37) ode2(besselode2, y, x); (%o37) y = %K2 %Y2(x) + %K1 %J2(x) ode2 can also handle translates of Besseľs equation; i.e., differential equations of the form (x - a)2y + (x - a)y + ((x-a)2 -2)y = 0 (%i38) besselode3:(x-1)^2*diff(y,x,2) + (x-1)*diff(y,x) + ((x-1)^2 - 4)*y=0; (%o38) (x - 1)2 d2 d x2 y + (x - 1) d d x y + (x - 1)2 - 4 y = 0 (%i39) ode2(besselode3, y, x); (%o39) y = %K2 %Y2(x - 1) + %K1 %J2(x - 1) If ode2 successfully solves the homogeneous part of an inhomogeneous equation, it then tries to find a particular solution using variation of parameters. Variation of Parameters. If y = k1y1 +k2y2 is the general solution of y + p(x)y +q(x)y = 0, then a particular solution of y + p(x)y +q(x)y = r(x) can be found by replacing the arbitrary constants k1 and k2 with arbitrary functions u1 and u2, and looking for a solution of the form y = u1y1 +u2y2. One such solution is given if u1 and u2 are antiderivatives of -y2r/(y1y2 -y2y1) and y1r/(y1y2 -y2y1), respectively. (%i40) varparode:diff(y,x,2) + 2*diff(y,x) + y = exp(x); (%o40) d2 d x2 y + 2 d d x y + y = ex (%i41) ode2(varparode,y,x); (%o41) y = ex 4 + (%K2 x + %K1) e-x In case the second order differential equation is not linear, ode2 will check to see if either the dependent variable or the independent variable is missing, in which case one of the following two methods will be used. CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 56 Missing dependent variable. If the undifferentiated dependent variable y is not present in a second order differential equation, the substitution v = y , v = y will reduce the differential equation to first order. This can be solved as above. Once v is obtained, y can be obtained by integrating v. (%i42) noyode:x*diff(y,x,2) + (diff(y,x))^2=0; (%o42) x d2 d x2 y + %DERIVATIVE2 ((y,x,1)) = 0 (%i43) ode2(noyode,y,x); (%o43) y = Z 1 logx + %K1 dx + %K2 Missing independent variable. If the independent variable x is not present in a second order differential equation, then making the dependent variable y a temporary independent variable and using the substitution v = y , v = y , the equation can again be reduced to first order. Since the derivatives are taken with respect to x, however, the new equation will involve three variables. This can be resolved by noting that v = dv/dx = (dv/dy)(dy/dx) = (dv/dy)v, and so v = dv/dx can be replaced by vdv/textrmdy. This will result in a first order differential equation with dependent variable v and independent variable y. This can be solved as above. Once v is obtained (in terms of y), y is a solution of the differential equation y = v(y), which can be solved as above. (%i44) noxode: y*diff(y,x,2) + (diff(y,x))^2 = 0; (%o44) y d2 d x2 y + %DERIVATIVE2 ((y,x,1)) = 0 (%i45) ode2(noxode,y,x); (%o45) y2 2 %K1 = x + %K2 Information on the Method The ode2 routine will store information about the method used to solve the differential equations in various variables. The variable method will keep track of the method used to solve the differential equations. (%i46) ode2(separableode, y, x); (%o46) y2 - y = x3 + 2 x2 + 2 x + %C (%i47) method; (%o47) SEPARABLE In the case where an integrating factor was used to make a differential equation exact, the variable intfactor will be the integrating factor used. (%i48) ode2(intfactorode, y, x); (%o48) x e2 y - logy = %C (%i49) method; CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 57 (%o49) EXACT (%i50) intfactor; (%o50) e2 y y When Bernoulli's equation or a generalized homogeneous equation is solved, the variable odeindex will be the index of the equation. (%i51) ode2(berode, y, x); (%o51) y = 1 2 5 x5 + %C x2 (%i52) method; (%o52) BERNOULLI (%i53) odeindex; (%o53) 3 When an inhomogeneous second degree linear differential equation is solved, the variable yp will be the particular solution arrived at by variation of parameters. (%i54) ode2(varparode, y, x); (%o54) y = ex 4 + (%K2 x + %K1) e-x (%i55) method; (%o55) VARIATIONOFPARAMETERS (%i56) yp; (%o56) ex 4 5.1.3. Solving Ordinary Differential Equations: desolve Using desolve Maxima can solve systems of linear ordinary differential equation with constant coefficients using the desolve command. The differential equations must be given using functional notation, rather than with dependent variables; i.e., diff(y(x),x) would have to be used rather than diff(y,x) to denote the derivative. The command desolve(delist, fnlist), will solve the system of differential equations given by the list delist, where fnlist is a list of the functions to be solved for. CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 58 (%i57) de1:diff(f(x),x)=diff(g(x),x)+sin(x); (%o57) d d x f (x) = d d x g(x) + sinx (%i58) de2:diff(g(x),x,2)=diff(f(x),x) - cos(x); (%o58) d2 d x2 g(x) = d d x f (x) - cosx (%i59) desolve([de1,de2],[f(x),g(x)]); (%o59) f (x) = ex d d x g(x) x=0 - d d x g(x) x=0 + f (0),g(x) = ex d d x g(x) x=0 - d d x g(x) x=0 + cosx + g(0) - 1 A single differential equation of a single unknown function can also be solved by desolve; in this case, it isn't necessary to enter them as lists. (%i60) de3:'diff(f(x),x,2)+ f(x) = 2*x; (%o60) d2 d x2 f (x) + f (x) = 2 x (%i61) desolve(de3, f(x)); (%o61) f (x) = sinx d d x f (x) x=0 - 2 + f (0) cosx + 2 x Initial Conditions Initial conditions can be specified for solutions given by desolve using the atvalue command. The conditions, however, can only be given at 0, and must be given before the equations are solved. (%i62) atvalue(f(x),x=0,1); (%o62) 1 (%i63) atvalue(g(x),x=0,2); (%o63) 2 (%i64) atvalue(diff(g(x),x),x=0,3); (%o64) 3 (%i65) desolve([de1,de2],[f(x),g(x)]); (%o65) [f (x) = 3 ex - 2,g(x) = cosx + 3 ex - 2] CHAPTER 5. ADVANCED MATHEMATICS - ODES AND BEYOND 59 desolve Method The desolve routine uses the LaPlace transform to solve the systems of differential equations. If f(t) is defined for all t 0, then the LaPlace transform of f is given by F(s) = L{ f(t)} = R 0 e-st f(t)dt. The LaPlace transform has the useful property that a derivative is transformed into multiplication by the variable; if L{ f(t)} = F(s), then L{ f (t)} = sF(s) - f(0). The LaPlace transform can thus transform a system of linear differential equations into a system of ordinary equations. (Note, however, that the LaPlace transform will transform multiplication by a variable into differentiation; if L{f(t)} = F(s), then L{t f(t)} = -F (s). The original differential equations need to have constant coefficients to prevent this.) If this new system can be solved, the LaPlace tranform can be inverted to give solutions of the original system of differential equations. CHAPTER 6 Matrix Operations and Vectors Maxima has many matrix capabilities. Unfortunately iťs vector handling at the time of this writing is not terribly robust, and the reader is cautioned to use it with care. 60 CHAPTER 7 Introduction to Maxima's Programming Language This chapter assumes that readers are familiar with the basic ideas of algebraic manipulation from Chapter ??, and know at least one programming language, and wish to use Maxima for more ambitious tasks than can be handled in a few sequential commands. If you are familiar with Pascal or Algol 60, you will probably find this adequate as a programming background. Familiarity with Fortran or Basic is less useful. Maxima's user-programming language1 is designed to allow you to define program modules or functions for algebraic manipulation. Each module uses zero or more arguments, and returns an algebraic expression. Since numbers are special cases of algebraic expressions, Maxima's user-language can be used as a numeric language too. Because the language is implemented as an interpreter it is usually more general than compiler-based languages, and also tends to be rather slow in tight inner loops of simple operations, by comparison. It has novel linguistic features, some of which are illustrated below. 7.1 Some Examples f1 and f2 defined below, are versions of the factorial function. Observe the punctuation carefully. Assignment is :, function definition is :=, statements are separated from one another by , (not terminated by commas). The label loop is set off by a comma as though it were a statement too. We have added indentation to make the programs conform to what you might expect, but extra spaces and tabs are optional. There is a conditional execution statement (the if) which has an optional else clause. (C1) f1(x):=if x<1 then 1 else x*f1(x-1)$ (C2) f1(5); (D2) 120 (C3) f2(x) := block ([temp], temp:1, while x > 1 do ( temp:x*temp, x:x-1), /*end while*/ temp)$ (C4) f2(5); (D4) 120 Observe the simplicity of using Maxima compared to, say, Pascal. There is no need to write a driver or main program with input or output commands. The input is provided through function application on an argument, and the output is the displayed value. 1The system-programming language used for implementing Maxima namely LISP, is quite different. If you wish to see how Maxima was constructed, you need to know LISP to understand the source code. 61 CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 62 Every command or function in Maxima has a value, and may, in addition, have some side-effects, such as the setting of variables, or the printing of messages. The block construction illustrated above is analogous to a procedure declaration. The first part of it is a list of local variables, and following that, expressions which are evaluated in order. Certain expressions or commands make sense only within a block, not at Maxima's command level: these are labels, return s and go s. The semantics of each of these commands conforms to the usual intuitive meaning. If the last statement in a block does not cause a transfer of control, and execution falls through the bottom, the value returned from the block is the value of the last expression evaluated. 7.2 Unconventional Conditionals The next example shows a function with a side-effect, but the major point is to illustrate some subtleties which you may not have thought about in conditional statements (if-then-else ). If you were asked, Is A greater than B, it would seem you could respond either yes or no. In your conventional programming language, certainly, that would be a reasonable assumption. But really, wouldn't it be appropriate in some circumstances for you to answer, How should I know?? That option is preserved in Maxima. If the flag prederror is set to true, the default, and if Maxima is unable to evaluate a predicate, it signals an error, and unless directed otherwise, returns control to the top-level Maxima command monitor. However, if the prederror flag is false, execution continues to the next statement, ignoring both then and else clauses! This is illustrated below: (C5) test(x,y):=block([], if x > y then print(x, "is greater than", y) else print(x, "is not greater than", y), return(alldone))$ (C6) test(4,3); 4 is greater than 3 (D6) alldone (C7) test(3,4); 3 is not greater than 4 (D7) alldone (C8) test(y^2+1,-y); MACSYMA was unable to evaluate the predicate: ERREXP1 #0: test(x=y^2+1,y=-y) -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) (C9) prederror:false$ (C10) test(y^2+1,-y); (D10) alldone Note that no message was printed for line d10, but the return value, alldone was displayed. 7.3 Assumptions What do YOU think? Is y2 + 1 > -y ? For this question to make sense, both sides of the inequality must be in the same ordered domain. We do not know, offhand, whether y can assume values which are matrices, complex numbers, sets, or even programs! If y were known to be real, or more specifically, positive real, a program could try some deduction. Maxima has some features of this nature, as illustrated below. (C11) assume(y>0); (D11) [y > 0] (C12) test(y^2+1,-y); CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 63 2 y + 1 is greater than - y (D12) alldone If we wish Maxima to forget that assumption, (C13) forget(y>0); (D13) [y > 0] 7.4 Arbitrary Numbers of Parameters Ambitious packages of programs have been written by many Maxima users. Sometimes the requirement that a command has a fixed number of arguments causes discomfort2. It is possible to write a Maxima program which counts the number of arguments it is given, sets default values for others, and does any number of clever things. A simple example is shown below. Note the way the left-hand-side of the := is set up. (C14) prog3([l]) := block( [], print ("l is bound to", l, "and l[1] is" ,l[1]), return(length(l)))$ (C15) prog3(a,b,c,d); l is bound to [a, b, c, d] and l[1] is a (D15) 4 (C16) prog3(a,b); l is bound to [a, b] and l[1] is a (D16) 2 7.5 Arrays Arrays are a useful data structure, and are provided in most programming languages. Maxima provides arrays, but does not require that they be declared, or that they have numeric (integer) index-sets. Rather than writing a program to fill up an array and then iterating through all elements, sometimes it is easier to describe a program to generate elements as they are called for. Such array-associated functions are often quite convenient. The usual way you set them up is to provide specific values for certain index values, and then let others be assigned as needed. Note carefully the use of := and :. (C17) a[4]:4*u; (D17) 4 u (C18) a[22/7]:%pi; (D18) %PI (C19) a[x]:mystery; (D19) mystery (C20) a[h]:=cos(h); (D20) a := COS(h) h (C21) a[3]; (D21) COS(3) (C22) a[x+1]; (D22) COS(x + 1) You might wonder what the value of a is, after all this. Most disappointing: (C23) a; (D23) a 2The language Pascal does not allow you to define a function f which can be used with a variable number of actual parameters, although the Pascal design includes built-in procedures with variable numbers of arguments (write for example). CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 64 The information that you are after is available this way: (C24) arrayinfo(a); 22 (D24) [HASHED, 1, [3], [--], [4], [x], [x + 1]] 7 The list of information supplied indicates several aspects of the array. It is hashed: uses the data-structure of a hash-table for storage (this is a common encoding trick discussed in data structure texts). The number of subscripts is 1. The specific indexes for which it has recorded values are listed. The array-associated function defined on line c20 can be displayed by dispfun. 7.6 Iteration For loops provide the major iteration facility in Maxima. Three examples which illustrate variants of this are the factorial functions below: (C25) f3(n) := block([temp], temp:1, for i:1 thru n do temp : temp*i, return(temp))$ (C26) f4(n) := block([temp], temp:1, for i:n step -1 thru 1 do temp : temp*i, return(temp))$ (C27) f5(n) := block([temp], temp:1, for i:n unless i <= 1 do (temp : temp*i , i:i-2), return(temp))$ Decrementing i by 2 in the previous program was needed because the default step size of 1 is added to i each time through. F5 is certainly a perverse program. Incidentally, return s from within a for exit from the loop, and not from the enclosing block. It is important to note that one can group a collection of statement to be done together with parentheses, as illustrated in f5. 7.7 Serious Business Most serious users of Maxima find that they are repeatedly using the same programs, and need to save them for another day. Some users also find they rarely get the programs or the data quite right the first time, and would rather type these things in to a text editor, and have Maxima gobble the text up from a file rather than the keyboard. Publication quality programs require comments and other features you are unlikely to want to type into Maxima interactively. Some people perfect a function or get an algebraic expression correct by typing definitions and commands into a file, say, newstuff, using an editor such as vi or emacs and then within Maxima type batch(newstuff);. If your filename has funny characters like periods or slashes, you must use quotes. For example, batch("/usr/local/maxima/demo.begin"). Maxima then reads the statements the file, assigning labels etc., and if there are syntax or other errors, prompts for help from the keyboard. If you want it to continue on to the next line after an error, type batcon(true);. Reading very large text files of programs and data can be slow using batch, and if you are not changing the text, you might prefer saving your environment in another way. You can use save(savedstuff,all); to save every named or labelled object in a Maxima system, on the file savedstuff in your working directory. (You better have write-permission or you'll get an error message.) Another time you can start up from where you left off by typing loadfile(savedstuff); into a fresh Maxima. You can load several saved files into a single system. Naturally if the files contain items with identical names, there is a potential for conflict. These will be resolved in favor of the last item read in. If you want to save only some material you have produced, say only the functions defined and the values of variables x, y, and z, you can type CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 65 save(savfunxyz,functions,x,y,z); A neat way to save a useful section of your environment is to (carefully) use kill to remove useless items first, and then save all that is left. Doing kill(labels,...) after first making sure that any useful result also has a name other than a c or d-label is sometimes a good start. You generally should not save every computation, since disk space is not infinite. 7.8 Hardcopy If you print out the files produced by save to show to your colleagues who are as yet unconvinced of the merits of Maxima you will be disappointed. While such files are ASCII character text, they do not make easy reading for persons uninitiated in various arcane matters. What you might want to do, then, is store human-readable versions of your output on a file. This is especially useful if you have a display terminal, or a slow hardcopy printer. Maxima provides a way of opening a file for echoing both input and output of the system. The command writefile(fn); where fn is a filename, starts the echoing, and closefile(); stops the echoing. If you want the output to be human-readable after being run through a typesetting program (TEX, for example), you can experiment with the output produced by setting typeset : true. (Suggestion: gcprint : false is a good idea.) 7.9 Return to Arrays and Functions Maxima provides a fascinating trick: arrays of functions. Imagine an array, each of whose elements is a function. For example, Legendre polynomials have an index, and an argument. Thus we can refer to P4(z2). Just as arrays can have an associated function, function-arrays can have such an associated function. Because we like to show off occasionally, and typesetting can be done by the Maxima system, we illustrate this feature here. This was produced by saving output in a file and running it through TEX. (%i31) p[n](x):=ratsimp(1/(2^n*n!)*diff((x^2-1)^n,x,n)); (%o31) pn(x) : = RATSIMP 1 2n n! DIFF x2 - 1 n ,x,n (%i32) p[4]; (%o32) LAMBDA [x], 35 x4 - 30 x2 + 3 8 (%i33) p[4](y+1); (%o33) 35 (y + 1)4 - 30 (y + 1)2 + 3 8 The lambda () notation for a raw function of one variable appears on line c29. This notation comes from the -calculus used to describe the programming language LISP, and illustrates the fact that Maxima can talk about such structures. You can generally ignore this fact, however. 