Communication Scientific CITS7200
Computer Science & Software Engineering
Lecture 1
A LTEX There are many computer tools available for the production of technical documents, and you must master those that you will need for your thesis. The tools I have chosen are commonly used by computer science graduates and researchers around the world and produce professional quality documents. A Thus, this lecture will briefly introduce L TEX and how to get started with A it, the ispell program for spell-checking L TEX source, xfig for generating A X document, and gnuplot for generdrawings that can be inserted in a L TE ating graphs. The use of bibliographic data files and the inclusion of images in your documents will also be covered.
1
A LTEX
A L TEX is a text mark-up language that produces professionally typeset docA uments. TEX, the predecessor of L TEX, was developed by Donald Knuth in the early 1970's for automatically typesetting books. It is a program that accepts a text file with certain formatting commands as input, and generates a dvi, or device independent, file as output. The dvi file can be readily translated into languages that can be understood by screens or printers. A L TEX, written by Leslie Lamport in the early 1980's, is a macro package A that sits on top of TEX. L TEX adds to TEX a collection of commands that
1
A simplify typesetting. L TEX is widely used by computer scientists around the world for a number of reasons, some of which are the following:
It is available on many different computers, including PCs and Macintoshes, and generates output on many different devices ranging from dot matrix printers to professional typesetting machines.
A Many journals and conferences accept papers in L TEX format or, inA deed, insist upon it. Direct reproduction from L TEX (or dvi or PostScript) source helps reduce errors.
The format of your document is governed by style options that are either specified at the beginning of your source, or included in your own customized style files. This means that if you need to change the format of your paper, you need only alter the style options.
A L TEX source is ASCII text. It can be sent by email, and the use of A L TEX formatting conventions is a standard way of sending mathematical symbols through email.
The dvi, PostScript or pdf translations are good for distributing your paper electronically, especially if you want to make it available via anonymous ftp across the Internet or down-loadable from your web page. Cross-references to equations, theorems, tables, figures and references are fully supported using symbolic labels. The re-numbering of crossreferences is done automatically when you add a new equation or reference to your paper.
A It is very easy to prepare overhead transparencies from your L TEX source by simply cutting and pasting the appropriate text into a slide file. Additionally, you can use TexPoint, which is a freeware available A for including L TEX directly into your PowerPoint slides.
Indexing is supported. You can use a key with the command \index so that items that should be included in an index to your document are noted with the page number on which they appear. Manual entries are facilitated.
A I have placed a number of links to L TEX on the unit webpage.
2
1.1
Getting started
A Usually you will prepare an ASCII file of L TEX source using your favourite editor. The source needs a preamble declaring the document style, and follows a \begin{document} command. For example, these lecture notes have the following preamble:
\documentclass[12pt, a4paper]{article} \usepackage{psfig} \parindent 0pt \parskip 5pt \title{Scientific Communication CITS7200} \author{Computer Science \& Software Engineering} \date{} \begin{document} \maketitle . . . \end{document} This preamble declares the document class as that of an article. Without any other parameters, the article class will produce 10pt text in U.S. letter format. This document has used 12pt text with the A4 format. The other commands say that the package psfig will be used for including postscript figures, paragraphs will have no indentation, and there will be a 5 point spacing between paragraphs, and it gives a title and author of the document, with no date included. If \date is left out then the current date will appear on the document. After the \begin{document} command the main part of your source text is entered, and the document concludes with a \end{document} command. The easiest way to process a document is to have the editor running in one window, to run latex file.tex in another, and to run xdvi file.dvi & in the background. This way you can edit and view your file almost simultaneously.
A Most L TEX commands describe the logical structure of the document.
To describe simple sentences, you merely type in the text as it comes. The ends of words and sentences are marked by spaces, and extra spaces are ignored. One or more blank lines denote the end of a paragraph. The tilde character ~ produces an ordinary inter-word space at which TEX will never 3
break a line. This is useful for terms such as "Chapter 1", which look silly if the "1" appears on the next line, or for citations.
A Emphasized text in L TEX is printed in italics. It is generated by the command \em. For example {\em This text is emphasized.}, produces This text is emphasized. Since emphasized text leans to the right, you sometimes need to add extra space. You do this with the \/ command, as in ``I {\em do}\/ like you.'' to produce "I do like you." The emphasis command is a toggle command, so that emphasis within emphasis reverts to plain text. Notice that opening quotes are produced by two single left quotes, whilst closing quotes are produced by two single right quotes (a double quote will do here too). Similarly, bold font is produced by the command {\bf } as in
``some of this {\bf text} is in bold.'' to produce "some of this text is in bold."
