Journal home page
General information
Submit an item
Download style files
Contact us
logo for The PracTeX Journal TUG logo

From the Editor: In this issue; Next issue: TeX for Editors; A Practical Approach to Graphics

Yuri Robbers

In this issue
Next issue: TeX for Editors
Editorial: A practical approach to graphics

In this issue

This 2007-1 issue marks the beginning of the Journal's third year. It also marks my debut as guest editor-in-chief, an opportunity Lance Carnes gracefully offered me, and for which I am grateful. My role for this issue has been the catalyst in discovering new tricks in LaTeX and HTML as well as in getting to know and working with several inspiring people.

The theme for this issue is "Graphics in TeX". Even though TeX was not originally designed for graphical work, it has acquired a lot of functionality in this area over the years. ConTeXt is especially noteworthy in this respect since it offers seamless integration of TeX and Metafont.

In addition to Metafont many other graphical add-on packages exist that extend the capabilities of the TeX-engine, such as PSTricks and PGF with TikZ. This issue introduces various of these packages and shows some of their many applications.

The papers in this issue are both individual submissions and selections from the presentations at last summer's PracTeX06 conference.

The articles at the top of the table of contents have something to do with the theme, and the ones further down deal with other subjects. Within both sets the first articles are for a general audience and beginning/intermediate users. As you progress towards the bottom of each category there are articles on more advanced topics. If you are an expert user or a beginning/intermediate user who would like a challenge, try reading about and possibly doing the projects described by these authors.

The PracTeX Journal now offers an RSS feed. If you are not familiar with RSS (an easy way to "subscribe" to a web site) see this introduction. The RSS link is an orange "XML" icon on the home page.

Next issue

Theme: TeX for editors
Editor: Paul Blaga

Most people use TeX, LaTeX or ConTeXt mainly for writing their own scientific articles. They don't need skills for fine tuning the visual aspect of their work, because this is what the editor of the journal does. The same goes for the books, if the publisher has an in-house TeX expert. However, sometimes the user needs more knowledge, for instance if he has to edit a volume or has to provide a camera-ready version of his book that has to meet certain standards. This issue of PracTeX is devoted to aspects of LaTeX that may be of help to these users. We have in mind, for instance, packages as geometry, caption, fancyhdr, but also explanations of the way somebody can change the aspects of chapters heads and others of the same kind, without becoming, really, a TeX wizard.

Of course we will also welcome any contributions that do not fit within the theme. Issue 2007-2 is scheduled to appear around May 20th. Please contact me if you have a topic for a "TeX for editors" article or other article. — Paul Blaga


This issue is a big one, with about 15 articles. Many thanks to the authors who wrote and typeset the pieces that appear in this issue.

A large part of the organizational work of assembling this issue was done by production editors Joe Hogg, Lance Carnes, Paul Blaga, Francisco Reinaldo, Will Robertson, and myself. These editors spent many hours coordinating with authors, reviewers, and proofreaders to get each article into finished form. My special thanks to Paul Blaga who has done a lot of work, editing three articles, authoring one, helping out with production editing and getting the job done even if it is sprung on him at extremely short notice.

Many thanks also to the reviewers and proofreaders who checked the articles and sent comments and corrections.

Dave Walden has solved one or two problems I have encountered while keeping the journal's web site on track, even though Dave officially resigned as the Journal's web master in August 2006. Many thanks, Dave.

Editorial: A practical approach to graphics

The truly practical approach to any given job does not necessarily include using the best tool in existence for this job. Sometimes it is far more practical to use a tool you know well rather than investing time in learning a new tool. Conceivably you could even research the various tools in existence before deciding which tool you should learn to use. By then we have definitely left the realm of practical approaches.

This raises the question of when it is worthwhile to learn a new tool and when it is not. To answer that question I would like to explore the world of Graphics in TeX (any flavour) a bit deeper.

Suppose you are asked to give a presentation on a subject rather dear to your heart — let's say the question of whether or not fish sleep. Further suppose that you are a scientist and used to working with LaTeX, but having no prior experience with Microsoft PowerPoint. The logical choice would be to use LaTeX for this presentation, since learning the Beamer class is probably going to be relatively easy, and it is much easier to reuse LaTeX material that one has used before in publications with Beamer than it would be to reuse this same material within PowerPoint.

