Talk:TeX

From Wikipedia, the free encyclopedia

You have new messages (last change).
Good articles TeX has been listed as a good article under the good-article criteria. If you can improve it further, please do.
If it no longer meets these criteria, you can delist it, or ask for a review.
Tux the penguin This article is within the scope of WikiProject Linux, an attempt to comprehensively cover Linux and related topics on Wikipedia. If you would like to participate, you can edit the article attached to this notice, or visit the project page, where you can join the project and/or contribute to the discussion.


Contents

[edit] Pronunciation of TeX

The pronunciation help is wrong. please cf. http://www.ling.hf.ntnu.no/ipa/full/ipachart_cons_pulm_fbmp3.html TeX derives from Greek technê and that is NOT pronounced with an IPA [x] or ['chi'] but with the fortis palatal fricative - IPA ["c with a diacritical ' at its bottom"]. So TeX is not pronounced like 'loch' or GER 'Bach' but like GER 'mich' or GER 'Technik'.

Knuth writes in the TeXbook: "It's the `ch' sound in Scottish words like loch or German words like ach; it's a Spanish `j' and a Russian `kh'." I understand this as a clear description of [x], not [ç]. Perhaps Knuth belongs to the 99% majority of people who never heard about palatalization in Greek. --EJ 13:50, 24 Jan 2005 (UTC)
Update: I just checked The Details of Modern Greek Phonetics and Phonology. Although it discusses Greek palatalization in depth, it nowhere suggests that consonants are palatalized after a front vowel, or when separated from the following front vowel by another consonant. Indeed, examples of velars in such a position show them unpalatalized: έχω [`εxo], γεγονός [ʝεγon`os], λίγο [ʎ`iγo], γυναίκα [ʝin`εka], προσεκτικά [prosεktik`a], κλείνω [kʎ`ino]. Thus, AFAICS, there is not a single reason why TeX should be pronounced with [ç]. Can someone who is both native Greek and a trained phonetician confirm that τεχνη is pronounced [tεxɲi]? --EJ 15:34, 24 Jan 2005 (UTC)
Surely the question is not how it might be pronounced in Modern Greek (Ancient Greek would in any case have been different), but how it's pronounced in English, on which Knuth is surely something of an authority. By the way I've converted the spelling guide to IPA, which is the Wikipedia standard, as well as being less ugly to look at adn probably less confusing. --rossb 19:56, 26 Feb 2005 (UTC)
I agree. The point was that even if someone disagrees with this viewpoint (such as, apparently, the anonymous first poster), it makes no sense to use an ich-laut.
BTW, I reverted the change you did in the sentence "The name is properly typeset ...", it was wrong. TeX is properly typeset either with a lowered capital "E" (preferred), or normal lowercase "e", but not with lowered lowercase "e". --EJ 13:34, 1 Mar 2005 (UTC)

Just to be clear here, are we talking about the Voiceless uvular fricative or the Voiceless velar fricative? --213.112.81.180 15:56, 15 May 2005 (UTC)

Velar. The question was whether it is velar or palatal. EJ 09:05, 17 May 2005 (UTC)
But German ach-laut certainly can be uvular. Is that the only example in Knuth's list that can be [χ] instead of [x]? --213.112.81.40 14:04, 17 May 2005 (UTC)
Hmm, I see. Russian [x] is definitely not velar. As for Spanish and Scottish (i.e., any of the at least 3 different languages or language varieties which could have been meant by this adjective), I don't know the languages myself, but all descriptions of their phonetics I've seen agree on [x], the uvular fricative [χ] not being even mentioned.
The discussion starts to get funny, I think it is time to use some common sense. There's no indication whatsoever that Knuth is a trained phonetician, and that he is aware of the differences between the various fricatives located near velum. In any case, the description he gives in TeXbook is not intended as an exact phonetical definition, but to give general (mostly North American) public a clue that the name is not to be pronounced [tɛk]. Inferring anything more from the description is of dubious value. --EJ 13:18, 18 May 2005 (UTC)
So I guess the conclusion is that you can use either one, and Knuth didn't make the distinction. --213.112.81.22 18:07, 18 May 2005 (UTC)
You can of course use whatever you want, as usual. It's just that the article should indicate some reasonable default pronunciation. --EJ 10:21, 19 May 2005 (UTC)

This is such a matter of controversy that some wikipedian whose native language includes the right 'ch' sound should record the correct pronunciation and put it on this page. Zaha 21:18, 30 May 2006 (UTC)