A L TEX recognises four kinds of dashes: the hyphen, such as in user-friendly; the dash between ranges, such as pages 1537; the dash between phrases, such as "Which do you prefer--the white or the black?"; and the mathematA ical minus sign, -1. These dashes are generated by the L TEX commands -, --, --- and $-$ respectively.
If you are using ellipses, as in "This program first processes a, then b, then c, . . . ", you must use the command \ldots to produce the three dots, rather A than just typing three dots. This is because L TEX produces the dots as they would be typeset, whereas typeset periods are set to occupy a small amount of space close to the preceding letter.
A L TEX provides three list environments: these are generated by the commands itemize, enumerate or description. They work as follows:
Type Programs are made up of \begin{itemize} \item Algorithms. This is the underlying idea that solves the problem. \item Data Structures. This is the format in which the data is best presented for this problem, and \item Data. This is the input to the algorithm. \end{itemize} to produce 4
Programs are made up of Algorithms. This is the underlying idea that solves the problem. Data Structures. This is the format in which the data is best presented for this problem, and Data. This is the input to the algorithm. The same input with enumerate in place of itemize gives: Programs are made up of 1. Algorithms. This is the underlying idea that solves the problem. 2. Data Structures. This is the format in which the data is best presented for this problem, and 3. Data. This is the input to the algorithm. The description command works as follows: Programs are made up of \begin{description} \item [Algorithms.] This is the underlying idea that solves the problem. \item [Data Structures.] This is the format in which the data is best presented for this problem, and \item [Data.] This is the input to the algorithm. \end{description} to produce: Programs are made up of Algorithms. This is the underlying idea that solves the problem. Data Structures. This the is format in which the data is best presented for this problem, and Data. This is the input to the algorithm. 5
A The power of L TEX becomes evident when you have mathematical symbols or equations in your paper. The math, displaymath, and equation environments put TEX in the mathematics mode. There are also short forms of these environments, the most common being $...$ and \[...\].
Subscripts and superscripts are made with the _ and ^ commands. The \frac command is used for large fractions in displayed formulas.
A There is an excellent summary of TEX and L TEX symbol commands given in your Readings and References handout for this unit. A Here is an example of L TEX source and the output:
Integrate $\int x^2e^{x^3}dx$. Let $u = x^3$. Then $du = 3x^2dx$, and we obtain \begin{equation} \int x^2e^{x^3}dx = \frac{1}{3} \int e^{x^3}(3x^2dx) = \frac{1}{3} \int e^udu = \frac{1}{3} e^u + C = \frac{1}{3} e^{x^3} + C \end{equation} Integrate obtain
3
x2 ex dx. Let u = x3 . Then du = 3x2 dx, and we 1 3
3
3
x2 ex dx =
ex (3x2 dx) =
1 3
1 1 3 eu du = eu + C = ex + C 3 3 (1)
A numbered displayed formula is produced in the equation environment, whereas the displaymath environment produces unnumbered formulas. Arrays are produced by the array environment. It has a single argument that specifies the number of columns and the alignment of items within the columns. Adjacent rows are separated by a \\ command (the newline command) and adjacent items within a row are separated by a & character. For example, $\begin{array}{lr} float & 10.0 \\ int & 10 \end{array}$ 6
produces f loat 10.0 int 10 The first column is left-justified, and the second column is right-justified. Note that because this is done in the maths environment $...$, the text is actually typeset as if it were a collection of mathematical variables. This can be corrected by specifying a font for the text, say {\rm float}, or better still using \textrm{...}, or by using the tabular environment. The tabular environment is useful for making arrays of ordinary text items. For example, the source \begin{center} \begin{tabular}{|c|l|c|} \hline Number of iterations & type & time \\ \hline 100 & random & 5.0 \\ 1000 & random & 120 \\ 100 & genetic & 0.1 \\ 1000 & genetic & 0.5 \\ \hline \end{tabular} \end{center} produces Number of iterations 100 1000 100 1000 type time random 5.0 random 120 genetic 0.1 genetic 0.5
A The input structure to L TEX should display as clearly as possible the logical structure of the document. For this reason, any structure that appears often in your text should be declared using the \def or the \newcommand command. These are usually placed at the beginning of your document.
Some examples are: \def\iff{\Leftrightarrow}
7
allows you to type \iff instead of \Leftrightarrow whenever you want the symbol . Similarly, \newcommand{\bi}{\begin{itemize}} \newcommand{\ei}{\end{itemize}} defines \bi ... \ei to be equivalent to \begin{itemize} ... \end{itemize}.