7.10 More Useful Examples As has been indicated earlier, procedures in Maxima can be used for almost any classical programming purposes (e.g. numerical techniques, combinatorial search). We have already indicated some differences from a purely numerical language, such as FORTRAN. We have seen that in Maxima there are no required type declarations, and floating-point CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 66 numbers, while contagious in the sense that mixed-mode (floating-point + integer) is converted to floating-point, do not necessarily arise from certain calculations. For example, sin(3) normally results in sin(3) in Maxima rather than its floating-point equivalent. sin(3.0) will, however, return a floating-point number. The numerical subroutine below, a Newton's method zero-finder sets the flag numer to true to force Maxima to convert most expressions free of indeterminates to numbers. This program uses Newton's method to find a zero of the expression exp which depends on the variable var. The iteration starts with var=x0 and terminates when the expression, evaluated at the trial-point, has absolute value less than eps. The derivative of exp is computed algebraically, and its value is computed at various points as the search is being conducted: (C1) newton(exp,var,x0,eps):= /* 1 */ block([xn,s,numer], /* 2 */ numer:true, /* 3 */ s:diff(exp,var), /* 4 */ xn:x0, /* 5 */ while abs(subst(xn,var,exp)) > eps do xn:xn-subst(xn,var,exp)/subst(xn,var,s), return(xn) ) /* 8 */$ This procedure for Newton's method uses an explicit expression for the first argument exp (e.g. sin(x)*erf(2*x)-%e^x). It is not a function name, e.g. f as we used before. The use of such an expression is straightforward and probably the best way for a beginner. The resulting program is somewhat verbose, because, as illustrated in lines 6 and 7 above, it is necessary to substitute values for variables in the expression and its derivative, s, to get numbers. Note the setting of numer on line 3 to assure that the if statement on line 6 would get numerical values for its test. The rest of the procedure is the classical Newton iteration. The advantage of this procedure over a purely numerical one is that it takes advantage of the ability to compute the derivative of exp algebraically and automatically, once, before evaluating it at any point. Another item to observe here is the use of comments in the text of programs which are batch ed in to the system. /* This is a comment */ . There are often many different ways of expressing the same computation, and some of them may be substantially more convenient or efficient than others. While it may not make much of a difference in efficiency in this case, the following revision of the newton procedure illustrates some techniques you may find useful in Maxima. (C2) newton(exp,var,x0,eps):= /* 1 */ block([ xn:x0, s:diff(exp,var), numer:true], /* 2 */ define(f(var), exp), /* 3 */ define(fprime(var),s), /* 4 */ loop, if abs(f(xn)) < eps then return(xn), /* 5 */ xn: xn - f(xn)/fprime(xn), /* 6 */ go (loop) ) /* 7 */ $ Observe the list of local names at the beginning of the block, which are initialized at the same time they are declared on line 2. Lines 3 and 4 are interesting because they define two new functions, f and fprime each to have as their bodies, the values of exp and s. The Newton iteration is more easily observed in functional rather than substitution notation. An even smaller version of newton could be defined by using a single function for exp/diff(exp,var). Let us try this last function: (C3) h:expand((x-1)*(x-3)*(x-5)); 3 2 (D3) x - 9 x + 23 x - 15 (C4) newton(h,x,3.5,1.0e-10); (D4) 2.999999999994028 You might wonder how many iterations that took. One way is to use the very convenient debugging feature provided by Maxima which allows you to watch the assignment of values to variables. You set the variable setcheck to a list of the variables you wish to watch. (or all to watch all the variables.) CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 67 (C5) setcheck:[xn]$ (C6) newton(h,x,3.5,1.0e-10); xn set to 3.5 xn set to 2.923076923076923 xn set to 3.000228597554008 xn set to 2.999999999994028 (D6) 2.999999999994028 (That message [*flonum: ...] is a message from the garbage collector, mentioned in Section ??.) That tells us that only three iterations were needed in this case. We claimed that two functions were defined in running this program. To display a function using the same syntax that you would use to type it in, you use the grind3 command. (C7) grind(fprime); Error: cursorpos doesn't know position Fast links are on: do (si::use-fast-links nil) for debugging Error signalled by MACSYMA-TOP-LEVEL. Broken at ERROR. Type :H for Help. 7.11 Part Hacking An important tool for applications programs in Maxima the ability to extract and test parts of expressions. These are used for the definition or extension of algorithms which do various conditional manipulation of formulas. For example, you can write a symbolic differentiation algorithm for expressions by applying the rules below: d dx x = 1 d dx y = 0 (y = x) (1) d dx (u + v) = d dx u + d dx v (2) d dx (u v) = v d dx u + u d dx v (3) The technique we shall consider for implementing a version of the differentiation program is to take the expression apart using the part command, and use the rules above to guide the manipulation. First, we check to see if the expression is an atom (e.g. number, variable). Thus we begin our differentiation program as follows: newdiff(expr,var):= if atom(expr)=true then (if expr=var then 1 else 0) This fragment implements both parts of rule 1. If the if statement falls through to the else clause, then we have a composite expression. We then check what its leading operator is by selecting its zeroth part via part(expr,0). Based on its value we apply the appropriate rule. else if part(expr,0)=}+} then newdiff(part(expr,1),var) + newdiff(part(expr,2),var) else if part(expr,0)=}*} then part(expr,2)*newdiff(part(expr,1),var) + part(expr,1)*newdiff(part(expr,2),var) else 'newdiff(expr,var)$ 3When computers were slow and programs were large, reformatting programs took a long time. The name of a program at MIT to grind out LISP function definitions was grindef. The name was carried over to Maxima. Now, most LISP systems call this reformatting prettyprinting. CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 68 Note the last clause which returns a 'newdiff form, that is, quoted, as a result when the program doesn't know how to handle the leading operator. With some thought, you should now be able to extend newdiff to accommodate expressions including sin and cos as well as sums and products with more than two terms. (The Maxima language does not at the moment have a case statement, which would make this particular program look better.) Is this stiil true ???? 7.12 User Representation of Data There is no record or structure data-type constructor in Maxima but there is a way of doing something similar. If you have used a language like Pascal which has such a feature, you may appreciate its usefulness. Modern dialects of LISP use such structures, but when Maxima was first designed and initially implemented (in 1967), this was not in widespread use. It also influenced the user-level programming language. It is natural to have to deal with collections of information in writing programs. For example, you might wish to provide as input or output data, a pair of expressions representing a lower and an upper bound on a formula. One way of handling this is by making up a new function name, say bounds and using it as though it were a record designator or constructor, but never associating it with an algorithm. Thus bounds(1,x) could be an expression representing the ordered pair < 1,x >. Another example would be the use of integer_vector(1,3,5,7) to designate a sequence (presumably of arbitrary length, in general), of integers. There is a built-in form in Maxima namely a list, which can be used to deal with such collections. It is part of the LISP heritage indicated earlier that there was initially only one type of structure: lists of any length or element type in Maxima. A list, when printed, looks like square-brackets enclosing a sequence of items. Thus you could express the bounds above as [1,x ]. However, if you used lists for all collections, you would not know, on the face of it, whether [1,2] was an instance of bounds or integer_vector. Maxima allows you to designate functions you intend to treat as lists, although you can use a different header like bounds with each different type. Maxima makes available certain built-in functions which can then be used on such list-like constructions such as integer_vector. These are declared by, for example, declare(integer_vector,list). The built-in operations include cons, append, member, endcons, map, rest. list2:cons(element,list1) returns a (new) list2 which has the given element inserted at the beginning of list1; List1 and list2 have a common shared tail. list3:append(list1,list2) returns a (new) list3 which is a combination of the two lists, sharing a common tail with list2. Member(element,list) returns true or false depending on a test for membership. Endcons(element,list) returns an (unshared) list where the given element has been attached to the end of given list Map(fn,list) returns a new list where each element has had the function fn applied to it in turn. Rest(list,n) returns the part of the list beginning n items from the front. These functions are parts of the fundamental repertoire of the programming language LISP. The use of list-like objects should be considered whenever you are dealing with a collection of elements: a set of coordinates, a series of coefficients, a system of equations, a set of solutions, etc. Independent of the whole list operations above, Maxima has some selection and alteration operations which are available on the same collections of data by the use of a numeric index. If you wish to use these indexing facilities, as we will illustrate for the notion of complex below, you declare(complex,list). Then, if you define a complex number by x:complex(3,4) meaning that x has real part 3, and imaginary part 4, the notation x[1]:10; is supported, and changes the value of x to complex(10,4). The declaration explains to the Maxima system that the data structure for complex will be implemented (in effect) as a list of items, and should be decomposable using the semantics of the Maxima list-handling commands. In fact, both selection and alteration is supported, and if you set the notation up by (real:1,imag:2)$ you can use the following command: x[imag]:-x[imag] to change x to its complex conjugate. An important caution must be observed. When Maxima deals with compound structures, they are usually not recopied, and if there are two names for the same object and the object is changed, then both names refer to the changed object. If x and y refer to the same complex number, then changes to x[real] are also made to the corresponding component of y. If these items are to be kept separate, x:copylist(y); will give x a different representation, but whose value is the same as y 's. You can then change their components separately. Maxima's built-in lists mentioned earlier, which use square-brackets, can be altered and selected by indexing. There are two other compound structures in Maxima which we mention here, which may be useful for collections of data: arrays and matrices. The matrix form is useful for two-dimensional rectangular tables of data. The matrix data format is supported by a number of special commands oriented toward the conventional interpretation of matrices in mathematics: inverses, determinants, etc. Matrices can be used for other kinds of data, but you should be cautious CHAPTER 7. INTRODUCTION TO MAXIMA'S PROGRAMMING LANGUAGE 69 about using them in a manner that is too far distant from the conventional: arithmetic operations in Maxima in particular, have already been defined. Arrays are unusual in their flexibility in Maxima. They are usually treated as global tables, although they can be declared to be local to a function; they cannot be passed around as parameters as is the case with matrices. The names of arrays may be passed from program to program, but the data itself is not recopied, nor is it convenient to even make a copy of the data into another array. Hashed arrays are particularly ingenious, and have been illustrated earlier. Functions can be associated with arrays to provide new values for entries as they are required. At this point you should be able to make use of the Maxima manual to learn more details for representation of new data types as you develop your application. The true programming buff may also be interested in the macro-expansion capabilities of Maxima and its extensible syntax. At this point we would discourage the use of these facilities by novices, but encourage their use by persons willing to experiment in providing the most versatile user-oriented packages within the Maxima framework. There is a compilation facility which allows users to translate Maxima code into potentially faster running code. Since most of the time in most programs is used by calls to LISP programs, this is usually ineffective. In general, this should be avoided by novices and most experienced users, since time spent on this is more wisely spent on mathematical restructuring of the solution method, or (in the case of primarily numerical computation), using a numerical library routine written in a suitable language. CHAPTER 8 Graphics and Forms of Output In addition to some options with respect to the mathematical output from Maxima on the command line, the system is also capable of interfacing with external programs to provide graphing capabilities. You can also save Maxima sessions to reload later. 8.1 Options on the Command Line While for normal use the default settings are likely to be preferable, Maxima allows you to set some options with regards to how output is returned. 8.1.1. 1D vs. 2D The default output Maxima gives you when you evaluate an expression is 2D output, which basically means you have some visual feedback on the structure of things like fractions and integrals as ascii art, and the output expression is not one string. Normally this is a good thing, but if for any reason you wish to turn that feature off, it is quite possible to do so. Maxima has an internal variable regulating this feature called DISPLAY2D. Ordinarily it is set to TRUE, providing the typical ascii Maxima output you would get in a terminal session. Setting it to FALSE will cause returns to be a single string. Here is an example: (C1) 2*y*3*x/(4*x+y+3/4); 6 x y (D1) ----------- 3 y + 4 x + - 4 (C2) DISPLAY2D:FALSE; (D2) FALSE (C3) 2*y*3*x/(4*x+y+3/4); (D3) 6*x*y/(y+4*x+3/4) One important note here is that if you only wish to change the dimension of your output for one entry, you must set this variable by hand both before and after the command - i.e., ev(2/3,DISPLAY2D:FALSE) will not result in 2/3 unless the DISPLAY2D:FALSE command has already been entered - ev will not use it in its evaluation. 8.1.2. TeX Strings as Output As you probably already know, Maxima is able to output expressions in TeX format. This ability is used transparently in several places, but you can also request this output manually, using the tex(expression) command. 70 CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 71 (C5) A1:integrate(2*x+x^3+3*x^2,x); 4 x 3 2 (D5) -- + x + x 4 (C6) tex('integrate(sin(x)/(2*x+x^3+3*x^2),x)); $$\int {{{\sin x}\over{x^{3}+3\>x^{2}+2\>x}}}{\>dx}$$ (D6) FALSE (C7) tex(A1); $${{x^{4}}\over{4}}+x^{3}+x^{2}$$ (D7) FALSE This is useful if you are writing a paper without the benefit of Emaxima and wish to include a Maxima result - the above can be pasted directly into a LaTeX environment. 8.1.3. Writing a Session to a File This is actually a bit tricky, especially if you decide halfway through a session you wish to make a record of it. Fortunately, however, there are some tools you can use. Note: Need to figure out the subtlties of save() Writefile The basic command you will want to start running when you want to record a session. The syntax works like this: writefile("/home/user/maxima sessionoutput.txt")$ This will record the entire session into a text file called maximasessionoutput.txt in the /home/user directory. If you wish close this file, you simply use the command closefile("/home/user/maxima sessionoutput.txt")$ These files are not loadable as commands - it is merely a transcript of a session as it occurred. This is useful for basic text documentation preperation, or just basic saving of a session for printing. If you wish to record lines that were entered before you began writing the file to disk, you can use a command called PLAYBACK to get them into the record. (Discuss playback options here.) Creating BATCHable files - Stringout If the user wishes to create a file containing session information which may be loaded again into the system using the BATCH, then STRINGOUT is the command you want to use. There are several choices here - you can save specific input (listing C labels), include all C input lines by supplying the arguement INPUT, all of the functions you have defined by supplying the argument FUNCTIONS, and also all the values you have defined by supplying the arguement VALUES. Since this is potentially quite important in the creation of packages and specialized invironments, we will go into this with some detailed examples. Leťs imagine we want to create a package to calculate the volume of a cube, with certain default values in place. We begin by entering these commands in Maxima: (C1) defaultlength:35$ (C2) defaultheight:45$ (C3) defaultwidth:65$ (C4) volume(length,width,height):=length*width*height; (D4) volume(LENGTH, width, height) := LENGTH width height (C5) defaultvolume:volume(defaultlength,defaultwidth,defaultheight)$ (C6) defaultvolume; (D6) 102375 Of course, the last line is not strictly necessary, but it will serve to illustrate the difference between various options. Now, as a first example we will save all the lines of input to the file cube.mac: (C7) stringout("cube.mac",INPUT); CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 72 (D7) /home/user/cube.mac Notice the return from this command is the directory where the file is located. If we look at the contents of that file: defaultlength:35$ defaultheight:45$ defaultwidth:65$ volume(LENGTH,width,height):=LENGTH*width*height; defaultvolume:volume(defaultlength,defaultwidth,defaultheight)$ defaultvolume; We see that this is the series of commands we entered. If we were to load this file with the batch command (reference this somewhere) we would get precisely the same thing we got before. Now leťs try the FUNCTIONS and VALUES arguements: (C8) stringout("cubefunctions.mac",FUNCTIONS); (D8) /home/user/cubefunctions.mac (C9) stringout("cubevalues.mac",VALUES); (D9) /home/user/cubevalues.mac And we see the contents of those files are: cubefunctions.mac volume(LENGTH,width,height):=LENGTH*width*height; cubevalues.mac defaultlength:35; defaultheight:45; defaultwidth:65; defaultvolume:102375; So those are the basics of how output works. Experiment a little before you begin creating batch files, so you know in more detail what is saved and what isn't by various options. In the case of FUNCTIONS, the following will tell you what you have defined: (C10) DISPFUN(ALL); (E10) volume(LENGTH, width, height) := LENGTH width height (D10) DONE You will also want to examine the VALUES variable and the use of the PACKAGEFILE variable. CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 73 8.2 Graphics Maxima has, via external programs, the ability to produce 2D and 3D graphs. The information will also be recorded in a file called maxout.openmath. The information in that file will always be the raw openmath format data of the last plot command. 8.2.1. 2D function plotting 2D graphs are produced with the plot2d command. Perhaps the simplest way to introduce this command is to show it in action. plot2d([sin(x),cos(x),tan(x)],[x,0,2*%Pi],[y,-1,1]); These are just the basics - there are many options which can be set, but most of them are part of an overall plot options system which we will discuss later. In the examples section at the end of this chapter we will show some more 2D plots with various options set. 8.2.2. 