[edit] Ilumination of spelling

What do people think of adding a picture that illustrates the "spelling" of TeX, similar to the one at the beginning of the LaTeX article? –Matt 09:56, 13 Jun 2004 (UTC)

I like the lion better :) Pity the thumbnailer makes it so faint... anyone know how to manipulate the original to fix that? --Lupin 10:43, 13 Jun 2004 (UTC)
You could simply ask the copyright holder of the image (Oneiros). It would have been nice if I'd been asked before the inclusion of the image here anyway. :-( --Oneiros 13:57, 2005 Apr 6 (UTC)

[edit] Vertical displacement in name spelling

Please avoid using the TeX logo when you cannot print it properly as in HTML. If you want to have the logo, upload a bitmap. – Torsten Bronger 12:18, 18 Jul 2004 (UTC)

Lovely. But it's not a logo, it's the properly formatted form of its name (comme "iTunes", not "ITunes" or "Itunes" or something else). Reinstated. --James F. (talk) 14:34, 18 Jul 2004 (UTC)
Vertical displacement is not part of orthography. Moreover it looks ugly in many browser configurations and inserts an enlarged line skip. All of this is unnecessary. Finally, in his book "Digital Typography" Knuth himself insists on a thorough positioning of the logo letters according to the current typeface. This is obviously not at all possible in HTML. – Torsten Bronger 08:54, 19 Jul 2004 (UTC)
Moreover, the current form in the article is really weird: The last letter seems to have been typed as a Greek letter, but the first two haven't. Anyway, if no real reason against it is given, I'll switch back to the plain version in the article. – Torsten Bronger 23:56, 20 Jul 2004 (UTC)
I concur with this. Indeed, in chapter 6 of "Digital Typography", Don Knuth insists on the correct typography for the name, and this is not what we get on the article by using a subscript -- so it can hardly be called "properly formatted". Interestingly, when Don writes the name in HTML, he uses "TeX", and does not play with the subscripts (see e.g. [1]). I suggest to follow Torsten Bronger's suggestion (especially given that I have indeed linked the bitmap logo as an image, but did not notice or remember this discussion until now). Opinions ? Schutz 23:59, 15 December 2005 (UTC)
I absolutely disagree. I think the properly typeset logo looks wonderful, even in HTML. Besides, if Knuth (along with Spivak) insists on lowering the E when using TeX, why can't we do it in HTML?
If you really don't like that extra interlinear space, we can keep the plain-text version in the opening paragraph, and attach one of those "wrong name" notices. --Siva 21:33, 18 December 2005 (UTC)
Knuth does not insist on lowering the E when using TeX; he insists on having the logo typeset correctly; the lowering of the E is only part of the story, there is also some backspacing between the letters so that they touch each other (see image at the top of the article, or type \show\TeX in a TeX session to see exactly how the logo is created), and we do not get this with HTML. Personaly, I don't care about the extra interlinear space, but I find the HTML version of the logo quite ugly. However, this is a question of taste; to me, the most important point is that Don Knuth himself does not lower the E in HTML. The "wrong name notice" idea sounds good to me. Schutz 21:55, 18 December 2005 (UTC)
Another thing that I just noticed: the logo is really awful when it is part of a link, such as TEX. This is not so much of a problem on the TeX article itself, but can be seen on other pages that link to it. We need a consistent naming, and I think we should go for the simple TeX form. Schutz 08:25, 21 December 2005 (UTC)
There has not been any discussion on this, so I will keep the first occurence of the "logo" version of TeX, and transform all the others to the plain text version. Also, I will make the change in other articles that mention TeX, especially if they have one of these awful links to this article. Schutz 14:39, 17 March 2006 (UTC)
I suggest that the specialised orthography be left out of the lead for a number of reasons:
  1. It does not scale at all with the font size being scaled as it ought to.
  2. Using the math markup for stylistic purposes unrelated to math is semantically problematic as it's not math.
  3. As someone mentioned above, orthography isn't part of spelling while capital/lowercase is considered part of spelling in English for proper names.
  4. As some other people have pointed out, it is not even a standard way of "spelling" even among authorities on TeX in prestigious TeX publications, so there is even less motive for Wikipedia to make this distinction.
  5. The Manual of Style makes no provisions for using stylistic devices in the leads of articles.
— Saxifrage ✎ 02:23, 8 August 2006 (UTC)
The distinction between name and logo is an appropriate one, and the matter is thus editorial first and typographical only as a by-product. There's a reason that "italic small caps" is an oxymoron. —The preceding unsigned comment was added by 66.166.67.2 (talk • contribs) August 19, 2006.