Creating the presentation turns out to be a breeze, but a nice background image, perhaps with some fish in faded colours that do not interfere with the text, is still lacking. Which tool should you use? If you have no prior experience creating graphics in LaTeX, but you are a little bit of a Photoshop wizard, my recommendation would be: use Photoshop. Don't spend time searching CTAN or the web, printing the PGF and TikZ manuals, learning the graphics language and applying it to this problem, since that is an awful lot of work for creating one picture. In Photoshop this background graphic can probably be produced in fifteen minutes, rather than the many hours needed to familiarize oneself with the intricacies and oddities of a completely new package, even if this package has been specifically designed to work well within a LaTeX environment.

Should you wish to give presentations often, or create a lot of graphics, then learning a new tool might become worthwhile. Suppose this lecture is widely acclaimed and you receive a prestigious grant to write the definitive standard work on sleep in fish. This book is obviously going to be hundreds of pages, and contain hundreds or even thousands of graphs. These graphs are based on work of other scientists, on your own experiments, on computer simulations, and on the output of various programs for statistical analysis. In other words: all these graphs are different. They use different fonts, different scales, different ways of indicating points of interest, etc.

In this case you cannot escape redrawing all graphs in a standardized way. Of course you could refer to your trusty statistical software of choice, but in this case I would not. TeX in most variations has a strong emphasis on separating content and form. Most users consider this to be one of its main advantages. In a document that has hundreds or thousands of graphs, you could truly benefit from this separation.

It is, for example, often considered best to have the fonts in the figures match the text fonts. But what if, near the end of the project, you discover that the publisher requires the text font to be Minion Pro rather than Palatino, because that is what the other volumes in the series use. If you have used statistical software such as R for all graphs, you would need to either redraw all graphs, or end up with a book that uses Minion Pro for the main text and the immensely different Palatino in all figures. If, however, you have made the more auspicious choice of investing some time in learning MetaPost, PSTricks, PGF or almost any other TeX-friendly graphics language, then the change is a matter of changing the font definition for all graphics along with the definition for all text, and Bob's your uncle (as the British so eloquently put it).

Another example, this time from the viewpoint of publishers of self-published authors. Suppose one writes a book, and needs to design a cover. It is possible to use any TeX-friendly language, such as PSTricks, or MetaPost. If you are a ConTeXt user, the "built-in" MetaPost integration called MetaFun would be the obvious choice. If this, however, is a new endeavour, using an already familiar program such as Photoshop might again be a lot easier. But if you intend to publish a whole series of books, with covers that are somewhat standardized, then the benefits of using MetaPost, PSTricks or some such language become obvious.

A new book in the series will have the same size, except perhaps width of the spine. If in the original design one has defined a new length variable, spine width, for this purpose then the change is easily made. Similarly, the title, author and publisher on front cover and spine will at least have the same font at the same size, and some of these are even likely to remain the same. With a cover designed in a TeX-like language, the necessary changes are easily made, and the new cover will match the old cover perfectly in all the necessary aspects while in other aspects it can easily be adapted exactly as you desire.

So far I have pretended that you should only use one tool. In practice most people will use a combination of tools for any given task. You could, for example, design a book cover in PSTricks, but use a graphic that has been created in Photoshop. You could use statistical graphs in a book that have been created in R, but with labels and other bits of text that are being typeset and positioned using AMS-LaTeX. Nevertheless, the principles outlined in this editorial apply equally well in such a hybrid environment. Any project can potentially benefit from some thought on which tools to use. Sometimes learning a new tool will easily be worthwhile, but at other times you would be better advised to stick with what you are already familiar with. Also, even though I myself often learn new things just for the fun of it, in critical projects you should probably refrain from using new tools just for the fun of it. As former US government official Bert Lance and famous proponent of the "Practical Approach" put it: "If it ain't broke, don't fix it!"

Yuri Robbers

Page generated June 9, 2010 ; TUG home page; search; contact webmaster.