3D Function Plotting This works just about like the 2D plotting, only you need to supply the proper parameters. Here again an example is the best teacher. Plot3d(r^.33*cos(th/3),[r,0,1],[th,0,6*%pi],['grid,12,80], ['transform_xy,polar_to_xy],['view_direction,1,1,1.4],['colour_z,true], [gnuplot_pm3d,true], [gnuplot_preamble,"set pm3d at s;unset surface"]); CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 74 8.3 Plot Options Maxima defines a list, called PLOT_OPTIONS, which controls most of the behavior of Maxima's plotting options. There are two ways of setting these options - you can set an option globally, using the SET_PLOT_OPTION command 1, or supply the new options as arguements to a plot command. If you want to check what your global options are currently set at, you can just type PLOT_OPTIONS; and the current state of the system will be listed. (%i1) PLOT_OPTIONS; (%o1) [[x, -3, 3], [y, - 3, 3], [t, - 3, 3], [GRID, 30, 30], [VIEW_DIRECTION, 1, 1, 1], [COLOUR_Z, FALSE], [TRANSFORM_XY, FALSE], [RUN_VIEWER, TRUE], [PLOT_FORMAT, GNUPLOT], [GNUPLOT_TERM, DEFAULT], [GNUPLOT_OUT_FILE, FALSE], [NTICKS, 10], [ADAPT_DEPTH, 10], [GNUPLOT_PM3D, FALSE], [GNUPLOT_PREAMBLE, ], [GNUPLOT_CURVE_TITLES, [DEFAULT]], [GNUPLOT_CURVE_STYLES, [with lines 3, with lines 1, with lines 2, with lines 5, with lines 4, with lines 6, with lines 7]], [GNUPLOT_DEFAULT_TERM_COMMAND, ], [GNUPLOT_DUMB_TERM_COMMAND, set term dumb 79 22], [GNUPLOT_PS_TERM_COMMAND, se# t size 1.5, 1.5;set term postscript eps enhanced color solid 24]] OK, leťs look at each of these options. * [x, - 3, 3] - This defines the X range. In order to change this range globally, use a command of the form SET_PLOT_OPTION([x,-5,5]); 1Setting an option globally only changes the option for the current Maxima session - upon restart, the original defaults will be restored. CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 75 * [y, - 3, 3] - This defines the Y range. In order to change this range globally, use a command of the form SET_PLOT_OPTION([y,-5,5]); * [t, - 3, 3] - t is the default range for the parameter in parametric plots. In order to change this range globally, use a command of the form SET_PLOT_OPTION([t,-5,5]); * [GRID, 30, 30] - This controls, in 3D plotting, the number of points used to draw the figure. The function is only calculated at a certain number of points - after that, linear approximations are drawn. Globally, usee a command of the form SET_PLOT_OPTION([GRID,40,35]); * [VIEW_DIRECTION, 1, 1, 1] - This option is specific to the case when the plot command outputs directly to postscript in 3D(See PLOT_FORMAT.) It determines the direction from which the 'camera' looks at the function, which is along a line parallel to the line from VIEW_DIRECTION to the origin. It only needs to be set in the case of postscript output; it is ignored otherwise. Globally, use a command of the form SET_PLOT_OPTION([VIEW_ DIRECTION,1.4,1.4,1.4]); * [COLOUR_Z, FALSE] - This refers also to the postscript output - if set to TRUE it provides a little color shading in the output. Form is SET_PLOT_OPTION([COLOUR_Z, TRUE]); * [TRANSFORM_XY, FALSE] - This appears to provide the ability to produce plots with different coordinate systems, but I am unsure of how to make it work. Need to get help here. * [RUN_VIEWER, TRUE] - If you only wish Maxima to output a file and not launch the graphical viewer, set this option to false. Remember, however, that if you wish to run multiple commands to generate data you will have to recover the information from the maxout.PLOT_FORMAT file each time, because each new plot command will overwrite it. Form is SET_PLOT_OPTION([RUN_VIEWER, FALSE]); * [PLOT_FORMAT, GNUPLOT] - This controls which program gets the output from Maxima for display. There are currently five viable options - GNUPLOT, OPENMATH, MGNUPLOT,, GEOMVIEW, and PS. PS is simply direct output to a postscript file, maxout.ps. Form is SET_PLOT_OPTION([PLOT_FORMAT, GEOMVIEW]); MGNUPLOT is a Tk based interface to GNUPLOT (included with Maxima) which provides a rudimentary GUI, but overall has fewer features. All of these programs are freely available. Geomview is currently Unix only, and is available at http://geomview.sourceforge.net as both source and binary. Openmath is distributed as part of Maxima. Gnuplot is widely available, with a homepage at http://www.gnuplot.info and the most current work at http://sourceforge.net/projects/gnuplot. Gnuplot can run on both Windows and Linux and is currently the default plotting solution. * [[GNUPLOT_TERM, DEFAULT] - This option is specific to the gnuplot plot format. It sets the output terminal type for gnuplot. Options are DEFAULT: The default gnuplot terminal is a separate graphical window, DUMB - produces an ASCII art approximation to graphics, and PS - use this option in conjunction with GNUPLOT_OUT_FILE in order to write gnuploťs output to a postscript file instead of the screen. Form is SET_PLOT_OPTION([GNUPLOT_TERM, PS]); * [GNUPLOT_OUT_FILE, FALSE] - This option is specific to the gnuplot plot format. It is used to set the name of the file to write to. Form is SET_PLOT_OPTION([GNUPLOT_OUT_FILE, "myplot.ps"]); * [NTICKS, 10] - Controls the initial number of points used for the adaptive plotting routine. Form is SET_PLOT_OPTION([NTICKS, 200]); * [ADAPT_DEPTH, 10] - Controls the maximum number of splittings used for the adaptive plotting routine. Form is SET_PLOT_OPTION([ADAPT_DEPTH, 5]); * [GNUPLOT_PM3D, FALSE] - This option is specific to the gnuplot plot format. Controls the usage PM3D mode, which has advanced 3D features. PM3D is only available in gnuplot versions after 3.7. Form is SET_PLOT_OPTION([GNUPLOT_PM3D, TRUE]); * [GNUPLOT_PREAMBLE, ] - This option is specific to the gnuplot plot format. Inserts gnuplot commands before the plot is drawn. Any valid gnuplot commands may be used. Multiple commands should be separated with a semi-colon. This is where any options not controlled by other Maxima gnuplot settings may be controlled. Form is SET_PLOT_OPTION([GNUPLOT_PREAMBLE, "set log y; set log x"]); CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 76 * [GNUPLOT_CURVE_TITLES, [DEFAULT]] - This option is specific to the gnuplot plot format. Controls the titles given in the plot key. The default value is DEFAULT, which automatically sets the title of each curve to the function plotted. If not DEFAULT, GNUPLOT_CURVE_TITLES should contain a list of strings. (To disable the plot key entirely, add "set nokey" to GNUPLOT_PREAMBLE.) Form is SET_PLOT_OPTION([GNUPLOT_CURVE_TITLES, ["my first function","my second function"]]); * [GNUPLOT_CURVE_STYLES, [with lines 3, with lines 1, with lines 2, with lines 5, with lines 4, with lines 6, with lines 7]] - This option is specific to the gnuplot plot format. This is a list of strings controlling the appearance of curves, i.e., color, width, dashing, etc., to be sent to the gnuplot plot command. See the gnuplot documentation for "plot" for more information. Form is SET_PLOT_OPTION([GNUPLOT_CURVE_STYLES, ["with lines 7", "with lines 2"]]); * [GNUPLOT_DEFAULT_TERM_COMMAND, ] - This option is specific to the gnuplot plot format. The gnuplot command to set the terminal type for the default terminal. The default value is "", i.e., use gnuploťs default. Form is SET_PLOT_OPTION([GNUPLOT_DEFAULT_TERM_COMMAND, "set term x11"]); * [GNUPLOT_DUMB_TERM_COMMAND, set term dumb 79 22] - This option is specific to the gnuplot plot format. The gnuplot command to set the terminal type for the dumb terminal. Form is SET_PLOT_OPTION([GNUPLOT_DUMB_TERM_COMMAND, "set term dumb 80 24"]); * [GNUPLOT_PS_TERM_COMMAND, set size 1.5, 1.5; set term postscript eps enhanced color solid 24] - This option is specific to the gnuplot plot format. The gnuplot command to set the terminal type for the postscript terminal. See the gnuplot documentation for "set term postscript" for more detailed information. CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 77 Figure 8.1: Graphing with Openmath, the default Maxima plotting tool CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 78 Figure 8.2: Graphing with gnuplot, using the mgnuplot utility CHAPTER 8. GRAPHICS AND FORMS OF OUTPUT 79 Figure 8.3: Graphing with GeomView. CHAPTER 9 Maxims for the Maxima User Some beginning users of algebraic manipulation systems find that their previous experiences with traditional programming systems do not translate easily into algebraic programming; others find Maxima descriptions inadequate because the emphasis is on the mixture of mathematical notations and algorithms, and not on efficient use of machine or human resources (no one likes to wait longer than necessary for an answer!). While we cannot provide a complete education in efficient and effective programming, we have collected a few maxims in an attempt to help you with some of these start-up problems. Algebraic manipulation is a new and different computational approach for which prior experience with computational methods may be misleading. You should attempt to learn the ways in which algebraic manipulation differs from other approaches. For example, consider the problem of inverting an n × n matrix. In numerical analysis we learn that the problem requires on the order of n3 operations. Theoreticians will point out that for n sufficiently large, one can reduce the number of multiplications below n3 to n2.8. This analysis is unfortunately not relevant in dealing with matrices with symbolic entries. Consider the number of terms in the determinant of the general n × n matrix whose elements are the symbols ai,j. When the inverse is written out in a fully expanded form, just the determinant (a necessary part of representing the inverse) has n! terms. It is impossible to compute this determinant in less than time proportional to n! In fact, for large n, it is just not feasible to compute this form explicitly on existing computers. The combinatorial or exponential character that some algebraic manipulation problems have when they are approached with an inefficient algorithm, makes for a vastly different game from, say, numerical computation, where the size of objects is generally known at the onset of the calculation, and does not increase. Needlessly generalizing a problem usually results in unnecessary expense. For example, if you wish to obtain determinants for a collection of matrices whose general pattern of entries is represented by parametric formulas, you might consider obtaining the determinant of the general matrix and substituting various values for the parameters into the result. This may work for matrices of low order, but is probably a poor plan for dealing with the exponential growth inherent in computing symbolic determinants. It would probably be better to substitute the parameters first, since this would drastically reduce the cost of the determinant calculation. Sometimes, when humans are dealing with formulas, it is preferable to use an indeterminate, say G in a formula which is really known to be, say, 3/5. On the other hand, it is likely (although not certain!) that the calculation using 3/5 will take less time than the calculation with G. Since the cost inherent in some computations is usually a function of the number of variables in the expression, it pays to reduce the number of variables in a problem as much as possible. You should be aware of the types of calculations which in the general case have exponential growth (e.g. many matrix calculations with symbolic entries, repeated differentiation of products or quotients, solution of systems of polynomial equations). Your should anticipate a certain amount of trial-and-error in calculations. 80 CHAPTER 9. MAXIMS FOR THE MAXIMA USER 81 Just as in other problem-solving activities, often the first technique that comes to mind is not the best. While it occasionally happens that brute force carries the day, cleverness in computing can be as important as cleverness in hand calculations. It is natural, during hand calculations, to apply critical simplificiations or substitutions in computations. These simplifications include collecting terms selectively or striking out terms which do not ultimately contribute to the final answer because of physical interpretations. Computer algorithms which do not incorporate these same tricks may bog down surprisingly soon. Thinking about these shortcuts may be important. In fact, it is one of the more rewarding aspects of computer algebra systems that they give the problem solver an opportunity to organize, encapsulate and distribute a particularly clever piece of mathematical manipulation. Try to reduce your problem so that it can be performed in a simpler domain. For example, if your problem appears to involve trigonometric functions, logs, exponentials, etc. see if you can reduce it to a rational function (ratio of polynomials) problem. If it appears to be a rational function, see if you can, by substitutions, make it into a polynomial problem, or a truncated power-series problem. If it appears to be a problem involving rational numbers, consider the use of floating-point numbers as an alternative, if the growth in the size of numbers presents difficulties. There are other special forms that are especially efficient. In a number of areas of investigation it pays to convert all expressions to the internal rational form (using rat) or into Poisson-series form (using intopois) to avoid the overhead the the general representation. The price you may pay here is thtat the structure of the formulas may be significantly different from those you began with: The canonical transformations used by these representations drastically re-order, expand, and modify the original expressions. Sometimes someone else has already started on your problem. You should look through the share directory programs available to see if there are contributed packages that might be of use either as subroutines or as models for programming. You should also consider writing programs that you develop in solving your problems in a form suitable for sharing with others. Pattern matching allows you to tune the system simplifier to your application, and develop rule-replacement programs. Learning to use the pattern-matching facilities effectively is a nontrivial task. Nevertheless if you have a fairly complex problem involvilng the recognition and application of identities, you should consider making an effort to use these facilities. In recent years, advocates of rule-based expert systems have claimed that this type of formalism can or should be used to incorporate varied types of knowledge. Algebraic manipulation programs have depended on pattern matching since at least 1961, for some of their power. Finally, we would like to point out that algebraic manipulation systems, in spite of their pitfalls, can be of major assistance in solving difficult problems. If you are willing to invest some time in learning, there may be enormous benefits in using such a system. We think it is unfortunate that some users reserve Maxima for difficult problems. Those of us who have grown up with Maxima near at hand find it of great use in routine computations as well. CHAPTER 10 Help Systems and Debugging Using the debugger and the online help system. 82 CHAPTER 11 Troubleshooting Common errors, their cause and solution. 83 CHAPTER 12 Advanced Examples These examples try to draw on everything in the manual to show you what can be done with Maxima in advanced usages. This is NOT a good starting point for new users. Establishing a Minimum for the Rayleigh Quotient We begin by defining the Rayleigh Quotient in general. From basic Regular Sturm-Liouville Eigenvalue principles, we know that the Rayleigh Quotient is defined as = -p d dx b a + R b a p d dx 2 - q2 dx R b a 2dx given the Sturm-Liouville differential equation d dx p(x) d dx + q(x) + (x) = 0 where a < x < b. Maxima RQ:(-p*('ev('ev(u(x)*'diff(u(x),x)),x=a)-'ev('ev(u(x)*diff(u(x),x)),x=b))+ 'integrate(p*'diff(u(x),x)^2-q*u(x)^2,x,a,b))/'integrate(u(x)^2*sigma,x,a,b); TEX Output R b a p d d x u(x) 2 - q u2 (x) dx - p EV EV u(x) d d x u(x) ,x = a - EV EV u(x) d d x u(x) ,x = b R b a u2 (x) dx Now we evaluate it. This must be done in stages, otherwise the ev command will not understand its arguements. Maxima ev(RQ,p=1,q=0,sigma=1,u(x)=x-x^2,a=0,b=1); ev(%,diff,integrate); ev(%,ev); TEX Output 84 CHAPTER 12. ADVANCED EXAMPLES 85 EV EV x - x2 d d x x - x2 ,x = 1 - EV EV x - x2 d d x x - x2 ,x = 0 + R 1 0 d d x x - x2 2 dx R 1 0 (x - x2)2 dx 30 EV EV (1 - 2 x) x - x2 ,x = 1 - EV EV (1 - 2 x) x - x2 ,x = 0 + 1 3 10 This can be checked by hand. Seeing that it is correct, we now can use it to search for the minimum eigenvalue on a more difficult problem: Maxima ev(RQ,p=1,q:-(x^2),sigma=1,u(x)=x-1,a=0,b=1)$ ev(%,diff,integrate)$ EV(%,EV,NUMER); TEX Output 6.1 Maxima ev(RQ,p=1,q:-(x^2),sigma=1,u(x)=-2*x^2+2,a=0,b=1)$ ev(%,diff,integrate)$ ev(%,ev,NUMER); TEX Output 2.6428571428571432 Maxima ev(RQ,p=1,q:-(x^2),sigma=1,u(x)=x^3+x^2-2,a=0,b=1)$ ev(%,diff,integrate)$ ev(%,ev,NUMER); TEX Output 2.7760840108401084 The smallest eigenvalue must therefore be less than or equal to 2.642857... Laplacian in Different Coordinate Systems This will probably go in the main documentation somewhere, but for now I'll stick it here. It is possible to express the Laplacian in different coordinate systems, provided you know how to define the coordinate system. We will use Spherical Coordinates for our first example: CHAPTER 12. ADVANCED EXAMPLES 86 Maxima load(vect)$ scalefactors([[rho*cos(theta)*sin(phi),rho*sin(theta)*sin(phi),rho*cos(phi)],rho,theta,phi]); depends(f,[rho,theta,phi]); express(laplacian(f)); ev(%,diff)$ ratexpand(%); TEX Output ; In: LAMBDA (X ANS A3) ; #'(LAMBDA (X ANS A3) NIL (COND # #)) ; Note: Variable A3 defined but never used. ; ; Note: Variable A3 defined but never used. ; ; Note: Variable A3 defined but never used. ; ; Note: Variable A3 defined but never used. DONE [f (,,)] d d d d f |sin| 2 + d d d d f |sin| sin2 + d d d d f |sin| |sin| 2 2 d d f + d d f cos sin 2 + d2 d 2 f sin2 2 + d2 d 2 f 2 + d2 d 2 f Part II External, Additional, and Contributed Packages 87 CHAPTER 13 The Concept of Packages - Expanding Maxima's Abilities This chapter will try to give some in depth information on the why and how of Maxima packages. It is true that many "standard" abilities of Maxima, such as vectors, are due to packages, but this part of the book will deal with more specialized contributed packages, such as elliptical integrals, special scientific packages, etc. That way the first part of the book can be rewritten less frequently, but we can still have all the information here. 88 CHAPTER 14 Algebra 89 CHAPTER 15 Calculus Packages: asympa, pdiff, qual 15.1 asympa 90 CHAPTER 15. CALCULUS 91 15.2 pdiff - Positional Derivatives Author: Barton Willis University of Nebraska at Kearney Kearney Nebraska Documentation adapted for the Maxima Book by CY Introduction Working with derivatives of unknown functions1 can be cumbersome in Maxima. If we want, for example, the first order Taylor polynomial of f(x+x2) about x = 1, we get (%i1) taylor(f(x + x^2),x,1,1); (%o1) f (2) + d d x f x2 + x x=1 (x - 1) + To "simplify" the Taylor polynomial, we must assign a gradient to f (%i2) gradef(f(x),df(x))$ (%i3) taylor(f(x+x^2),x,1,1); (%o3) f (2) + 3 df(2) (x - 1) + This method works well for simple problems, but it is tedious for functions of several variables or high order derivatives. The positional derivative package pdiff gives an alternative to using gradef when working with derivatives of unknown functions. Usage To use the positional derivative package, you must load it from a Maxima prompt. Assuming pdiff.lisp is in a directory that Maxima can find, this is done with the command (%i4) kill(all)$ (%i1) load("pdiff.lisp")$ Use the full pathname if Maxima can't find the file. Note that the kill(all) is needed because the gradef definition will conflict with the proper functioning of the diff commands. Loading pdiff.lisp sets the option variable use_pdiff to true; when use_diff is true, Maxima will indicate derivatives of unknown functions positionally. To illustrate, the first three derivatives of f are (%i2) [diff(f(x),x), diff(f(x),x,2), diff(f(x),x,3)]; (%o2) f(1)(x), f(2)(x), f(3)(x) The subscript indicates the order of the derivative; since f is a function of one variable, the subscript has only one index. When a function has more than one variable, the subscript has an index for each variable 1By unknown function, we mean a function that isn't bound to a formula and that has a derivative that isn't known to Maxima. CHAPTER 15. CALCULUS 92 (%i3) [diff(f(x,y),x,0,y,1), diff(f(y,x),x,0,y,1)]; (%o3) f(0,1)(x,y), f(1,0)(y,x) Setting use_pdiff to false (either locally or globally) inhibits derivatives from begin computed positionally (%i4) diff(f(x,x^2),x), use_pdiff : false; (%o4) d d x f x,x2 (%i5) diff(f(x,x^2),x), use_pdiff : true; (%o5) f(1,0)(x,x2 ) + 2x f(0,1)(x,x2 ) Taylor polynomials of unknown functions can be found without using gradef. An example (%i6) taylor(f(x+x^2),x,1,2); (%o6) f (2) + 3 f(1)(2) (x - 1) + 2 f(1)(2) + 9 f(2)(2) (x - 1)2 2 + Additionally, we can verify that y = f(x-ct)+g(x+ct) is a solution to a wave equation without using gradef (%i7) y : f(x-c*t) + g(x+c*t)$ (%i8) ratsimp(diff(y,t,2) - c^2 * diff(y,x,2)); (%o8) 0 (%i9) remvalue(y)$ Expressions involving positional derivatives can be differentiated (%i10) e : diff(f(x,y),x); (%o10) f(1,0)(x,y) (%i11) diff(e,y); (%o11) f(1,1)(x,y) The chain rule is applied when needed (%i12) [diff(f(x^2),x), diff(f(g(x)),x)]; (%o12) 2x f(1)(x2 ),g(1)(x) f(1)(g(x)) The positional derivative package doesn't alter the way known functions are differentiated (%i13) diff(exp(-x^2),x); CHAPTER 15. CALCULUS 93 (%o13) -2xe-x2 To convert positional derivatives to standard Maxima derivatives, use convert_to_diff (%i14) e : [diff(f(x),x), diff(f(x,y),x,1,y,1)]; (%o14) f(1)(x), f(1,1)(x,y) (%i15) e : convert_to_diff(e); (%o15) d d x f (x), d2 d yd x f (x,y) To convert back to a positional derivative, use ev with diff as an argument (%i16) ev(e,diff); (%o16) f(1)(x), f(1,1)(x,y) Conversion to standard derivatives sometimes requires the introduction of a dummy variable. Here's an example (%i17) e : diff(f(x,y),x,1,y,1); (%o17) f(1,1)(x,y) (%i18) e : subst(p(s),y,e); (%o18) f(1,1)(x, p(s)) (%i19) e : convert_to_diff(e); (%o19) d2 d %x0 d x f (x,%x0) [%x0=p(s)] Dummy variables have the form ci, where i=0,1,2...and c is the value of the option variable dummy_char. The default value for dummy_char is %x. If a user variable conflicts with a dummy variable, the conversion process can give an incorrect value. To convert the previous expression back to a positional derivative, use ev with diff and at as arguments (%i20) ev(e,diff,at); (%o20) f(1,1)(x, p(s)) Maxima correctly evaluates expressions involving positional derivatives if a formula is later given to the function. (Thus converting an unknown function into a known one.) Here is an example; let (%i21) e : diff(f(x^2),x); CHAPTER 15. CALCULUS 94 (%o21) 2x f(1)(x2 ) Now, give f a formula (%i22) f(x) := x^5; (%o22) f (x) : = x5 and evaluate e (%i23) ev(e); (%o23) 10x9 This result is the same as (%i24) diff(f(x^2),x); (%o24) 10x9 In this calculation, Maxima first evaluates f(x) to x10 and then does the derivative. Additionally, we can substitute a value for x before evaluating (%i25) ev(subst(2,x,e)); (%o25) 5120 We can duplicate this with (%i26) subst(2,x,diff(f(x^2),x)); (%o26) 5120 (%i27) remfunction(f); (%o27) [f] We can also evaluate a positional derivative using a local function definition (%i28) e : diff(g(x),x); (%o28) g(1)(x) (%i29) e, g(x) := sqrt(x); (%o29) 1 2 x CHAPTER 15. CALCULUS 95 (%i30) e, g = sqrt; (%o30) 1 2 x (%i31) e, g = h; (%o31) h(1)(x) (%i32) e, g = lambda([t],t^2); (%o32) 2x The pderivop function If F is an atom and i1,i2,...in are nonnegative integers, then pderivop(F,i1,i2,...in), is the function that has the formula i1+i2++in xi1 1 xi2 2 xin n F(x1,x2,...xn). If any of the derivative arguments are not nonnegative integers, we'll get an error (%i33) pderivop(f,2,-1); Each derivative order must be a nonnegative integer -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) The pderivop function can be composed with itself (%i34) pderivop(pderivop(f,3,4),1,2); (%o34) f(4,6) If the number of derivative arguments between two calls to pderivop isn't the same, Maxima gives an error (%i35) pderivop(pderivop(f,3,4),1); The function f expected 2 derivative argument(s), but it received 1 -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) When pderivop is applied to a known function, the result is a lambda form2 (%i36) f(x) := x^2; (%o36) f (x) : = x2 (%i37) df : pderivop(f,1); (%o37) ([G2491],2G2491) 2If you repeat theses calculations, you may get a different prefix for the gensym variables. CHAPTER 15. CALCULUS 96 (%i38) apply(df,[z]); (%o38) 2z (%i39) ddf : pderivop(f,2); (%o39) ([G2492],2) (%i40) apply(ddf,[10]); (%o40) 2 (%i41) remfunction(f); (%o41) [f] If the first argument to pderivop is a lambda form, the result is another lambda form (%i42) f : pderivop(lambda([x],x^2),1); (%o42) ([G2493],2G2493) (%i43) apply(f,[a]); (%o43) 2a (%i44) f : pderivop(lambda([x],x^2),2); (%o44) ([G2494],2) (%i45) apply(f,[a]); (%o45) 2 (%i46) f : pderivop(lambda([x],x^2),3); (%o46) ([G2495],0) (%i47) apply(f,[a]); (%o47) 0 (%i48) remvalue(f)$ If the first argument to pderivop isn't an atom or a lambda form, Maxima will signal an error (%i49) pderivop(f+g,1); Non-atom g+f used as a function -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) CHAPTER 15. CALCULUS 97 You may use tellsimpafter together with pderivop to give a value to a derivative of a function at a point; an example (%i50) tellsimpafter(pderivop(f,1)(1),1)$ (%i51) tellsimpafter(pderivop(f,2)(1),2)$ (%i52) diff(f(x),x,2) + diff(f(x),x)$ (%i53) subst(1,x,%); (%o53) 3 This technique works for functions of several variables as well (%i54) kill(rules)$ (%i55) tellsimpafter(pderivop(f,1,0)(0,0),a)$ (%i56) tellsimpafter(pderivop(f,0,1)(0,0),b)$ (%i57) sublis([x = 0, y = 0], diff(f(x,y),x) + diff(f(x,y),y)); (%o57) b + a TEX-ing positional derivatives Several option variables control how positional derivatives are converted to TEX. When the option variable tex_uses_prime_for_deriv is true (default false), makes functions of one variable TEX as superscripted primes (%i58) tex_uses_prime_for_derivatives : true$ (%i59) tex(makelist(diff(f(x),x,i),i,1,3))$ f (x), f (x), f (x) (%i60) tex(makelist(pderivop(f,i),i,1,3))$ f , f , f When the derivative order exceeds the value of the option variable tex_prime_limit, (default 3) derivatives are indicated with parenthesis delimited superscripts (%i61) tex(makelist(pderivop(f,i),i,1,5)), tex_prime_limit : 0$ f(1) , f(2) , f(3) , f(4) , f(5) (%i62) tex(makelist(pderivop(f,i),i,1,5)), tex_prime_limit : 5$ f , f , f , f , f The value of tex_uses_prime_for_derivatives doesn't change the way functions of two or more variables are converted to TEX. (%i63) tex(pderivop(f,2,1))$ f(2,1) When the option variable tex_uses_named_subscripts_for_derivatives (default false) is true, a derivative with respect to the i-th argument is indicated by a subscript that is the i-th element of the option variable tex_diff_var_names. An example is the clearest way to describe this. (%i64) tex_uses_named_subscripts_for_derivatives : true$ (%i65) tex_diff_var_names; CHAPTER 15. CALCULUS 98 (%o65) [x,y,z] (%i66) tex([pderivop(f,1,0), pderivop(f,0,1), pderivop(f,1,1), pderivop(f,2,0)])$ [fx, fy, fxy, fxx] (%i67) tex_diff_var_names : [a,b]; (%o67) [a,b] (%i68) tex([pderivop(f,1,0), pderivop(f,0,1), pderivop(f,1,1), pderivop(f,2,0)])$ [fa, fb, fab, faa] (%i69) tex_diff_var_names : [x,y,z]; (%o69) [x,y,z] (%i70) tex([diff(f(x,y),x), diff(f(y,x),y)])$ [fx(x,y), fx(y,x)] When the derivative order exceeds tt tex_prime_limit, revert to the default method for converting to TEX (%i71) tex(diff(f(x,y,z),x,1,y,1,z,1)), tex_prime_limit : 4$ fxyz(x,y,z) (%i72) tex(diff(f(x,y,z),x,1,y,1,z,1)), tex_prime_limit : 1$ f(1,1,1)(x,y,z) A longer example We'll use the positional derivative package to change the independent variable of the differential equation (%i73) de : 4*x^2*'DIFF(y,x,2) + 4*x*'DIFF(y,x,1) + (x-1)*y = 0; (%o73) 4x2 d2 d x2 y + 4x d d x y + (x - 1) y = 0 With malice aforethought, we'll assume a solution of the form y = g(xn), where n is a number. Substituting y g(xn) in the differential equation gives (%i74) de : subst(g(x^n),y,de); (%o74) 4x2 d2 d x2 g(xn ) + 4x d d x g(xn ) + (x - 1) g(xn ) = 0 (%i75) de : ev(de, diff); (%o75) 4x2 n2 x2n-2 g (xn ) + (n - 1) nxn-2 g (xn ) + 4nxn g (xn ) + (x - 1) g(xn ) = 0 Now let x t1/n (%i76) de : radcan(subst(x^(1/n),x, de)); CHAPTER 15. CALCULUS 99 (%o76) 4n2 x2 g (x) + 4n2 xg (x) + x 1 n - 1 g(x) = 0 Setting n 1/2, we recognize that g is the order 1 Bessel equation (%i77) subst(1/2,n, de); (%o77) x2 g (x) + xg (x) + x2 - 1 g(x) = 0 Limitations * Positional derivatives of subscripted functions are not allowed. * Derivatives of unknown functions with symbolic orders are not computed positionally. * The pdiff.lisp code alters the Maxima functions mqapply and sdiffgrad Although the author is unaware of any problems associated with these altered functions, there may be some. Setting use_pdiff to false should restore mqapply and sdiffgrad to their original functioning. CHAPTER 15. CALCULUS 100 15.3 qual CHAPTER 16 Combinatorics 101 CHAPTER 17 Differential Equations 102 CHAPTER 18 Graphics 103 CHAPTER 19 Integequations 104 CHAPTER 20 Integration 105 CHAPTER 21 Macro 106 CHAPTER 22 Matrix 107 CHAPTER 23 Numeric 108 CHAPTER 24 Physics Packages: dimen, dimension, physconst 24.1 dimen 109 CHAPTER 24. PHYSICS 110 24.2 dimension - Advanced Dimensional Analysis Author: Barton Willis University of Nebraska at Kearney Kearney Nebraska Documentation adapted for the Maxima Book by CY Introduction This document demonstrates some of the abilities of a Maxima package called dimension. Not surprisingly, its purpose is to perform dimensional analysis. Maxima comes with an older package dimensional analysis (dimen) that is similar to the one that was in the commercial Macsyma system. The software described in this document differs greatly from the older one. Usage To use the package, you must first load it. From a Maxima prompt, this is done using the command (%i1) load("dimension.mac")$ To begin, we need to assign dimensions to the variables we want to use. Use the qput function to do this; for example, to declare x a length, c a speed, and t a time, use the commands (%i2) qput(x, "length", dimension)$ (%i3) qput(c, "length" / "time", dimension)$ (%i4) qput(t, "time", dimension)$ We've defined the dimensions length and time to be strings; doing so reduces the chance that they will conflict with other user variables. To declare a dimensionless variable , use 1 for the dimension. Thus (%i5) qput(sigma,1,dimension)$ To find the dimension of an expression, use the dimension function. For example (%i6) dimension(4 * sqrt(3) /t); (%o6) 1 time (%i7) dimension(x + c * t); (%o7) length (%i8) dimension(sin(c * t / x)); (%o8) 1 (%i9) dimension(abs(x - c * t)); (%o9) length (%i10) dimension(sigma * x / c); (%o10) time CHAPTER 24. PHYSICS 111 (%i11) dimension(x * sqrt(1 - c * t / x)); (%o11) length dimension applies logcontract to its argument; thus expressions involving a difference of logarithms with dimensionally equal arguments are dimensionless; thus (%i12) dimension(log(x) - log(c*t)); (%o12) 1 dimension is automatically maps over lists. Thus (%i13) dimension([42, min(x,c*t), max(x,c*t), x^^4, x . c]); (%o13) 1,length,length,length4 , length2 time When an expression is dimensionally inconsistent, dimension should signal an error (%i14) dimension(x + c); Expression is dimensionally inconsistent. #0: dimension(e=x+c)(dimension.mac line 154) -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) (%i15) dimension(sin(x)); Expression is dimensionally inconsistent. #0: dimension(e=SIN(x))(dimension.mac line 229) -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) An equation is dimensionally correct when either the dimensions of both sides match or when one side of the equation vanishes. For example (%i16) dimension(x = c * t); (%o16) length (%i17) dimension(x * t = 0); (%o17) length time When the two sides of an equation have different dimensions and neither side vanishes, dimension signals an error (%i18) dimension(x = c); Expression is dimensionally inconsistent. #0: dimension(e=x = c)(dimension.mac line 175) -- an error. Quitting. To debug this try DEBUGMODE(TRUE);) CHAPTER 24. PHYSICS 112 The function dimension works with derivatives and integrals (%i19) dimension('diff(x,t)); (%o19) length time (%i20) dimension('diff(x,t,2)); (%o20) length time2 (%i21) dimension('diff(x,c,2,t,1)); (%o21) time length (%i22) dimension('integrate (x,t)); (%o22) length time Thus far, any string may be used as a dimension; the other three functions in this package, dimension_as_list , dimensionless, and natural_unit all require that each dimension is a member of the list fundamental_dimensions. The default value is of this list is (%i23) fundamental_dimensions; (%o23) [mass,length,time] A user may insert or delete elements from this list. The function dimension_as_list returns the dimension of an expression as a list of the exponents of the fundamental dimensions. Thus (%i24) dimension_as_list(x); (%o24) [0,1,0] (%i25) dimension_as_list(t); (%o25) [0,0,1] (%i26) dimension_as_list(c); (%o26) [0,1,-1] (%i27) dimension_as_list(x/t); (%o27) [0,1,-1] (%i28) dimension_as_list("temp"); (%o28) [0,0,0] CHAPTER 24. PHYSICS 113 In the last example, "temp" isn't an element of fundamental_dimensions; thus, dimension_as_list reports that "temp" is dimensionless. To correct this, append "temp" to the list fundamental_dimensions (%i29) fundamental_dimensions : endcons("temp", fundamental_dimensions); (%o29) [mass,length,time,temp] Now we have (%i30) dimension_as_list(x); (%o30) [0,1,0,0] (%i31) dimension_as_list(t); (%o31) [0,0,1,0] (%i32) dimension_as_list(c); (%o32) [0,1,-1,0] (%i33) dimension_as_list(x/t); (%o33) [0,1,-1,0] (%i34) dimension_as_list("temp"); (%o34) [0,0,0,1] To remove "temp" from fundamental_dimensions, use the delete command (%i35) fundamental_dimensions : delete("temp", fundamental_dimensions)$ The function dimensionless finds a basis for the dimensionless quantities that can be formed from a list of dimensioned quantities. For example (%i36) dimensionless([c,x,t]); Dependent equations eliminated: (1) (%o36) ct x ,1 (%i37) dimensionless([x,t]); Dependent equations eliminated: (1) (%o37) [1] In the first example, every dimensionless quantity that can be formed as a product of powers of c,x, and t is a power of ct/x; in the second example, the only dimensionless quantity that can be formed from x and t are the constants. CHAPTER 24. PHYSICS 114 The function natural_unit(e,[v1,v2,...,vn]) finds powers p1, p2,... pn such that dimension(e) = dimension(v p1 1 v p2 2 ...vpn n ). Simple examples are (%i38) natural_unit(x,[c,t]); Dependent equations eliminated: (1) (%o38) [ct] (%i39) natural_unit(x,[x,c,t]); Dependent equations eliminated: (1) (%o39) [x] Here is a more complex example; we'll study the Bohr model of the hydrogen atom using dimensional analysis. To make things more interesting, we'll include the magnetic moments of the proton and electron as well as the universal gravitational constant in with our list of physical quantities. Let h be Planck's constant, e the electron charge, e the magnetic moment of the electron, p the magnetic moment of the proton, me the mass of the electron, mp the mass of the proton, G the universal gravitational constant, and c the speed of light in a vacuum. For this problem, we might like to display the square root as an exponent instead of as a radical; to do this, set sqrtdispflag to false (%i40) SQRTDISPFLAG : false$ Assuming a system of units where Coulomb's law is force = product of charges distance2 , we have (%i41) qput(%hbar, "mass" * "length"^2 / "time",dimension)$ (%i42) qput(%%e, "mass"^(1/2) * "length"^(3/2) / "time",dimension)$ (%i43) qput(%mue, "mass"^(1/2) * "length"^(5/2) / "time",dimension)$ (%i44) qput(%mup, "mass"^(1/2) * "length"^(5/2) / "time",dimension)$ (%i45) qput(%me, "mass",dimension)$ (%i46) qput(%mp, "mass",dimension)$ (%i47) qput(%g, "length"^3 / ("time"^2 * "mass"), dimension)$ (%i48) qput(%c, "length" / "time", dimension)$ The numerical values of these quantities may defined using numerval. We have (%i49) numerval(%%e, 1.5189073558044265d-14*sqrt(kg)*meter^(3/2)/sec)$ (%i50) numerval(%hbar, 1.0545726691251061d-34*kg*meter^2/sec)$ (%i51) numerval(%c, 2.99792458d8*meter/sec)$ (%i52) numerval(%me, 9.1093897d-31*kg)$ (%i53) numerval(%mp, 1.6726231d-27*kg)$ To begin, leťs use only the variables e,c,h,me, and mp to find the dimensionless quantities. We have (%i54) dimensionless([%hbar, %me, %mp, %%e, %c]); (%o54) me mp , c h e2 ,1 CHAPTER 24. PHYSICS 115 The second element of this list is the reciprocal of the fine structure constant. To find numerical values, use float (%i55) float(%); (%o55) 5.446169970987487 × 10-4 ,137.03599074450503,1.0 The natural units of energy are given by (%i56) natural_unit("mass" * "length"^2 / "time"^2, [%hbar, %me, %mp, %%e, %c]); (%o56) c2 me, c3 h mp e2 Leťs see what happens when we include e,p, and G. We have (%i57) dimensionless([%hbar, %%e, %mue, %mup, %me, %mp, %g, %c]); (%o57) p e , c2 me e e3 , c2 mp e e3 , e4 G c4 2 e , c h e2 ,1 To find the natural units of mass, length, time, speed, force, and energy, use the commands (%i58) natural_unit("mass", [%hbar, %%e, %me, %mp, %mue, %mup, %g, %c]); (%o58) mp, c2 m2 e e e3 , c2 m2 e p e3 , G m3 e e2 , c h me e2 (%i59) natural_unit("length", [%hbar, %%e, %me, %mp, %mue, %mup, %g, %c]); (%o59) e2 mp c2 m2 e , e e , p e , G me c2 , h c me (%i60) natural_unit("time", [%hbar, %%e, %me, %mp, %mue, %mup, %g, %c]); (%o60) e2 mp c3 m2 e , e e c , p e c , G me c3 , h c2 me (%i61) natural_unit("mass" * "length" / "time"^2, [%hbar, %%e, %me, %mp, %mue, %mup, %g, %c]); (%o61) c4 me mp e2 , c6 m3 e e e5 , c6 m3 e p e5 , c4 G m4 e e4 , c5 h m2 e e4 (%i62) natural_unit("mass" * "length"^2 / "time"^2, [%hbar, %%e, %me, %mp, %mue, %mup, %g, %c]); (%o62) c2 mp, c4 m2 e e e3 , c4 m2 e p e3 , c2 G m3 e e2 , c3 h me e2 CHAPTER 24. PHYSICS 116 The first element of this list is the rest mass energy of the proton. The dimension package can handle vector operators such as dot and cross products, and the vector operators div, grad, and curl. To use the vector operators, we'll first declare them (%i63) prefix(div)$ (%i64) prefix(curl)$ (%i65) infix("~")$ Leťs work with the electric and magnetic fields; again assuming a system of units where Coulomb's law is force = product of charges distance2 the dimensions of the electric and magnetic field are (%i66) qput(e, sqrt("mass") / (sqrt("length") * "time"), dimension)$ (%i67) qput(b, sqrt("mass") / (sqrt("length") * "time"),dimension)$ (%i68) qput(rho, sqrt("mass")/("time" * "length"^(3/2)), dimension)$ (%i69) qput(j, sqrt("mass") / ("time"^2 * sqrt("length")), dimension)$ Finally, declare the speed of light c as (%i70) qput(c, "length" / "time", dimension); (%o70) length time Leťs find the dimensions of E 2,EB, B 2, and E×B/c. We have (%i71) dimension(e.e); (%o71) mass length time2 (%i72) dimension(e.b); (%o72) mass length time2 (%i73) dimension(b.b); (%o73) mass length time2 (%i74) dimension((e ~ b) / c); (%o74) mass length2 time The physical significance of these quantities becomes more apparent if they are integrated over R3. Defining (%i75) qput(v, "length"^3, dimension); (%o75) length3 CHAPTER 24. PHYSICS 117 We now have (%i76) dimension('integrate(e.e, v)); (%o76) length2 mass time2 (%i77) dimension('integrate(e.b, v)); (%o77) length2 mass time2 (%i78) dimension('integrate(b.b, v)); (%o78) length2 mass time2 (%i79) dimension('integrate((e ~ b) / c,v)); (%o79) length mass time Iťs clear that E 2,EB and B 2 are energy densities while E×B/c is a momentum density. Leťs also check that the Maxwell equations are dimensionally consistent. (%i80) dimension(DIV(e)= 4*%pi*rho); (%o80) mass 1 2 length 3 2 time (%i81) dimension(CURL(b) - 'diff(e,t) / c = 4 * %pi * j / c); (%o81) mass 1 2 length 3 2 time (%i82) dimension(CURL(e) + 'diff(b,t) / c = 0); (%o82) mass 1 2 length 3 2 time (%i83) dimension(DIV(b) = 0); (%o83) mass 1 2 length 3 2 time CHAPTER 24. PHYSICS 118 24.3 physconst - Definitions for Physical Constants CHAPTER 25 Simplification 119 CHAPTER 26 Special Functions 120 CHAPTER 27 Sym 121 CHAPTER 28 Tensor 122 CHAPTER 29 Trigonometry 123 CHAPTER 30 Utils 124 CHAPTER 31 Vector 125 Part III Installing, Resources, Misc. 126 CHAPTER 32 Installing Maxima 32.1 Requirements Your basic needs are a computer running Windows, Linux or MacOSX and a supported Lisp implimentation. Currently on Linux Maxima will build on CMUCL (18e recommended), GCL (2.5.0 or greater) and CLisp (2.29, 2.31 or greater - 2.30 won't work properly). On Windows builds have been achieved with Clisp and GCL - GCL is used in the standard binary. On MacOSX Maxima is compiled using OpenMCL. Note that Lisp choice is not an either/or situation - if you have multiple lisp implimentations available you can build on all of them and select at runtime which Lisp you would like to use. While it is possible to build your own Windows or Mac OS X binary, it is quite difficult to do so (especially on Windows) and unless there is a real need we recommend you use the provided binaries for those platforms. For Windows a Wizard based install has been created using InnoSetup which should look and feel very familiar to most Windows users. On MacOSX you need to use either the DarwinPorts? or Other way? tools to download the binaries. 