A Citations to bibliographic entries can be handled automatically in L TEX, along with other types of cross-referencing. To handle cross-referencing, A L TEX requires two passes over the input: one pass to find the information and a second pass to put it into the text.
Here is an example of how cross-referencing to sections and equations works: Equation~\ref{eq:euler} in Section~\ref{sec-early} below is Euler's famous result. \subsection{Early Results} \label{sec-early} Euler's equation \begin{equation} e^{i\pi} + 1 = 0 \label{eq:euler} \end{equation} combines the five most important numbers in mathematics into a single equation. giving Equation 2 in Section 1.2 below is Euler's famous result.
1.2
Early Results
ei + 1 = 0 (2)
Euler's equation combines the five most important numbers in mathematics into a single equation. Bibliographic citations are handled in a similar fashion. The source list is created with a thebibliography environment, which is like the enumerate environment except that: 8
List items are begun with the \bibitem command. Its argument is a key by which the source can be cited with a \cite command. The thebibliography environment has an argument that should be a piece of text the same width or wider than the widest item label in the source list. For example, I can cite items from the reference list to these notes as illustrated in the following example: Leslie Lamport makes a strong argument~\cite{lamport2} for the use of logical word-processing systems, and \LaTeX~\cite{lamport} is a good example of a (semi-) logical system. The document by Ken Wessen~\cite{ken} shows how \LaTeX\/ can be used to produce a thesis. Leslie Lamport makes a strong argument [4] for the use of logical A word-processing systems, and L TEX [2] is a good example of a (semi-) logical system. The document by Ken Wessen [3] shows A how L TEX can be used to produce a thesis. This example uses part of the following bibliography: \begin{thebibliography}{9} \bibitem{knuth} D. E. Knuth. {\em The \TeX~Book.}\/ Addison-Wesley, Reading, Massachusetts, 1984. \bibitem{lamport} L. Lamport. {\em \LaTeX\/: A Document Preparation System}.\/ Addison-Wesley, Reading, Massachusetts, 1986. \bibitem{ken} Ken Wessen. Preparing a thesis using \LaTeX. Private communication, 1994. \bibitem{lamport2} L. Lamport. Document production: Visual or logical. {\em Notices of the Amer. Maths. Soc.},\/ Vol. 34, 1987, pp. 621-624. \end{thebibliography}
A In a later lecture we will cover BIBTEX, L TEX's full bibliographic database environment.
9
2
ispell
A There are many spelling checkers that can be used on L TEX source; the one I use is called ispell. If you use either a -t flag or have a .tex extension on A your file, ispell will ignore most L TEX commands. A typical session might look like:
Marr 0: 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: 11: 12: Barr Carr Mar Marc Mare Mark Marry Mars Mart Marx Mary Mawr Parr
File: image.tex
luminance events that occur infrequently, but which somehow capture a high proportion of the image information. David MARR The online manual gives all the information you could need for ispell.
A For other spelling checkers and word counting, you can detex your L TEXsource using the command detex file.tex.
3
Psfig
Psfig/TEX is a macro package for TEX that facilitates the inclusion of Postscript figures into your documents. The post-processor dvips has full Psfig support, so that Postscript figures are automatically scaled and positioned on the page. To include a Postscript figure in your document, the top of your file should begin with the following declaration: 10
\documentclass[...]{article} \usepackage{psfig} and then, when you want to include an image or figure, use \psfig{figure=input} where input is the name of the PostScript file. Psfig provides options for scaling the height and width of your figure, rotating it by an arbitrary angle, clipping, and many other options. There is a link on the unit webpage to the Psfig User's Guide.
4
Xfig
Xfig is a menu-driven tool that allows you to draw and manipulate objects interactively in an X window. The on-line manual gives you plenty of information on how to use it. In particular, you can export your figures as A encapsulated PostScript files and then include them in your L TEX documents using Psfig.
5
gnuplot
Gnuplot is a command-driven interactive function plotting program. There is an on-line manual entry, but the gnuplot help command will give you all the information you need. In particular, if you have produced a plot you wish to include in a document, you can save the output in Postscript and then use Psfig again. For example, the command gnuplot> plot sin(x); will generate a Gnuplot window containing a plot of the function. If you then type gnuplot> set output "file.ps"; gnuplot> set terminal postscript eps; gnuplot> replot; the same plot will be sent to a Postscript file. You should then type set terminal x11 to return to plotting functions on your terminal rather than sending the results to a file.
11