32.2 Source Based Installation on Linux Note: we assume here that your machine has development libraries and tools installed. If you get file/feature not found failures during configure it is probably because your distribution doesn't have the development libraries and tools you need installed. 32.2.1. Configure (Note - this discussion assumes you are using Maxima 5.9.0 or greater to build with. Older versions have a terrible build system and are no longer] supported.) Your first task (assuming you are in the top level directory of the maxima source file hierchy) is to determine which version(s) of Lisp you intend to build on. If you are building with multiple Lisps, you should specify which one you want to be your default Lisp (the Lisp Maxima will run with if you do not tell it otherwise.) In the example below, options are given to enable building on all three Lisps supported by Maxima 5.9.0. The default Lisp of CLisp is selected. ./configure -enable-gcl -enable-cmucl -enable-clisp -with-default-lisp=clisp You should see some output scroll by and then a summary similar to the following: Summary: clisp enabled. Executable name: "clisp" CMUCL enabled. Executable name: "lisp" 127 CHAPTER 32. INSTALLING MAXIMA 128 GCL enabled. Executable name: "gcl" default lisp: clisp wish executable name: "wish" The wish executable is related to Tcl/Tk used for the Xmaxima gui. If you encounter a case where a Lisp executable name is not found or you wish to use a different version of a particular lisp, you can specify the location of the executible you wish to use. For example, if you have a different copy of CMUCL you wish to use with an executable name of cmulisp instead of lisp, you can specify that with the following: ./configure -enable-gcl -enable-cmucl -enable-clisp -with-cmucl=/usr/local/bin/cmulisp -with-default-lisp=clisp Summary: clisp enabled. Executable name: "clisp" CMUCL enabled. Executable name: "/usr/local/bin/cmulisp" GCL enabled. Executable name: "gcl" default lisp: clisp wish executable name: "wish" There are other options available for configure, but these should be enough to get you started on a standard Linux system. If you need more options, check the output from ./configure -help 32.2.2. Make Once you have configured the program to your satisfaction, simply type make. You will see a very long series of outputs as Maxima is compiled on each Lisp platform you have selected. This is a long process even on fairly fast machines. Once this process is done, you should run make check. This will run each build of Maxima through a series of mathematical tests to ensure your Maxima build succeeded. You will see something similar to the following, depending on which Lisp(s) you compiled with. (Plus some make related output about checking directories for tasks which doesn't matter): make[1]: Entering directory `/home/user/maxima/tests' echo "Running test suite with clisp..."; /bin/sh ../maxima-local -lisp=clisp -batch-lisp=tests.lisp > tests-clisp.log < /dev/null 2>&1; ./summarize-log tests-clisp.log Running test suite with clisp... *** Summary results for tests recorded in *** log file tests-clisp.log: Error summary: Error(s) found in rtest15.mac: (4) Expected failures (known bugs in this version of Maxima): rtest15.mac: (4) Timing: Real time: 9.218797 sec. Run time: 9.01 sec. GC: 60, GC time: 0.64 sec. CHAPTER 32. INSTALLING MAXIMA 129 *** end of summary for tests-clisp.log echo "Running test suite with cmucl..."; /bin/sh ../maxima-local -lisp=cmucl -batch-lisp=tests.lisp > tests-cmucl.log < /dev/null 2>&1; ./summarize-log tests-cmucl.log Running test suite with cmucl... *** Summary results for tests recorded in *** log file tests-cmucl.log: Error summary: Error(s) found in rtest15.mac: (4) Expected failures (known bugs in this version of Maxima): rtest15.mac: (4) Timing: ; 3.91f0 seconds of real time ; 3.34f0 seconds of user run time ; 0.49f0 seconds of system run time ; [Run times include 0.21f0 seconds GC run time] *** end of summary for tests-cmucl.log echo "Running test suite with gcl..."; /bin/sh ../maxima-local -lisp=gcl -batch-lisp=tests.lisp > tests-gcl.log < /de v/null 2>&1; ./summarize-log tests-gcl.log Running test suite with gcl... *** Summary results for tests recorded in *** log file tests-gcl.log: Error summary: Error(s) found in rtest15.mac: (4) Expected failures (known bugs in this version of Maxima): rtest15.mac: (4) Timing: real time : 8.690 secs run time : 7.160 secs *** end of summary for tests-gcl.log If all these tests are passed and the number of expected failures is the same as the number of known failures, everything has tested out correctly and you can proceed to the install, which is simply the standard make install. Getting advanced features like the Emacs modes to work may require a little extra work - see the documentation on those specific modes for details about making them work. 32.3 Source Based Installation on Windows Someone want to write this one up? I am most definitely not qualified. 32.4 Source Based Installation on MacOSX Someone who has done it? List of Figures 2.1 Maxima-mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2 Enhanced Terminal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.3 An Emaxima Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.4 TeXmacs with a Maxima Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 8.1 Graphing with Openmath, the default Maxima plotting tool . . . . . . . . . . . . . . . . . . . . . . . 77 8.2 Graphing with gnuplot, using the mgnuplot utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 8.3 Graphing with GeomView. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 130 Index Compiling Linux, 127 Clisp, 127 CMUCl, 127 GCL, 127 Debugging Exiting, 30 Exiting, 30 Hiding output, 31 Quitting, 30 131 Bibliography [AAC93] AACC, editor. Proceedings of the 1993 American Control Conference: The Westin St. Francis Hotel, San Francisco, California, June 2­4, 1993, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1993. IEEE Computer Society Press. Three volumes. 32.4 [ABC+88] O. Akhrif, G. L. Blankenship, P. Chancelier, C. Gomez, J. P. Quadrat, and A. Sulem. Integration of symbolic and numerical processing in control engineering design. In IEEE, editor, Digest of Papers: Compcon Spring '88. Thirty-Third IEEE Computer Society International Conference, 29 February ­ 3 March 1988, Cathedral Hill Hotel, San Francisco, California, pages 482­485, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. [ACM85] Proceedings of the ACM SIGPLAN 85 Symposium on Language Issues in Programming Environments: papers presented at the symposium in Seattle, Washington, 25­28 June, 1985, New York, NY, USA, 1985. ACM Press. Published in ACM SIGPLAN notices, volume 20, number 7. 32.4 [ACM89] ACM, editor. Proceedings of the ACM-SIGSAM 1989 International Symposium on Symbolic and Algebraic Computation, ISSAC '89, 17­19 July 1989, Portland, OR, USA, New York, NY, USA, 1989. ACM Press. 32.4 [ACM94] ACM, editor. ISSAC '94: Proceedings of the 1994 International Symposium on Symbolic and Algebraic Computation: July 20­22, 1994, Oxford, England, United Kingdom, New York, NY, USA, 1994. ACM Press. 32.4 [AG87] George C. Atallah and James Geer. Non-linear oscillations with multiple forcing terms. International Journal of Non-Linear Mechanics, 22(6):439­449, 1987. [Akm88] Varol Akman. Geometry and graphics applied to robotics. In Earnshaw [Ear88], pages 619­638. [AM89] H. Ashrafiuon and N. K. Mani. Applications of symbolic computing for numerical analysis of mechanical systems. American Society of Mechanical Engineers, Design Engineering Division (Publication) DE, 19-3(pt 3):141­149, 1989. [AM90] H. Ashrafiuon and N. K. Mani. Analysis and optimal design of spatial mechanical systems. Journal of Mechanisms, Transmissions, and Automation in Design, 112(2):200­207, June 1990. [And84] George E. Andrews. Ramanujan and SCRATCHPAD. In Golden and Hussain [GH84], pages 383­?? [ANGK+87] V. C. Aguilera-Navarro, R. Guardiola, C. Keller, M. de Llano, M. Popovic, and M. Fortes. Van der Waals perturbation theory for Fermion and Boson ground-state matter. Phys. Rev. A, 35(9):3901­3910, May 1987. [Ano75] Anonymous. MACSYMA primer. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1975. [Ano78] Anonymous. MACSYMA primer. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, revised edition, 1978. 132 BIBLIOGRAPHY 133 [Ano80] Anonymous, editor. Conference record of the 1980 LISP Conference: papers presented at Stanford University, Stanford, California, August 25­27, 1980, New York, NY, USA, 1980. ACM Press. 32.4 [Ano84] Anon. 54th symposium on shock and vibration. Shock and Vibration Bulletin, June 1984. [Ano88] Anonymous. MACSYMA user's guide. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, 1988. [Ano92] Anonymous. MACSYMA user's guide. Macsyma, Inc., Arlington, MA, USA, 1992. [Ano95a] Anonymous. Mathematik -- komfortabel und preiswert: Macsyma 2.0 für Windows. (German) [Mathematics -- comfortable and inexpensive: Macsyma 2.0 for Windows]. Elektronik, 44(5):154­??, 1995. [Ano95b] Anonymous. Software forum. Design news, 50(8):134­??, April 1995. [Ano97] Anonymous. Macsyma. Jones and Bartlett, Sudbury, MA, USA, 1997. CD-ROM and booklet. [AP90] M. D. Abouzahra and R. Pavelle. Computer algebra applied to radiation from microstrip discontinuities. Journal of Symbolic Computation, 10(5):525­528, November 1990. [Ari89] N. Ari. Application of computer code MACSYMA for electromagnetics. In Hamza [Ham89], pages 77­80. [Baj86] Chanderjit Bajaj. Proving geometric algorithm non-solvability: An application of factoring polynomials. Journal of Symbolic Computation, 2(1):99­102, March 1986. [Ban84] Kenneth A. Bannister. MACSYMA-aided large deformation analysis of a cylindrical shell under pure bending. In Golden and Hussain [GH84], pages 140­?? [Bar90] Jon Barwise. Computers and mathematics. Notices Amer. Math. Soc., 37(1):7­??, January 1990. [Bau88] H. H. Bau. Symbolic computation -- an introduction for the uninitiated. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:1­10, 1988. [BBR85] N. Bessis, G. Bessis, and D. Roux. Closed-form expressions for the Dirac-Coulomb radial rt integrals. Phys. Rev. A, 32:2044­2050, 1985. [Ben98] Hans Benker. Ingenieurmathematik mit Computeralgebra-Systemen: AXIOM, DERIVE, MACSYMA, MAPLE, MATHCAD, MATHEMATICA, MATHLAP und MuPAD in der Anwendung. (German) [Engineering mathematics with computer algebra systems: AXIOM, DERIVE, MACSYMA, MAPLE, MATHCAD, MATHEMATICA, MATHLAP and MuPAD in their use]. Friedrich Vieweg und Sohn, Braunschweig, Germany, 1998. [Ber84] Robert H. Berman. Measuring the performance of a computational physics environment. In Golden and Hussain [GH84], pages 244­?? [Bey79] W. A. Beyer. Lie group theory for symbolic integration of first order ordinary differential equations. In Lewis [Lew79], pages 362­384. [Bey84] William A. Beyer. Solution of simultaneous polynomial equations by elimination in MACSYMA. In Golden and Hussain [GH84], pages 110­?? [BFBZ92] Eva Bozoki, Aharon Friedman, and Ilan Ben-Zvi. ASAP -- A symbolic algebra package for accelerator design. In IEEE [IEE92a], pages 272­274. Five volumes. IEEE catalog no. 91CH3038-7. [BFHS92] W. A. Beyer, L. R. Fawcett, L. P. Harten, and B. K. Swartz. The volume common to two congruent circular cylinders. Journal of Symbolic Computation, 13(2):221­230, February 1992. [BFMS87] W. A. Beyer, L. R. Fawcett, R. D. Mauldin, and B. K. Swartz. The volume common to two congruent circular cones whose axes intersect symmetrically. Journal of Symbolic Computation, 4(3):381­390, December 1987. BIBLIOGRAPHY 134 [BG83] Richard Bogen and Jeffrey Golden. MACSYMA reference manual. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, version 10, first printing edition, 1983. [BGGD77] Richard Bogen, Jeffrey Golden, Michael Genesereth, and Alexander Doohovskoy. MACSYMA reference manual: version nine. Massachusetts Institute of Technology, Cambridge, MA, USA, 1977. [BGV94] W. N. Borst, V. V. Goldman, and J. A. Van Hulzen. GENTRAN 90: a REDUCE package for the generation of Fortran 90 code. In ACM [ACM94], pages 45­51. [BH87] W. A. Beyer and L. Heller. A Steiner tree associated with tree quarks. Journal of Symbolic Computation, 3(3):283­289, June 1987. [BH95] Stefan Braun and Harald Häuser. Macsyma Version 2: systemnahe und praxisnahe Einführung mit Anwendungsbeispielen. (German) [Macsyma Version 2: Hands-on practical use: Introduction with sample applications]. Addison-Wesley, Bonn, Germany, 1995. [BHY88] Haim H. Bau, Thorwald Herbert, and M. M. Yovanovich, editors. Symbolic Computation in Fluid Mechanics and Heat Transfer, volume 105 of American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 345 E. 47th St., New York, NY 10017, USA, 1988. American Society of Mechanical Engineers. [BIG01] A. Ben-Israel and R. Gilbert. Computer-supported calculus. Texts and monographs in symbolic computation. Springer Verlag, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 2001. [BKK76] A. (Abraham) Bers, John Laurence Kulp, and C. F. F. (Charles F. F.) Karney. Symbolic computation of nonlinear wave interactions on MACSYMA. Plasma research report PRR 76/17, Massachusetts Institute of Technology, Research Laboratory of Electronics, Cambridge, MA, USA, 1976. [BMM90] John S. Baras, David C. MacEnany, and Robert L. Munach. Fast error-free algorithms for polynomial matrix computations. Proceedings of the IEEE Conference on Decision and Control, 2:941­946, 1990. IEEE catalog number 90CH2917-3. [BMS88] Christopher I. Byrnes, Clyde F. Martin, and Richard E. Saeks, editors. Analysis and Control of Nonlinear Systems, Amsterdam, The Netherlands, 1988. North-Holland. Selected papers from the 8th International Symposium on the Mathematics of Networks and Systems, held in Phoenix, June 15­19, 1987. 32.4 [Bog83] Richard Bogen. MACSYMA Reference Manual, Version 10. Massachusetts Institute of Technology, Computer Science Lab., Cambridge, MA, USA, December 1983. 2nd Printing, Vol. I. [Bog86] Richard Bogen. MACSYMA reference manual. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, version 12 edition, 1986. [BR73] A. Bhushan and N. Ryan. RFC 578: Using MIT-Mathlab MACSYMA from MIT-DMS Muddle, October 1973. [Bra89] Kevin F. Bratcher. Utilization of the MACSYMA software for instructional programming. Thesis (m. eng.), Department of Mechanical Engineering, University of Louisville, Louisville, KY, USA, 1989. [Bre84] Richard L. Brenner. Simplifying large algebraic expressions by computer. In Golden and Hussain [GH84], pages 50­?? [BS84] Johnnie W. Baker and Oberta Slotterberg. Providing a complex number environment for MACSYMA and VAXIMA. In Golden and Hussain [GH84], pages 39­?? [BSZ93] J. Birk, J. Schaffner, and M. Zeitz. Rule-based selection of nonlinear observer design methods. In Fliess [Fli93], pages 231­238. [BT88] Berta Buttarazzi and Antonio Tornambe. Expert system for the asymptotic estimators. In IEEE [IEE88c], pages 1153­1156. Two volumes. BIBLIOGRAPHY 135 [Buc85] Bruno Buchberger, editor. Proceedings of Eurocal '85, Vol. I, volume 203 of Lecture Notes in Computer Science, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1985. Springer Verlag. 32.4 [Cah90] Kevin Cahill. How to use MACSYMA to write long FORTRAN codes for noncompact simulations of gauge theories. Computers in Physics, 4(2):159­165, March­April 1990. [Cal82] Jacques Calmet, editor. Computer algebra: EUROCAM'82, European Computer Algebra Conference, Marseille, France, 5­7 April, 1982, volume 144 of Lecture Notes in Computer Science, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1982. Springer Verlag. [Car84] George J. Carrette. Results in unexpected MACSYMA implementation environments. In Golden and Hussain [GH84], pages 292­?? [CCF84] J. N. L. Connor, P. R. Curtis, and D. Farrelly. The uniform asymptotic swallowtail approximation: Practical methods for oscillating integrals with four coalescing saddle points. J. Phys. A, 17:283­310, 1984. [CD82] Daniel Claude and Pierre Dufresne. Application of Macsyma to nonlinear systems decoupling. Lecture Notes in Computer Science, 144:294­301, 1982. [CD87] Harvey Cohn and Jesse Ira Deutsch. Application of symbolic manipulation to the Hecke transformations of modular forms in two variables, II. Journal of Symbolic Computation, 4(1):35­40, August 1987. [CDA90] Henri Cabannes and Jean-Paul Duruisseau-Alyod. Construction, using Macsyma, of exact solutions for some equations of discrete kinetic theory of gases. American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP, 205:277­284, 1990. [CDW90] Shui-Nee Chow, Byron Drachman, and Duo Wang. Computation of normal forms. Journal of Computational and Applied Mathematics, 29(2):129­143, February 1990. [Cel84] Pedro Celis. Remark: Corrections and errors in John Ivie's some MACSYMA programs for solving recurrence relations. ACM Transactions on Mathematical Software, 10(4):477­478, December 1984. See [Ivi78]. 32.4 [CF80] J. A. Campbell and J. P. Fitch. Symbolic computing with and without LISP. In Anonymous [Ano80], pages viii + 247. [CFG+84] B. Char, G. Fee, K. O. Geddes, G. H. Gonnet, M. B. Monagen, and S. M. Watt. On the design and performance of the Maple system. In Golden and Hussain [GH84], pages 199­?? [CG90] G. Chen and I. Gil. Implementation of an algorithm in Macsyma. computing the formal solutions of differential systems in the neighborhood of regular singular point. In Watanabe and Nagata [WN90], pages 307­?? [Cha86] Bruce W. Char, editor. Proceedings of the 1986 Symposium on Symbolic and Algebraic Computation: Symsac '86, July 21­23, 1986, Waterloo, Ontario, New York, NY, USA, 1986. ACM Press. 32.4 [Cha92] T. F. Chan. Analysis of self-excited induction generators using symbolic programming. International Journal of Electrical Engineering Education, 29(4):329­338, October 1992. [Che88] K. J. Cheng. Perturbation solution to heat transfer in wedge-type flow using the computer algebra system MACSYMA. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:47­55, 1988. [CHH91] F. A. Chedid-Helou and J. H. Hemann. Mathematical modeling of wave propagation in anisotropic media. Materials Evaluation, 49(6):708­715, June 1991. BIBLIOGRAPHY 136 [CHW91] B. Champagne, W. Hereman, and P. Winternitz. The computer calculation of Lie point symmetries of large systems of differential equations. Computer Physics Communications, 66(2-3):319­340, September­October 1991. [Cla89a] M. Clarkson. MACSYMS's inverse Laplace transform. SIGSAM Bulletin, 23(1):33­38, January 1989. [Cla89b] M. E. Clarkson. An improved Laplace transform package for MACSYMA. SIGSAM Bulletin, 19(2):31­33, May 1989. [Cla90] M. E. Clarkson. Praxis: An expert system for Macsyma. In Miola [Mio90], pages 264­265. [Coo84] Gene Cooperman. An alternative top-level for MACSYMA. In Golden and Hussain [GH84], pages 356­?? [Cre89] M. R. M. Crespo da Silva. Analysis of the non-planar response of a cantilever, with the aid of computerized symbolic manipulation. In Nelson, Jr. [Nel89], pages 61­70. [Cre90a] M. R. M. Crespo da Silva. Formulation and analysis of the nonlinear dynamics of a beam with the aid of symbolic computation. American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP, 205:149­174, 1990. [Cre90b] M. R. M. Crespo da Silva. On the use of symbolic computation for automating the analysis of problems in dynamics. In Kinzel et al. [KRB+90], pages 593­600. Two volumes. [CS87] Jean-Philippe Chancelier and Agnes Sulem. MacroTEX: a LATEX code generator in Macsyma. Rapports techniques 93, INRIA (Institut National de Recherche en Informatique et en Automatique), Le Chesnay, France, 1987. [CS89] J. Ph. Chancelier and A. Sulem. MACROTEX : un générateur de code LATEX implémenté en MACSYMA. (French) [MACROTEX: a code generator implemented in MACSYMA]. Cahiers GUTenberg, 3:32­39, octobre 1989. [CS90] L. S. Chien and C. T. Sun. Parallel computation using boundary elements in solid mechanics. Collection of Technical Papers -- AIAA/ASME/ASCE/AHS Structures, Structural Dynamics & Materials Conference, Pt 2:644­651, 1990. [CTZY90] T. Y. Chang, H. Q. Tan, D. Zheng, and M. W. Yuan. Application of symbolic method to hybrid/mixed finite elements and computer implementation. Computers and Structures, 35(4):293­299, 1990. [Cut84] Elizabeth Cuthill. Evaluating infinite integrals using MACSYMA. In Golden and Hussain [GH84], pages 291­?? [CVG94] John Crow, Mario Vassallo, and Gustaf Gripenberg. Three software reviews comprise this month's column. John Crow presents the second installment of his two-part comparative review of Maple and Macsyma. Mario Vassallo reports on his experiences with spreadsheets and mathematics. Gustaf Gripenberg takes a look at MAX. Notices Amer. Math. Soc., 41(4):299­306, April 1994. [DCC85] Sandra J. DeLoatch, Langley Research Center, and Norfork State College. MACSYMA usage at Langley. Central Scientific Computing Complex document Z-1; NASA contractor report NASA CR 172518, Norfolk State College, Norfolk, VA, USA, 1985. [Dek83] Joseph Deken. Symbolic computing and statistics. In Heiner et al. [HSW83], pages 70­73. [DeL87] Sandra J. DeLoatch. MACSYMA usage at Langley. NASA contractor report NASA-CR 172518, NASA, Washington, DC, USA, 1987. [Del91] M. Delest. Enumeration of polyominoes using Macsyma. Theoret. Comput. Sci., 79(1):209­226, February 1991. [Deu93] Jesse Ira Deutsch. Identities arising from Hecke transformations of modular forms over q( 2) and q( 3). Journal of Symbolic Computation, 15(3):315­324 (or 315­323??), March 1993. BIBLIOGRAPHY 137 [Dev94a] Keith Devlin. Computers and mathematics. Notices Amer. Math. Soc., 41(3):195­??, March 1994. [Dev94b] Keith Devlin. Computers and mathematics. Notices Amer. Math. Soc., 41(4):299­??, April 1994. [Dev94c] Keith Devlin. Computers and mathematics. Notices Amer. Math. Soc., 41(7):772­??, September 1994. [DJ89] M. L. Dudley and R. W. James. Computer-aided derivation of spherical harmonic spectral equations in astro-geophysics. Journal of Symbolic Computation, 8(4):423­427, October 1989. [DM88] A. K. Dhingra and N. K. Mani. Finite and multiply separated kinematic synthesis of link and geared mechanisms. American Society of Mechanical Engineers, Design Engineering Division (Publication) DE, 15-1(PT1):317­326, 1988. [DM93] A. K. Dhingra and N. K. Mani. Finitely and multiply separated synthesis of link and geared mechanisms using symbolic computing. Journal of Mechanical Design, Transactions Of the ASME, 115(3):560­567, September 1993. [Dri84] R. Drew Drinkard, Jr. A tutorial on particular uses of MACSYMA. In Golden and Hussain [GH84], pages 186­?? [DS81] R. Drew Drinkard and Nancy K. Sulinski. MACSYMA: a program for computer algebraic manipulation (Demonstrations and Analysis). Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, 1981. [DST87] James H. Davenport, Yvon Siret, and Évelyne Tournier. Calcul formel: systmes et algorithmes de manipulations algébriques. (French) [Formal calculus: systems and algorithms for algebraic manipulation]. Études et recherches en informatique. Masson, Paris, France, 1987. 32.4 [DST88] James H. Davenport, Yvon Siret, and Évelyne Tournier. Computer Algebra: Systems and Algorithms for Algebraic Computation. Academic Press, New York, USA, 1988. Translated from the French [DST87] by A. Davenport and J. H. Davenport. [Ear88] Rae A. Earnshaw, editor. Theoretical foundations of computer graphics and CAD: Proceedings of the NATO Advanced Study Institute held at Il Ciocco, Italy, July 4-17, 1987, volume 40 of NATO ASI series. Series F, Computer and systems sciences, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1988. Springer Verlag. 32.4 [EF89] T. H. Einwohner and R. J. Fateman. A MACSYMA package for the generation and manipulation of Chebyshev series. In ACM [ACM89], pages 180­185. [EM87] C. Easwaran and S. R. Majumdar. Application of Macsyma in solving the laminar flow of micropolar fluid in a meandering channel. Canadian Journal of Chemical Engineering, 65(4):529­535, August 1987. [Eng84] Paul D. Engelman. Two to three dimensional mapping. In Golden and Hussain [GH84], pages 313­?? [ET88a] H. K. Eldeib and S. Tsai. Symbolic computing in computer-aided control system analysis and design. In IEEE, editor, Proceedings of the IEEE 1988 National Aerospace and Electronics Conference, NAECON 1988, 7­10 November 1988, Convention Center, Santa Clara, California, pages 488­491, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. IEEE catalog number 88CH2657-5. [ET88b] H. K. Eldeib and S. Tsai. Symbolic computing in computer-aided control system analysis and design. In IEEE, editor, Proceedings of the IEEE 1988 National Aerospace and Electronics Conference, 1988. NAECON 1988, 23­27 May 1988, Dayton Convention Center, Dayton, OH, USA, volume 2, pages 428­433, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. BIBLIOGRAPHY 138 [ET89] H. K. Eldeib and S. Tsai. Applications of symbolic manipulation in control system analysis and design. In IEEE, editor, IEEE International Symposium on Intelligent Control, 1988. Proceedings, 24­26 August 1988, Arlington, Virginia, pages 269­274, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1989. IEEE Computer Society Press. [F+77] R. J. Fateman et al., editors. Proceedings of the 1977 MACSYMA Users' Conference, held at Berkeley, California, July 27­29, 1977, number CP-2012 in NASA conference publication, Washington, DC, USA, July 1977. NASA. 32.4 [Fab92] James A. Fabunmi. System parameters of output feedback controlled flexible structures. Journal of Intelligent Material Systems and Structures, 3(2):316­332, April 1992. [Fab93] James A. Fabunmi. Feedback gain sensitivities of closed-loop modal parameters of controlled structures. Journal of Guidance, Control, and Dynamics, 16(5):892­898, September­October 1993. [Far89] S. Faraji. Higher order theories for thick layers of transversely isotropic material with loading symmetric about the middle plane. In Topping [Top89], pages 287­292. [Fat82a] Richard J. Fateman. Addendum to the MACSYMA reference manual for the VAX. Technical report, Computer Science Division, University of California, Berkeley, 1982. [Fat82b] Richard J. Fateman. MACSYMA primer for VAX/UNIX. Computing Services, University of California, Berkeley, Berkeley, CA, USA, 1982. [Fat87] Richard J. Fateman. TEX output from MACSYMA-like systems. SIGSAM Bulletin, 21(4):1­5, November 1987. [Fat89] Richard J. Fateman. A review of Macsyma. IEEE Transactions on Knowledge and Data Engineering, 1(1):133­145, March 1989. [Fav79] John M. Favaro. An interactive symbolic executor based on MACSYMA. Master of science, plan ii., Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1979. [Fel98] Richard N. Fell. Macsyma tutorial for calculus. Jones and Bartlett, Sudbury, MA, USA, 1998. [FF81] J. K. Foderaro and R. J. Fateman. Characterization of VAX Macsyma. In Wang [Wan81], pages 14­19. ACM order no. 505810. [FFF+84] R. J. Fateman, J. Foderaro, G. Foster, R. McGeer, N. Soiffer, and C. J. Williamson. Research in algebraic manipulation at the University of California, Berkeley. In Golden and Hussain [GH84], pages 188­?? [FG80] Richard J. Fateman and Mathlab Group. Addendum to the Mathlab/MIT MACSYMA reference manual for VAX/UNIX "VAXIMA". Computing Services, University of California, Berkeley, Berkeley, CA, USA, 1980. [fHR93] Shi fang Han and K. G. Roesner. Unsteady flow of polymer fluid between coaxial cylinders. Journal of Hydrodynamics, 5(2):52­62, February 1993. [Fit73] John Fitch. Problems #3 and #4 in REDUCE and MACSYMA. SIGSAM Bulletin, pages 10­11, 1973. [FKM95] Robert Fournier, Norbert Kajler, and Bernard Mourrain. Visualization of mathematical surfaces: the IZIC server approach. Journal of Symbolic Computation, 19(1/2/3):159­174 (or 159­173??), January, February, March 1995. Design and implementation of symbolic computation systems (Gmunden, 1993). [Fli93] M. Fliess, editor. Nonliner control systems design 1992: selected papers from the IFAC symposium, Bordeaux, France, 24­26 June 1992, volume 7 of IFAC symposia series, Oxford, UK, 1993. Pergamon Press. 32.4 BIBLIOGRAPHY 139 [Fod78] John Keith Foderaro. Typesetting MACSYMA equations. Master of science, plan ii., Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1978. [Fra84] Joseph P. Frakes. Numeric-analytic solutions with the aid of MACSYMA of the Von Karman equations for a circular plate under a concentrated load. Thesis (m.s.a. math.), University of Virginia, Charlottesville, VA, USA, 1984. [Fre81] Daniel Freedman. An integrated plotting package for VAX/ Unix Macsyma. Master of science, plan ii., Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1981. [GA+86] G. Gupta, K. S. Ahluwalia, et al., editors. Computers in engineering, 1986: proceedings of the 1986 ASME International Computers in Engineering Conference and Exhibition, July 20­24, 1986, Chicago, Illinois, 345 E. 47th St., New York, NY 10017, USA, 1986. American Society of Mechanical Engineers. Three volumes. 32.4 [Gat84] Barbara Louise Gates. The design and implementation of an automatic RATFOR code generator for VAXIMA. Thesis (m.s.), Kent State University, Kent, OH, USA, 1984. [GBC92] L. Gerbaud, J. Bigeon, and G. Champenois. Modular approach to describe electromechanical systems. using Macsyma to generate global approach simulation software. PESC record, II(??):1189­1196, 1992. [GBoT74] Mathlab Group, Richard Bogen, and Project Mac (Massachusetts Institute of Technology). MACSYMA reference manual. Massachusetts Institute of Technology, Project MAC, Cambridge, MA, USA, version 6 edition, 1974. [GD93] Ronald E. Graham and Valerie J. DuFore. Arrow mode reboost analysis for space station freedom. In AACC [AAC93], pages 2485­2488. IEEE catalog number 93CH3225-0. [GE94] G. C. Goodwin and R. J. Evans, editors. Automatic Control. World Congress 1993. Proceedings of the 12th Triennial World Congress of the International Federation of Automatic Control. Vol.2. Robust Control, Design and Software, Sydney, Australia, 18­23 July 1993, Oxford, UK, 1994. Pergamon Press. 32.4 [Gen79] M. R. Genesereth. The role of plans in automated consultation. In IJCAI79 [IJC79], pages 311­319. User consultant for MACSYMA. [Gen83] James E. Gentle, editor. Computer science and statistics: proceedings of the Fifteenth Symposium on the Interface, Houston, Texas, March 1983, Amsterdam, The Netherlands, 1983. North-Holland. 32.4 [GF80] Mathlab Group and Richard J. Fateman. An introduction to MACSYMA for VAX/UNIX. Computing Services, University of California, Berkeley, Berkeley, CA, USA, 1980. [GFB+93] L. Gerbaud, F. Pazos Flores, A. Bolopion, Y. Baudon, and J.-P. Ferrieux. On the use of MACSYMA symbolic language to generate simulation softwares for the control of converter-machine drives. In IEE [IEE93], pages 396­401. [GH84] V. Ellen Golden and M. A. Hussain, editors. Proceedings of the 1984 MACSYMA Users' Conference: Schenectady, New York, July 23­25, 1984, Schenectady, NY, USA, 1984. General Electric. 32.4 [Gil95] David E. Gilsinn. Constructing Galerkin's approximations of invariant tori using MACSYMA. Nonlinear Dynamics, 8(2):269­305, September 1995. [GJK88] P. W. Grant, C. P. Jobling, and Y. W. Ko. A comparison of rule-based and algebraic approaches to the computer-aided manipulation of linear dynamic system models. In International Conference on Control 88, 13­15 April 1988, number 285 in IEE conference publication, pages 87­92, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. BIBLIOGRAPHY 140 [GM82] V. Ellen Golden and Mathlab Group. Introductory MACSYMA documentation: a collection of papers: (a) An introduction to ITS for the Macsyma user, (b) ITS easy once ITS explained, and (c) Macsyma primer. Massachusetts Institute of Technology, Laboratory for Computer Science, Cambridge, MA, USA, revised edition, 1982. [Gol77a] J. P. Golden. MACSYMA's differential equation solver. In Fateman et al. [F+77], pages xi + 501. [Gol77b] V. Ellen Golden. An introduction to ITS for the MACSYMA user. Mathlab memo 3, Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1977. [Gol79] V. Ellen Golden. An introduction to ITS for the MACSYMA user. Mathlab memo 3, Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1979. [Gol82] V. Ellen Golden. An Introduction to ITS for the MACSYMA user. Massachusetts Institute of Technology, Laboratory for Computer Science, Cambridge, MA, USA, 1982. [Gol84] V. Ellen Golden. Computational geography -- the habitats of the migratory Macsyma. In Golden and Hussain [GH84], pages 362­?? [Gol85] Jeffery P. Golden. Differentiation of unknown functions in MACSYMA. SIGSAM Bulletin, 19(2):19­ 24, May 1985. [Gol86] J. P. Golden. An operator algebra for Macsyma. In Char [Cha86], pages 244­246. [Gon83] Gail Gong. Letting Macsyma help. In Gentle [Gen83], pages 237­244. [GoTLfCS83] Mathlab Group, Massachusetts Institute of Technology. Laboratory for Computer, and Science. MACSYMA reference manual. Massachusetts Institute of Technology, Cambridge, MA, USA, version ten edition, 1983. [Gra94] L. P. Grayson, editor. Proceedings. Frontiers in Education. Twenty-Fourth Annual Conference. Educating Engineers for World Competition (Cat. No.94CH35723), 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1994. IEEE Computer Society Press. 32.4 [Grä96] Hans-Gert Gräbe. About the polynomial system solve facility of Axiom, Macsyma, Maple, Mathematica, MuPAD, and Reduce. Technical report 96-11, Institut für Informatik, Universität Leipzig, Leipzig, Germany, 1996. [Gro78] Mathlab Group. MACSYMA primer. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1978. [Gui89] M. Guizani. A review of Macsyma. IEEE Transactions on Knowledge and Data Engineering, 1(1):133, March 1989. [GW84] B. L. Gates and Paul S. Wang. A LISP-based RATFOR code generator. In Golden and Hussain [GH84], pages 319­?? [H+91] T. C. Huang et al., editors. Machinery dynamics and element vibrations: presented at the 1991 ASME design technical conferences­13th Biennial Conference on Mechanical Vibration and Noise, September 22­25, 1991, Miami, Florida, volume 36 of Design engineering, 345 E. 47th St., New York, NY 10017, USA, 1991. American Society of Mechanical Engineers. 32.4 [Ham89] M. H. Hamza, editor. Proceedings of the IASTED International Symposium. Expert Systems Theory and Applications, Anaheim, CA, USA, 1989. Acta Press. 32.4 [Har84] Leo Harten. Applications of MACSYMA in solving lineaar systems of differential equations. In Golden and Hussain [GH84], pages 122­?? [HB87] W. Hereman and E. Van Den Bulck. MACSYMA program for the Painlevé test of nonlinear ordinary and partial differential equations. Report AD-A194 293; CMS-TSR-88-21, University of Wisconsin at Madison. Center for Mathematical Sciences, Madison, WI, USA, 1987. BIBLIOGRAPHY 141 [HC85] Leo P. Harten and George J. Carrette. Share Library in DOE-MACSYMA. ISA, Research Triangle Park, NC, USA, 1985. [HCR91] Klaus A. Hoffmann, Ting-Lung Chiang, and Walter H. Rutledge. Computation of flowfields for projectiles in hypersonic chemically reacting flows. Journal of Spacecraft and Rockets, 28(1):23­30, January­February 1991. [Hei87] R. M. Heiberger, editor. Computer Science and Statistics: Proceedings of the 19th Symposium on the Interface, Alexandria, VA, USA, 1987. American Stat. Assoc. 32.4 [Hel91] Barbara Heller. MACSYMA for statisticians. Wiley series in probability and mathematical statistics. Applied probability and statistics. John Wiley and Sons, New York, NY, USA; London, UK; Sydney, Australia, 1991. [Her83] R. Hermann. Geometric construction and properties of some families of solutions of nonlinear partial differential equations. J. Math. Phys., 24(3):510­521, 1983. [Her88] T. Herbert. Symbolic computations with spectral methods. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:25­31, 1988. [Her91] Willy Hereman. Exact solitary wave solutions of coupled nonlinear evolution equations using MACSYMA. Computer Physics Communications, 65(1-3):143­150, April 1991. [HFO94] Toshiaki Hata, Masashi Furumaki, and Kazuhumi Ohenoki. Transient analysis of stress wave penetration in a plate subjected to a stress pulse (application of symbolic manipulation system to ray theory). Nippon Kikai Gakkai Ronbunshu, A Hen/Transactions of the Japan Society of Mechanical Engineers, Part A, 60(576):1800­1806, August 1994. [HJL89] P. Hansen, B. Jaumard, and S. H. Lu. Automated procedure for globally optimal design. Journal of Mechanisms, Transmissions, and Automation in Design, 111(3):361­367, September 1989. [HJL91] Pierre Hansen, Brigitte Jaumard, and Shi-Hui Lu. Analytical approach to global optimization. Mathematical Programming, 52(2):227­254, August 1991. [HLTH94] D. W. C. Ho, J. Lam, S. K. Tin, and C. Y. Han. Recent applications of symbolic computation in control system design. In Goodwin and Evans [GE94], pages 567­570. [HN83] M. A. Hussain and Ben Noble. Applications of MACSYMA to calculations in applied mathematics. Corporate Research and Development Report 83CRD054 (Technical information series), General Electric, Schenectady, NY, USA, 1983. [HN85] M. A. Hussain and B. Noble. Equivalent formulation of equations of motion for complex dynamical systems using computer algebra. Computers in Engineering, Proceedings of the International Computers in Engineering Conference and, 1:483­489, 1985. [HNS87] W. Hereman, Y. Nagel, and J. Strikwerda. MACSYMA at CMS. Version 309.3. Report AD-A193 476; CMS-TSR-88-3, University of Wisconsin at Madison. Center for Mathematical Sciences, Madison, WI, USA, 1987. [Hol86] Ulf Holmberg. Some MACSYMA functions for analysis of multivariable linear systems. Technical Report LUFTD2/(TFRT-7333)/1040/(1986), Department of Automatic Control, Lund Institute of Technology, Lund, Sweden, 1986. [Hol88] M. H. Holmes. Application of symbolic manipulation in the study of soft tissues. American Society of Mechanical Engineers, Bioengineering Division (Publication) BED, 9:111­122, 1988. [HSW83] Karl W. Heiner, Richard S. Sacher, and John W. Wilkinson, editors. Computer Science and Statistics: proceedings of the 14th Symposium on the Interface. Springer Verlag, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1983. 32.4 BIBLIOGRAPHY 142 [HSW94] J. W. Helton, M. Stankus, and J. Wavrik. Computer simplification of engineering systems formulas. In IEEE, editor, Proceedings of the 33rd IEEE Conference on Decision and Control, 14­16 December 1994, volume 2, pages 1893­1898, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1994. IEEE Computer Society Press. IEEE catalog number 94CH3460-3. [HSW98] J. W. Helton, M. Stankus, and J. J. Wavrik. Computer simplification of formulas in linear systems theory. IEEE Transactions on Automatic Control, 43(3):302­314, March 1998. [HT84] P. Hollis and D. L. Taylor. Hopf bifurcation in multi-degree-of-freedom systems using MACSYMA. In Golden and Hussain [GH84], pages 169­?? [HT90a] W. Hereman and M. Takaoka. Solitary wave solutions of nonlinear evolution and wave equations using a direct method and MACSYMA. J. Phys. A, 23(21):4805­??, November 1990. [HT90b] P. Hollis and D. L. Taylor. Applications of computer algebra in journal bearing analysis. In Kinzel et al. [KRB+90], pages 601­606. Two volumes. [HvH+83] B. Hulshof, A. van Hulzen, et al. REDUCE, 1983. [HWH91] David Harper, Chris Wooff, and David Hodgkinson. A Guide to Computer Algebra Systems. John Wiley and Sons, New York, NY, USA; London, UK; Sydney, Australia, 1991. [IEE86] IEEE, editor. Proceedings / 1986 IEEE International Conference on Robotics and Automation, April 7­10, 1986, the San Francisco Hilton and Tower, San Francisco, California, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1986. IEEE Computer Society Press. Three volumes. IEEE Computer Society order number 695. 32.4 [IEE88a] IEEE, editor. Proceedings / 1988 IEEE International Conference on Robotics and Automation, April 24­29, 1988, Franklin Plaza Hotel, Philadelphia, Pennsylvania, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. Three volumes. 32.4 [IEE88b] IEEE, editor. Proceedings / the Fourth Conference on Artificial Intelligence Applications. Sheraton Harbor Island Hotel, San Diego, California, March 14­18, 1988, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. IEEE catalog number 88CH2552-8. 32.4 [IEE88c] IEEE, editor. Proceedings of the 1988 IEEE International Conference on Systems, Man and Cybernetics, August 8­12, 1988, Beijing and Shenyang, China. Beijing, China, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. Two volumes. 32.4 [IEE91] IEEE, editor. 1991 IEEE International Joint Conference on Neural Networks: the Westin Stamford and Westin Plaza, 18­21 November 1991, Singapore, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1991. IEEE Computer Society Press. Three volumes. IEEE catalog number 91CH3065- 0. 32.4 [IEE92a] IEEE, editor. Conference record of the 1991 IEEE Particle Accelerator Conference: accelerator science and technology, May 6­9, 1991, San Francisco, California, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1992. IEEE Computer Society Press. Five volumes. IEEE catalog no. 91CH3038-7. 32.4 [IEE92b] IEEE, editor. Proceedings, Fourth International Conference on Software Engineering and Knowledge Engineering: SEKE 92, June 15­20, 1992, Europa Palace Hotel, Capri, Italy, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1992. IEEE Computer Society Press. IEEE Catalog number 92THO438-2. 32.4 [IEE93] IEE, editor. Fifth European Conference on Power Electronics and Applications: 13­16 September 1993: venue, Brighton Conference Centre, UK, volume Eight volumes., London, UK, 1993. IEE. 32.4 BIBLIOGRAPHY 143 [IJC79] IJCAI-79: Proceedings of the Sixth International Joint Conference on Artificial Intelligence (Tokyo, August 20­23, 1979), Tokyo, Japan, 1979. Joho Shori Gakkia. Two volumes. 32.4 [IL88] R. A. Ibrahim and W. Li. Structural modal interaction with combination internal resonance under wide-band random excitation. Journal of Sound and Vibration, 123(3):473­495, June 1988. [ILT87] D. Ionescu, P. Lethebinh, and I. Trif. Expert system for computer process control design. In Meystel and Luh [ML87], pages 185­193. IEEE Service Cent. Piscataway, NJ, USA. [Ivi77] John Ivie. Some MACSYMA programs for solving recurrence relations: research project. Master of science, plan ii, Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1977. [Ivi78] John Ivie. Some Macsyma programs for solving recurrence relations. ACM Transactions on Mathematical Software, 4(1):24­33, March 1978. See also [Cel84]. 32.4 [Jen74] R. D. Jenks, editor. Proceedings of Eurosam '74, Royal Institute of Technology, Stockholm, Sweden, August 1­2, 1974, New York, NY, USA, 1974. ACM Press. Published in ACM SIGSAM Bulletin, volume 8, number 3. [Jen84] Richard D. Jenks. The new SCRATCHPAD language and system for computer algebra. In Golden and Hussain [GH84], pages 409­?? [JL94] M. Jerosolimski and L. Levacher. New method for fast calculation of Jacobian matrices: Automatic differentiation for power system simulation. IEEE Transactions on Power Systems, 9(2):700­706, May 1994. [JM85] M. S. Ju and J. M. Mansour. Comparative studies of formulating the dynamics of rigid-body systems using Macsyma -- a case study. Developments in Mechanics, 13:185­186, 1985. [JM87] Ming-Shaung Ju and Joseph M. Mansour. Application of simulation sensitivity analysis to functional neuromuscular stimulation of gait. AMD (Symposia Series) (American Society of Mechanical Engineers, Applied Mechanics Division), 84:327­330, 1987. [JM88] Ming-Shaung Ju and J. M. Mansour. Simulation of the double limb support phase of human gait. Journal of Biomechanical Engineering, Transactions of the ASME, 110(3):223­229, August 1988. [JS87] J. Jacky and D. Schuler, editors. Directions and Implications of Advanced Computing, Palo Alto, CA, USA, 1987. Comput. Professionals Social Responsibility. 32.4 [Kea91] G. Keady. GENTRANs from REDUCE and from MACSYMA. Technical report, University of Waikato, Mathematics Department, Waikato, New Zealand, September 1991. [KLW90] Bernhard Kutzler, Franz Lichtenberger, and Franz Winkler. Softwaresysteme zur Formelmanipulation: praktisches Arbeiten mit den Computer-Algebra-Systemen REDUCE, MACSYMA, und DERIVE. (German) [Software systems for formula manipulation: practical work with the computer algebra systems REDUCE< MACSYMA, and DERIVE]. Expert Verlag, Ehningen Bei Boeblingen, Germany, 1990. [Koe92] Wolfram Koepf. Power series in computer algebra. Journal of Symbolic Computation, 13(6):581­604 (or 581­603??), June 1992. [Koe93] Wolfram Koepf. Examples for the algorithmic calculation of formal Puisieux, Laurent and power series. SIGSAM Bulletin, 27(1):20­32, January 1993. [KoTRLoE77] John Laurence Kulp, Massachusetts Institute of Technology. Research Laboratory of, and Electronics. Ray trajectories in a torus -- an application of MACSYMA to complex numerical computation. Plasma research report PRR 77/9, Massachusetts Institute of Technology, Research Laboratory of Electronics, Cambridge, MA, USA, 1977. BIBLIOGRAPHY 144 [Kow86] Janusz S. Kowalik. Knowledge based problem solving. Prentice-Hall, Englewood Cliffs, NJ 07632, USA, 1986. 32.4 [KRB+90] G. L. Kinzel, S. M. Rohde, D. W. Bennett, et al., editors. Computers in engineering, 1990: proceedings of the 1990 ASME International Computers in Engineering Conference and Exposition, August 5­9, Boston, Massachusetts, 345 E. 47th St., New York, NY 10017, USA, 1990. American Society of Mechanical Engineers. Two volumes. 32.4 [KSB92] A. K. Kaw, A. S. Sclvarathinaru, and G. H. Besterfield. Comparison of interphase models for a crack in fiber reinforced composite. Theoretical and Applied Fracture Mechanics, 17(2):133­147, July 1992. [Kut88] B. Kutzler. muMATH-algebra facilities on microcomputers. Mini Micro Magazin, 4(9):40­43, September 1988. [Kwo91] Yue-Kuen Kwok. Application of MACSYMA to solutions of ordinary differential equations. International journal of mathematical education in science and technology, 22(6):877­??, November 1991. [KWW92] B. (Bernhard) Kutzler, Bernhard Wall, and Franz Winkler. Mathematische Expertensysteme: Praktisches Arbeiten mit den Computer Algebra-System MACSYMA, Mathematica und DERIVE. (German) [Mathematical Expert Systems: Practical Work with the Computer Algebra Systems MACSYMA, Mathematica, and DERIVE]. Number 430 in Kontakt und Studium. Expert Verlag, Ehningen Bei Boeblingen, Germany, 1992. [Lan80] Douglas H. Lanam. A package for generating and executing Fortran programs with Macsyma. Master of science, plan ii., Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1980. [Lan87] S. Landau. The responsible use of 'experť systems. In Jacky and Schuler [JS87], pages 167­181. [Lan88] R. H. Lance. Symbolic computation and the instruction of engineering undergraduates. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:21­24, 1988. [Lan89] Richard H. Lance. Computer algebra and undergraduate engineering teaching. Frontiers in Education Conference, pages 180­186, 1989. IEEE catalog number 89CH2737-5. [Lew75] Ellen Lewis. An introduction to ITS for the MACSYMA user. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1975. [Lew76a] E. Lewis. An introduction to ITS for the MACSYMA user. Mathlab Memo 3, Massachusetts Institute of Technology, A. I. Lab., Cambridge, MA, USA, April 1976. [Lew76b] Ellen Lewis. An introduction to ITS for the MACSYMA user. Mathlab memo 3, Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1976. [Lew78] Ellen Lewis. An introduction to ITS for the MACSYMA user. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, revised edition, 1978. [Lew79] V. Ellen Lewis, editor. Proceedings of the 1979 MACSYMA Users Conference: [held in] Washington, DC, USA, June 20­22, 1979, Cambridge, MA, USA, 1979. Massachusetts Institute of Technology, Laboratory for Computer Science. 32.4 [LF87] Zvi Ladin and Woodie Flowers. Propagation of kinematic disturbances in gait models. IEEE/Engineering in Medicine and Biology Society Annual Conference, pages 469­470, 1987. IEEE Service Cent. Piscataway, NJ, USA. [LG86] Francois LeGland and Antoine Gondel. Systematic numerical experiments in nonlinear filtering with automatic Fortran code generation. Proceedings of the IEEE Conference on Decision and Control Including The Symposium on Adaptive Pro, pages 638­642, 1986. IEEE Service Cent. Piscataway, NJ, USA. BIBLIOGRAPHY 145 [LH84] M. C. Leu and N. Hemati. Automated symbolic derivation of dynamic equations of motion for robotic manipulators. In Stelson and Sweet [SS84], pages 193­206. [LH86] M. C. Leu and N. Hemati. Automated symbolic derivation of dynamic equations of motion for robotic manipulators. Journal of Dynamic Systems, Measurement, and Control, 108(3):172­179, September 1986. [LHC92] Robert Y. Liang, Jialou Hu, and Fred Choy. Theoretical study of crack-induced eigenfrequency changes on beam structures. Journal of Engineering Mechanics, 118(2):384­394, February 1992. [Lit76] David Joseph Littleboy. An interactive primer for MACSYMA. Thesis (b.s.), Massachusetts Institute of Technology, Dept. of Electrical Engineering, 1976. [LMR90] Witold Litwin, Leo Mark, and Nick Roussopoulos. Interoperability of multiple autonomous databases. ACM Computing Surveys, 22(3):267­293, September 1990. Also published in/as: University of Maryland, Systems Research Center, TR-89-12 and CS TR-2188, March 1989. [LS96] X. Y. Lu and S. K. Spurgeon. Symbolic computation for dynamic sliding mode controller design. In IEE, editor, IEE Colloquium on Symbolic Computation for Control (Digest No: 1996/078), 2 April 1996, pages 4/1­4/5, London, UK, 1996. IEE. [Lue77] E. Lueken. Überlegungen zur Implementierung eines Formelmanipulationssystemes. (German) [Considerations for the implementation of formula manipulation systems]. Master's thesis, Technische Universität Braunschweig (??), Braunschweig, Germany, 1977. [Lun86] Michael Lund. Using MACSYMA to evaluate likelihood functions. Technical report 7317, Institutionen för reglerteknik, Lunds tekniska högskola, Lund, Sweden, 1986. [Mac83] Glenn A. Macala. Symbod: a computer program for the automatic generation of symbolic equations of motion for systems of hinge-connected rigid bodies. AIAA Paper, 1983. [MAC84] MACSYMA Applications Newsletter, July 1984. [Mac88] Macsyma, Inc. Macsyma mathematics reference manual. Macsyma, Inc., Arlington, MA, USA, version 13 edition, 1988. [Mac93a] Macsyma, Inc. Macsyma mathematics reference manual. Macsyma, Inc., Arlington, MA, USA, version 14 edition, 1993. [Mac93b] Macsyma, Inc. Macsyma system reference manual. Macsyma, Inc., Arlington, MA, USA, version 14 edition, 1993. [Mac94] Macsyma, Inc. Macsyma, 1994. 8 computer disks. [Mac95a] Macsyma, Inc. Macsyma graphics and user interface reference manual. Macsyma, Inc., Arlington, MA, USA, fifteenth edition, 1995. [Mac95b] Macsyma, Inc. Macsyma mathematics and system reference manual. Macsyma, Inc., Arlington, MA, USA, fifteenth edition, 1995. [Mac95c] Macsyma, Inc. Macsyma user's guide. Macsyma, Inc., Arlington, MA, USA, second edition, 1995. [Mac97a] Macsyma, Inc. Introduction to Macsyma. Jones and Bartlett, Sudbury, MA, USA, 1997. [Mac97b] Macsyma, Inc. Macsyma: with the new Math Tips natural language query, you can ask for help in your own words!; Version 2.2. Macsyma, Inc., Arlington, MA, USA, 1997. [Mac98] Macsyma Inc. Macsyma [Medienkombination]: for Windows 3.1, 95, & NT: with the new Math Tips natural language query, you can ask for help in your own word!: version 2.3. Springer Verlag, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1998. CD-ROM. BIBLIOGRAPHY 146 [Mae87] R. E. Maeder. Solving boundary-value problems with perturbations. SIGSAM Bulletin, 21(3):16­18, August 1987. [Mag89] J. F. Magnan. A MACSYMA program for the multiple bifurcation analysis of double-diffusive convection. Journal of Symbolic Computation, 7(2):189­198 (or 189­197??), February 1989. [Man93] Yiu-Kwong Man. Computing closed form solutions of first order ODEs using the Prelle-Singer procedure. Journal of Symbolic Computation, 16(5):423­444 (or 423­443??), November 1993. [Mr86] Bengt Mrtensson. Automatic TEX code generation from Macsyma and CTRL-C. Technical report 7334, Institutionen för reglerteknik, Lunds tekniska högskola, Lund, Sweden, 1986. [Mar87] Jocelyne Helene Marrannes Marbeau. Towards symbolic Kriging with the help of MACSYMA. Thesis (m.a.), University of Denver, Denver, CO, USA, 1987. [Mat71] Mathlab Group. The MACSYMA papers 1970. Massachusetts Institute of Technology, Mathlab Group, Cambridge, MA, USA, 1971. [Mat74] Mathlab Group. MACSYMA primer: introductory section. Cambridge, MA, USA, 1974. [Mat75] Mathlab Group. MACSYMA primer. Report, Massachusetts Institute of Technology, A. I. Lab., Cambridge, MA, USA, October 1975. [Mat77] Mathlab Group. MACSYMA Reference Manual, Version 9. Massachusetts Institute of Technology, Computer Science Lab., Cambridge, MA, USA, December 1977. [Mat80] Mathlab Group. Addendum to the Mathlab/MIT MACSYMA reference manual for VAX/UNIX version and VAX/VMS version. Computing Services, University of California, Berkeley, Berkeley, CA, USA, 1980. [MAT83a] MATHLAB Group. MACSYMA Reference Manual. Massachusetts Institute of Technology, Computer Science Lab., Cambridge, MA, USA, tenth edition, January 1983. [Mat83b] Mathlab Group. MACSYMA Reference Manual, Version Ten. Massachusetts Institute of Technology, Computer Science Lab., Cambridge, MA, USA, 1983. [Mat89a] J. Mathews. Using a symbol manipulation program in statics to compute centroids and moments. CoED, 9(1):52­55, January­March 1989. [Mat89b] J. H. Mathews. Using a computer algebra system to teach second order differential equations. CoED, 9(4):7­10, October­December 1989. [MB75] Mathlab Group and Richard A. Bogen. MACSYMA reference manual. Massachusetts Institute of Technology, Project MAC, Cambridge, MA, USA, version eight edition, 1975. [MD88] Jane Macfarlane and Max Donath. Automated symbolic derivation of state equations for dynamic systems. In IEEE [IEE88b], pages 215­222. IEEE catalog number 88CH2552-8. [Meh86] Unmeel Mehta. Knowledge based systems for computational aerodynamics and fluid dynamics. In Knowledge based problem solving [Kow86], pages 183­212. [Mej84] Raymond Mejia. Some applications of symbolic manipulation in biomathematics. In Golden and Hussain [GH84], pages 35­?? [MH83] Johan Mller-Holst. Introduksjon til MACSYMA. Kjeller: Forsvarets forskningsinstitut, ???, Norway, 1983. [Mig87] Maurice Mignotte. Inequalities about factors of integer polynomials. SIGSAM Bulletin, 21(4):24­24, November 1987. [Mil93] H. S. D. Mills. Symbolically precise solutions to a homogeneous second order matrix ordinary differential equation with Macsyma. Journal of Symbolic Computation, 15(1):91­98, January 1993. BIBLIOGRAPHY 147 [Mio90] A. Miola, editor. Design and implementation of symbolic computation systems: International Symposium DISCO '90, Capri, Italy, April 10­12, 1990: proceedings, volume 429 of Lecture Notes in Computer Science, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1990. Springer Verlag. 32.4 [ML87] A. Meystel and J. Y. S. Luh, editors. Proceedings / IEEE International Symposium on Intelligent Control 1987, 19­20 January 1987, Philadelphia, Pennsylvania, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1987. IEEE Computer Society Press. 32.4 [MM70] W. A. Martin and J. Moses. Mathlab (MACSYMA). MAC Programming Report V, Massachusetts Institute of Technology, A. I. Lab., Cambridge, MA, USA, July 1969 to December 1970 1970. [Mon92] Michael B. Monagan. A heuristic irreducibility test for univariate polynomials. Journal of Symbolic Computation, 13(1):47­58 (or 47­57??), January 1992. [Mos74] Joel Moses. MACSYMA -- the fifth year. In Richart D. Jenks, editor, Proceedings of EUROSAM '74, Royal Institute of Technology, Stockholm, Sweden, August 1­2, 1974, volume 8(3) of SIGSAM Bulletin, pages ??­??, New York, NY, USA, 1974. ACM Press. [Mos75] J. Moses. A MACSYMA primer. Mathlab Memo 2, Massachusetts Institute of Technology, Computer Science Lab., 1975. [MR85a] L. A. Month and R. H. Rand. Stability of a rigid body with an oscillating particle: an application of Macsyma. American Society of Mechanical Engineers (Paper), 1985. [MR85b] L. A. Month and R. H. Rand. Stability of a rigid body with an oscillating particle: an application of Macsyma. Journal of Applied Mechanics, Transactions ASME, 52(3):686­692, September 1985. [MSI90] Rafi Manor, Beni Shalom, and Adrian Ioinovici. Simulation of cyclically switching systems using the generalized alternor definition. International Journal of Systems Science, 21(7):1281­1287, July 1990. [MT94] P. Mitic and P. G. Thomas. Pitfalls and limitations of computer algebra. Computers and Education, 22(4):355­361, May 1994. [Mur85] F. A. Murzin. Syntactic properties of the REFAL language. International J. Computer Mathematics, 17:123­139, 1985. [NA79] A. K. Noor and C. M. Andersen. Computerized symbolic manipulation in structural mechanics -- progress and potential. Computers and Structures, 10:95­118, 1979. [NC79] E. Ng and Bruce W. Char. Gradient and Jacobian computation for numerical applications. In Lewis [Lew79], pages 604­621. [Nei80] Anne D. Neirynck. Partial fraction expansion routines for Vaxima. Master of science, plan ii., Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1980. [Nel89] James K. Nelson, Jr., editor. Computer utilization in structural engineering: proceedings of the sessions related to computer utilization at Structures Congress '89: San Francisco Hilton, San Francisco, CA, May 1­5, 1989, New York, NY, USA, 1989. American Society of Civil Engineers. 32.4 [NMM90] Glen C. Nielsen, Mark O. McLinden, and Graham Morrison. Use of computer algebra to locate critical loci in fluid mixtures. Journal of Symbolic Computation, 10(5):499­508, November 1990. [NNM91] J. F. Nayfeh, A. H. Nayfeh, and D. T. Mook. Nonlinear response of thick laminated composite plates. In Huang et al. [H+91], pages 321­?? [NTT86] S. Nicosia, P. Tomei, and A. Tornambe. Dynamic modelling of flexible robot manipulators. In IEEE [IEE86], pages 365­372. IEEE Service Cent. Piscataway, NJ, USA. BIBLIOGRAPHY 148 [NTT90] S. Nicosia, P. Tomei, and A. Tornambe. Discrete-time modeling of flexible robots. Proceedings of the IEEE Conference on Decision and Control, 2:539­544, 1990. IEEE catalog number 90CH2917-3. [NW83] Arthur C. Norman and Paul S. Wang. A comparison of the Vaxima and REDUCE factorization packages. SIGSAM Bulletin, 17(1):28­30, February 1983. [OA89] E. Oberaigner and K. Aziz. Use of symbolic computation in petroleum engineering. Society of Petroleum Engineers of AIME, (Paper) SPE, 1989. [OA91] Eduard Oberaigner and Khalid Aziz. Use of symbolic computation in petroleum engineering. Journal of Petroleum Science & Engineering, 5(3):237­246, April 1991. [Ols92] Andrew M. Olson. Object-oriented analysis model of an iconic interface to Macsyma. In IEEE [IEE92b], pages 253­260. IEEE Catalog number 92THO438-2. [OM92] Paul D. Orkwis and D. Scott McRae. Newton's method solver for high-speed viscous separated flowfields. American Institute of Aeronautics and Astronautics Journal, 30(1):78­85, January 1992. [Ous91a] N. E. Oussous. Computation, on Macsyma, of the minimal differential representation of noncommutative polynomials. Theoret. Comput. Sci., 79(1):195­207, February 1991. [Ous91b] N. E. Oussous. Macsyma computation of local minimal realization of dynamical systems of which generating power series are finite. Journal of Symbolic Computation, 12(1):115­126, July 1991. [Pai92] J. F. Painter. The matrix editor for symbolic Jacobians in ALPAL. In Wang [Wan92], pages 312­319. ACM order number: 505920. [Par84] Paradigm Associates. MACSYMA applications newsletter, 1984. [Par86] Paradigm Associates, Inc. MACSYMA Applications Newsletter, 1986. [Pav85a] R. Pavelle. Macsyma -- capabilities and applications to problems in engineering and the sciences. In Buchberger [Buc85], pages 19­32. [Pav85b] Richard Pavelle. MACSYMA: capabilities and applications to problems in engineering and the sciences. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, 1985. [Pav86] Richard Pavelle. Computer algebra: Capabilities and applications to problems in engineering and the sciences. In Pierce and Hohne [PH86], pages 100­110. [PEK91] D. W. Pepper, Ashley F. Emery, and Matthew B. Kelleher, editors. Computational techniques and numerical heat transfer on PCs and workstations: presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, Atlanta, Georgia, December 1­6, 1991, volume 185 of Heat Transfer Division, 345 E. 47th St., New York, NY 10017, USA, 1991. American Society of Mechanical Engineers. 32.4 [Pet88] R. Petti. Role of symbolic mathematics software in mathematical modeling. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:13­20, 1988. [PF95] Jerry Place and Sue Fitzgerald. Using a computer algebra system (Maple) to teach elementary queueing theory. Computer Applications in Engineering Education, 3(1):65­73, 1995. [PH86] Thomas H. Pierce and Bruce A. Hohne, editors. Artificial intelligence applications in chemistry, volume 306 of ACS symposium series, Washington, DC, USA, 1986. American Chemical Society. 32.4 [Pit79] K. M. Pitman. A FORTRAN - LISP translator. In Lewis [Lew79], pages ??­?? [PK88] A. R. Phelps and A. J. Krener. Computation of observer normal form using Macsyma. In Byrnes et al. [BMS88], pages 475­482. Selected papers from the 8th International Symposium on the Mathematics of Networks and Systems, held in Phoenix, June 15­19, 1987. BIBLIOGRAPHY 149 [PL87] Alkesh Punjabi and Maria Lam. Solutions of some problems in applied mathematics using MACSYMA. NASA contractor report NASA-CR 180299, NASA, Washington, DC, USA, 1987. [Pow84] Carl Robert Powell. An automatic testing facility for Vaxima. Thesis (m.s.), Kent State University, Kent, OH, USA, 1984. [Pro74] Project Mac (Massachusetts Institute of Technology). Mathlab Group. MACSYMA reference manual: version seven. Massachusetts Institute of Technology, Cambridge, MA, USA, 1974. [Pur85] James Purtilo. Polylith: An environment to support management of tool interfaces. In ACM SIGPLAN 85 [ACM85], pages 12­18. Published in ACM SIGPLAN notices, volume 20, number 7. [PW85a] Richard Pavelle and Paul S. Wang. MACSYMA from F to G. Journal of Symbolic Computation, 1(1):69­100, March 1985. [PW85b] Richard Pavelle and Paul S. Wang. MACSYMA from F to G. Journal of Symbolic Computation, 1(1):69­100, March 1985. [PW85c] Richard Pavelle and Paul S. Wang. MACSYMA from F to G. Academic Press, New York, USA, 1985. [RA87] Richard H. Rand and Dieter Armbruster. Perturbation methods, bifurcation theory, and computer algebra. Number 65 in Applied mathematical sciences. Springer Verlag, Berlin, Germany / Heidelberg, Germany / London, UK / etc., 1987. [Raj87] N. S. Rajaram. Expert systems development: Current problems, future needs. InTech, 34(4):25­26, April 1987. [RAKK88] Dietmar Rempfer, Monika Auweter-Kurtz, and Hans J. Käppeler. Aufstellung und Lösung der Dispersionsgleichung zur Untersuchung von Instabilitäten in MPD-Triebwerken mit Hilfe eines MACSYMA-FORTRANHybrid-Codes. (German) [List and solution of dispersion equations for the investigation of instabilities in mpd-engines with the help of macsyma-fortran hybrid codes]. Report 88-S18, Institut für Raumfahrtsysteme, Universität Stuttgart, Stuttgart, Germany, 1988. [Ran84] R. H. (Richard H.) Rand. Computer algebra in applied mathematics: an introduction to MACSYMA. Number 94 in Research notes in mathematics. Pitman Publishing Ltd., London, UK, 1984. [Ran87] R. H. Rand. Computer algebra applications using MACSYMA. In Heiberger [Hei87], pages 231­236. [Ran88] R. H. Rand. Use of symbolic computation in perturbation analysis. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:41­45, 1988. [Ran94] Richard H. Rand. Topics in nonlinear dynamics with computer algebra, volume 1 of Computation in education. Langhorne: Gordon and Breach Science, Newark, NJ, USA, 1994. [Ray88] Mohamed Omar Rayes. Enhancement of the complex environment of MACSYMA. Thesis (m.a.), Dept. of Mathematical Sciences, Kent State University, Kent, OH, USA, 1988. [RB91] G. J. Reid and A. Boulton. Reduction of systems of differential equations to standard form and their integration using directed graphs. In Watt [Wat91], pages 308­312. [RCF98] Darren Redfern, Edgar Chandler, and Richard N. Fell. Macsyma ODE lab book. Jones and Bartlett, Sudbury, MA, USA, 1998. [RDBE87] M. L. Rehak, F. L. Dimaggio, H. Benaroya, and I. Elishakoff. Random vibrations with Macsyma. Computer Methods in Applied Mechanics and Engineering, 61(1):61­70, March 1987. [Rei81] Allan Reiman. Computer-aided closure of the Lie algebra associated with a nonlinear partial differential equation. Computers and Mathematics with Applications, 7(5):387­393, ???? 1981. BIBLIOGRAPHY 150 [RH86] Raymond Rishel and Lawrence Harris. Algorithm for a solution of a stochastic adaptive linear quadratic optimal control problem. IEEE Transactions on Automatic Control, AC-31(12):1165­1170, December 1986. [RK87] R. H. Rand and W. L. Keith. Determinacy of degenerate equilibria with linear part x = y and y = 0 using Macsyma. Applied Mathematics and Computation, 21(1):1­20, January 1987. [Roe95] K. G. Roesner. Verified solutions for parameters of an exact solution for non-Newtonian liquids using computer algebra. Zeitschrift für Angewandte Mathematik und Mechanik, 75(suppl. 2):S435­438, 1995. [RS85] Patrick J. Roache and Stanly Steinberg. New approach to grid generation using a variational formulation. AIAA Paper, pages 360­370, 1985. [RS89] K. T. Rowney and R. D. Silverman. Finite field manipulations in Macsyma. SIGSAM Bulletin, 23(1):39­48, January 1989. [RW86] D. W. Rand and P. Winternitz. Odepainleve -- a Macsyma package for Painlevé analysis of ordinary differential equations. Computer Physics Communications, 33(12):359­383, December 1986. [SB89] Wayne C. Schou and Kevin A. Broughan. The Risch algorithms of MACSYMA and SENAC. SIGSAM Bulletin, 23(3):19­22, July 1989. [SC87] A. F. Saleeb and T. Y. Chang. Efficient quadrilateral element for plate bending analysis. International Journal for Numerical Methods in Engineering, 24(6):1123­1155, June 1987. [SC90] Zhengwei Su and Philip Coppens. Closed-form expressions for Fourier-Bessel transform of Slatertype functions. Journal of applied crystallography, 23(1):71­73, February 1990. [SCG88] R. Shtokhamer, B. F. Caviness, and R. P. Gilbert. An introduction to applied symbolic computation using MACSYMA, 1988. [Sch84] F. Schwarz. Algebra mit dem Computer. GMD-SPIEGEL, 2:4­6, 1984. [SDF88] S. J. Seda, M. G. R. Degrauwe, and W. Fichtner. A symbolic analysis tool for analog circuit design automation. In Digest of technical papers / IEEE International Conference on Computer-Aided Design, ICCAD-88, a conference for the EE CAD professional, November 7­10, 1988, Convention Center, Santa Clara, California, pages 488­491, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. Computer Society order number 869. IEEE catalog number 88CH2657-5. [SH84] J. F. Schenck and M. A. Hussain. Application of MACSYMA to a boundary value problem arising in nuclear magnetic resonance imaging. In Golden and Hussain [GH84], pages 38­?? [Sir70] Yvon Siret. Contribution au calcul formel sur ordinateur. (Frencg) [Contribution to formal calculus on a computer]. Thése s Sciences appliquées, Université de Grenoble, Grenoble, France, 1970. [SK86] N. Sreenath and P. S. Krishnaprasad. Dynaman: a tool for manipulator design and analysis. In IEEE [IEE86], pages 836­842. IEEE Service Cent. Piscataway, NJ, USA. [Ski93] L. A. Skinner. Matched asymptotic expansions of integrals. IMA Journal of Applied Mathematics (Institute of Mathematics & Its Applications), 50(1):77­90, 1993. [Slo86] N. J. A. Sloane. My friend MACSYMA. Notices Amer. Math. Soc., 33(1):40­43, January 1986. [SM84] Symbolics, Inc. and Mathlab Group. MACSYMA reference manual. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, version ten edition, 1984. Two volumes. [SM85] Symbolics, Inc. and Mathlab Group. VAX UNIX MACSYMA reference manual. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, version 11 edition, 1985. BIBLIOGRAPHY 151 [SP93] R. A. Smith and A. N. Palazotto. Comparison of eight variations of a higher-order theory for cylindrical shells. American Institute of Aeronautics and Astronautics Journal, 31(6):1125­1132, June 1993. [Spi86] Lilly Spirkovska. MUFIE, Macsyma's User-Friendly Interactive Executive: research project. Master of sciences, plan ii, Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA, 1986. [SR84] Stanley Steinberg and P. Roache. Using VAXIMA to write Fortran code. In Golden and Hussain [GH84], pages 1­?? [SR86] Stanly Steinberg and Patrick J. Roache. Using Macsyma to write FORTRAN subroutines. Journal of Symbolic Computation, 2(2):213­216, June 1986. [SR88] S. Steinberg and P. Roache. Automatic generation of finite difference code. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:81­86, 1988. [SR90] Stanly Steinberg and Patrick J. Roache. Using MACSYMA to write finite-volume based PDE solvers. American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP, 205:81­96, 1990. [Sre92] N. Sreenath. Hybrid computation environment for multibody simulation. Mathematics and Computers in Simulation, 34(2):121­140, August 1992. [SS84] K. A. Stelson and L. M. Sweet, editors. Sensors and controls for automated manufacturing and robotics: presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, New Orleans, Louisiana, December 9­14, 1984, 345 E. 47th St., New York, NY 10017, USA, 1984. American Society of Mechanical Engineers. 32.4 [Sta84] Donald F. Stanat. A functional language machine and its programming. In Golden and Hussain [GH84], pages 371­?? [Sto84] David R. Stoutemyer. Which polynomial representation is best? In Golden and Hussain [GH84], pages 221­?? [Str85] Nicholas Strauss. Jordan form of (i + j, j) over Zp. Working Paper 275, Massachusetts Institute of Technology, Computer Science Lab., Cambridge, MA, USA, July 1985. [Sua84] M. L. Suarez. Computer algebra applied to Kalman filtering. In Golden and Hussain [GH84], pages 188­?? [Sym84] Symbolics, Inc. Computer Aided Mathematics Group. MACSYMA newsletter, 1984. [Sym85a] Symbolics, Inc. An Introduction to MACSYMA. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, 1985. [Sym85b] Symbolics, Inc. An introduction to UNIX MACSYMA. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, version 3.0 edition, 1985. [Sym86a] Symbolics, Inc. MACSYMA Newsletter, 1986. [Sym86b] Symbolics, Inc., Computer Aided Mathematics Group. An Introduction to MACSYMA for Symbolics computers. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, version 4.0 edition, 1986. [Sym87] Symbolics, Inc., Computer Aided Mathematics Group. Bibliography of publications referencing MACSYMA. Symbolics, Inc., 11 Cambridge Center, Cambridge MA 02142, USA, 1987. [Sym88a] Symbolics, Inc., Burlington, MA, USA. Macsyma User's Guide, 1988. [Sym88b] Symbolics, Inc., Computer Aided Mathematics Group. MACSYMA reference manual. Symbolics, Inc., Burlington, MA, USA, version 13 edition, 1988. BIBLIOGRAPHY 152 [TD90] H. Q. Tan and X. Dong. Optimization techniques for symbolic equation solver in engineering applications. In Watanabe and Nagata [WN90], pages 305­?? [TDA88] H. Q. Tan, X. Dong, and S. M. Arnold. Symbolic derivation of constitutive equations. American Society of Mechanical Engineers, Heat Transfer Division, (Publication) HTD, 105:103­109, 1988. [TG90] P. Thejll and R. P. Gilbert. The use of MACSYMA for solving elliptic boundary value problems. Zeitschrift für Angewandte Mathematik und Mechanik, 70(11):479­??, 1990. [TH90] S. Toyama and S. Hatae. SYMTOM: an algebraic robotic manipulator modelling program. In IEEE, editor, IEEE International Workshop on Intelligent Robots and Systems '90. 'Towards a New Frontier of Applications', Proceedings. IROS '90, July 3-6, 1990, Tsuchiura-shi, Ibaraki, Japan, volume 1, pages 197­204, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1990. IEEE Computer Society Press. [TH94] M. M. Tomovic and V. S. Hillman. Computer algebra systems enhance teaching engineering technology courses. In Grayson [Gra94], pages 161­164. [Tha89a] C. Thas. A collection of REDUCE and MACSYMA programs about college geometry. part 1. Technical Report 5, State University of Gent, Gent, Belgium, September 1989. [Tha89b] C. Thas. A collection of REDUCE and MACSYMA programs about college geometry. part 2. Technical Report 5, State University of Gent, Gent, Belgium, September 1989. [The83a] The Mathlab Group, Lab. for Computer Science, MIT. MACSYMA Reference Manual, Version 10, Volume I. Symbolics, Inc., Burlington, MA, USA, second printing edition, December 1983. [The83b] The Mathlab Group, Lab. for Computer Science, MIT. MACSYMA Reference Manual, Version 10, Volume II. Symbolics, Inc., Burlington, MA, USA, second printing edition, December 1983. [TM85] Ph. Tombal and A. Moussiaux. MACSYMA computation of the Dirac-Bergmann algorithm for Hamiltonian systems with constraints. Journal of Symbolic Computation, 1(4):419­421, December 1985. [TM89] Ph. Tombal and A. Moussiaux. Algebraic programming of geometrical calculus and Clifford algebra. Journal of Symbolic Computation, 7(1):85­92 (or 85­91??), January 1989. [TMH99] M. O. Tokhi, Z. Mohamed, and A. W. I. Hashim. IEE colloquium on symbolic computation for control (ref. no. 1999/088), 17 June 1999. In IEE, editor, Application of symbolic manipulation to the analysis of dynamic characteristics of a flexible robot manipulator, pages 7/1­7/5, London, UK, 1999. IEE. [Top89] B. H. V. Topping, editor. Civil-Comp 89: proceedings of the Fourth International Conference on Civil and Structural Engineering Computing: Sep 19­21 1989: London, England, Edinburgh, Scotland, 1989. Civil-Comp Limited. 32.4 [TRC92] Matthew J. Targett, William B. Retallick, and Stuart W. Churchill. Solutions in closed form for a double-spiral heat exchanger. Industrial and engineering chemistry research, 31(3):658­669, March 1992. [TS00] B. Turetken and S. Eren San. Comparison of symbolic computation techniques for problems in electromagnetics. In IEEE, editor, International Conference on Mathematical Methods in Electromagnetic Theory, 2000. MMET 2000, 12­15 September 2000, volume 1, pages 172­174, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 2000. IEEE Computer Society Press. IEEE catalog number 00EX413. [TT88] Patrizio Tomei and Antonio Tornambe. Approximate modeling of robots having elastic links. IEEE Transactions on Systems, Man, and Cybernetics, 18(5):831­840, September­October 1988. [TTDD91] Kevin G. TeBeest, Steven A. Trogdon, Jeffrey R. Dahl, and Rod W. Douglass. Natural convection within spherical annuli by symbolic algebra on workstations. In Pepper et al. [PEK91], pages 101­ 106. BIBLIOGRAPHY 153 [TV89] E. Tunstel and N. Vira. Mechanization of manipulator kinematic equations via MACSYMA. Computers in Engineering, Proceedings of the International Computers in Engineering Conference and Exhibit, pages 649­655, 1989. [TYL88] Anthony P. Tzes, Stephen Yurkovich, and F. Dieter Langer. Symbolic manipulation package for modeling of rigid or flexible manipulators. In IEEE [IEE88a], pages 1526­1531. IEEE Service Cent. Piscataway, NJ, USA. [TYL89] Anthony P. Tzes, Stephen Yurkovich, and F. Dieter Langer. Method for solution of the Euler-Bernoulli beam equation in flexible-link robotic systems. In Proceedings of the IEEE International Conference on Systems Engineering August 24-26 1989 Dayton, Ohio (Aug 24­26 1989: Fairborn, OH, USA), pages 557­560, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1989. IEEE Computer Society Press. IEEE catalog number 89CH2767-2. [Una88] A. Unal. An algebraic criterion for the onset of chaos in nonlinear dynamical systems. In IEEE, editor, IEEE International Symposium on Circuits and Systems, Helsinki University of Technology, Espoo, Finland, June 7-9, 1988, volume 1, pages 19­22, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1988. IEEE Computer Society Press. [van82] J. A. van Hulzen. Computer algebra systems viewed by a notorious user. Lecture Notes in Computer Science, 144:166­180, 1982. [VGT90] N. Vira, T. Gill, and E. Tunstel. Application of symbolic computation in robot pose error modeling. Journal of Symbolic Computation, 10(5):509­524 (or 509­523??), November 1990. [VT92] N. Vira and E. Tunstel. Use of symbolic computation in robotics education. IEEE Transactions on Education, 35(1):18­30, February 1992. [Wan79] P. Wang. Parallel p-adic construction in the univariate polynomial factoring algorithm. In Lewis [Lew79], pages 310­317. [Wan81] Paul S. Wang, editor. SYMSAC '81: proceedings of the 1981 ACM Symposium on Symbolic and Algebraic Computation, Snowbird, Utah, August 5­7, 1981, New York, NY, USA, 1981. ACM Press. ACM order no. 505810. 32.4 [Wan82] Paul S. Wang. Hacijan's algorithm in Vaxima: Improvements and difficulties. Lecture Notes in Computer Science, 144:135­143, 1982. [Wan84] Paul S. Wang. MACSYMA-aided finite element analysis. In Golden and Hussain [GH84], pages 23­?? [Wan85] Paul S. Wang. Combining symbolic and numerical computational techniques on modern workstations. Proceedings of the Hawaii International Conference on System Science, pages 248­??, 1985. [Wan91] D. Wang. A toolkit for manipulating indefinite summations with application to neural networks. In Watt [Wat91], pages 462­463. [Wan92] Paul S. Wang, editor. Proceedings of ISSAC '92. International Symposium on Symbolic and Algebraic Computation, New York, NY, USA, 1992. ACM Press. ACM order number: 505920. 32.4 [Wan94] Dong Ming Wang. Differentiation and integration of indefinite summations with respect to indexed variables -- some rules and applications. Journal of Symbolic Computation, 18(3):249­264 (or 249­ 263??), September 1994. [War90] Thomas L. Ward. Symbolic mathematical computation in engineering economy. CoED (Journal) (Computers in Education Division of ASEE), 10(1):14­18, January­March 1990. [Wat91] Stephen M. Watt, editor. Proceedings of the 1991 International Symposium on Symbolic and Algebraic Computation. ISSAC '91, July 15­17, 1991, Bonn, Germany, New York, NY, USA, 1991. ACM Press. 32.4 BIBLIOGRAPHY 154 [Way90] Peter Wayner. Mainframe math on a PC: Macsyma, the grande dame of computer algebra, is finally available for PCs. BYTE Magazine, 15(1):207­208, 210, January 1990. [WH84] Ralph Wilcox and Leo Harten. Analytical solutions to some linear systems of differential equation. In Golden and Hussain [GH84], pages 138­?? [Whi77a] J. L. White. LISP: Data is program, a tutorial in LISP. In Fateman et al. [F+77], pages ??­?? [Whi77b] J. L. White. LISP: Program is data, a historical perspective on MACLISP. In Fateman et al. [F+77], pages ??­?? [WKB86] S. J. Watowich, J. L. Krause, and R. S. Berry. Stability analysis of an optimally controlled light-driven engine. Journal of Symbolic Computation, 2(1):103­108, March 1986. [WN90] Shunro Watanabe and Morio Nagata, editors. ISSAC '90: proceedings of the International Symposium on Symbolic and Algebraic Computation: August 20­24, 1990, Tokyo, Japan, New York, NY, USA and Reading, MA, USA, 1990. ACM Press and Addison-Wesley. 32.4 [Wol79] Steve Wolfram. MACSYMA tools for Feynman diagram calculations. In Lewis [Lew79], pages ??­ ?? [Wol84] Steve Wolfram. Five years of SMP. In Golden and Hussain [GH84], pages 220­?? [Woo84] David L. Wood. An overdetermined system of partial differential equations. In Golden and Hussain [GH84], pages 121­?? [WS83] Michael Wester and Stanly Steinberg. An extension to MACSYMA's concept of functional differentiation. SIGSAM Bulletin, 17(3/4):25­30, August/November 1983. [WS84] Michael Wester and Stanly Steinberg. A survey of symbolic differentiation implementations. In Golden and Hussain [GH84], pages 330­?? [WS88] C. F. Weggel and D. P. Schwartz. New analytical formulas for calculating magnetic field. IEEE Transactions on Magnetics, 24(2 (part 2)):1544­1547, March 1988. [WS91a] D. Wang and B. Schurmann. Computer aided investigations of artificial neural systems. In IEEE, editor, International Joint Conference on Neural Networks, 1991., IJCNN-91-Seattle, 8­14 July 1991, volume 2, pages 981­981, 1109 Spring Street, Suite 300, Silver Spring, MD 20910, USA, 1991. IEEE Computer Society Press. [WS91b] Dongming Wang and Bernd Schurmann. Computer aided investigations of artificial neural systems. In IEEE [IEE91], pages 2325­2330. Three volumes. IEEE catalog number 91CH3065-0. [WS92] D. Wang and B. Schurmann. Computer aided analysis and derivation for artificial neural systems. IEEE Transactions on Software Engineering, 18(8):728­735, August 1992. [Yag84] J. J. Yagla. Stability criteria for finite difference equations. In Golden and Hussain [GH84], pages 294­?? [YP91] Robert J. Yamartino and Richard Pavelle. An application of computer algebra to a problem in stratified fluid flow. Journal of Symbolic Computation, 12(6):669­672, December 1991. [yS88] Chiung yu Syu. An application of MACSYMA in fluid mechanics. Thesis (m.s.m.e.), University of Massachusetts at Amherst, Amherst, MA, USA, 1988. [YW85] Douglas A. Young and Paul S. Wang. An improved plotting package for Vaxima. In Buchberger [Buc85], page 431. [YW87] D. A. Young and P. S. Wang. GI/S: a graphical user interface for symbolic computation systems. Journal of Symbolic Computation, 4(3):365­380, December 1987. [ZZ86] S. W. Zewari and J. M. Zugel. Prolog implementation in robot kinematics. In Gupta et al. [GA+86], pages 133­136. Three volumes.