[edit] Example section concerns

I am not too convinced about the section TeX#TeX examples. Do we really want to write this section as a tutorial, with the "hello" code, the details of commands to type to compile the source, etc ? I prefer the way the MetaPost article does it: provide one relatively complex source code, along with the result it produces when compiled. Opinions? --Schutz 01:04, 22 Jul 2004 (UTC)

I think that's a good idea. – Torsten Bronger 17:31, 22 Jul 2004 (UTC)
Seconded. <disclaimer> I wrote MetaPost. 8-) </disclaimer> Maybe two similar short documents, one plain TeX, one LaTeX with the sample output from one or both of them. There are quite a few short example documents around (e.g. Lamport's sample.tex or others from a TeX distribution or off the web) that could be used for inspiration -- preferably giving a single small page (e.g. A5 paper). Not every feature of TeX/LaTeX needs to be included- just enough to demonstrate the principles and maybe the differences between plain & La. --Andrew Kepert 08:23, 7 Apr 2005 (UTC)
I have had a first go at a LaTeX version. It is A4, but 12pt. See User:AndrewKepert/TeX sample. --Andrew Kepert 10:04, 7 Apr 2005 (UTC)
This looks good to me. We should maybe go for more complicated mathematics typing; for example, the cumulative_distribution_function of the normal distribution is something that always look good as an example: an integral sign, some greek letters, fraction, sqrt, it's all there. And/or maybe a complicated summation ? Schutz 22:52, 19 December 2005 (UTC)

[edit] Licence

To 83.135.160.19: TeX is not in public domain, it is copyrighted by Knuth. Here is the licence notice:

% This program is copyright (C) 1982 by D. E. Knuth; all rights are reserved. 
% Copying of this file is authorized only if (1) you are D. E. Knuth, or if
% (2) you make absolutely no changes to your copy. (The WEB system provides
% for alterations via an auxiliary file; the master file should stay intact.)
% See Appendix H of the WEB manual for hints on how to install this program.
% And see Appendix A of the TRIP manual for details about how to validate it.

% TeX is a trademark of the American Mathematical Society.
% METAFONT is a trademark of Addison-Wesley Publishing Company.

--EJ 10:02, 20 May 2005 (UTC)

I have just edited the licence section of the article, but given everything that Don Knuth has written on the topic, it looks like TeX in indeed in the public domain; please have a look and make any comment if you don't agree with my comments. Schutz 00:16, 15 November 2005 (UTC)
IANAL, but I think that if TeX were truly in public domain, there couldn't be any restriction on its distribution (not even the weak "keep unchanged or rename"), and it couldn't hold a copyright notice, as it would no longer have any legal connection with Knuth. Knuth's statements about "public domain" thus directly contradict both the copyright notice in the source code, and his own words in the fourth paragraph of "The future of TeX and Metafont". My understanding is that TeX is not in public domain de jure, but for most practical purposes, it can be used as if it were de facto public domain, and that Knuth uses the words "public domain" in this second, informal sense. It may be just a terminological confusion, keep in mind that terms like free software were not widely (if at all) used in 1982, thus "public domain" could have been the closest two-word approximation to "copyrighted, but freely usable, and almost freely distributable". Also, I suspect that Knuth simply doesn't care about the exact legal status of TeX.
I agree with you. Just a few details:
  • The "Public Domain" thing does not date from 1982, but from 1990, which is long after the copyright notice was affixed to the file.
  • The weak "keep unchanged or rename", as you write, is guaranteed by trademark rather that copyright anyway.
  • I am planning to see a talk by Don Knuth next week, but indeed I am pretty confident that he would not be interested by a question on this topic ;-)
Schutz 22:54, 15 November 2005 (UTC)
Back to the content of the article. The current licence section is a bit too hairy, and its wording seems to suggest that there is some kind of ongoing controversy, whereas in reality the licence is a non-issue. I'd suggest to strip it down, and resolve the question of "PD or not PD" simply by avoiding the term. -- EJ 21:09, 15 November 2005 (UTC)
Well, I remember hearing about this discussion outside of Wikipedia; while it is certainly not an ongoing controversy (it is an error on my side if the text implies this), it is somewhat confusing (the best proof being this discussion ;-). If we have a full section to talk about the licence, I think it is fair enough to say if you look closely, you will notice this term and that term, which are not consistent and give some details. I think it would be going too far to avoid the term PD, given that it appears in Don Knuth's writings. Schutz 22:54, 15 November 2005 (UTC)
OK, this sounds fair enough. I was a bit afraid that it is tricky to mention PD in the section without making it either confusing, or dangerously close to original research. Now, if you can ask Knuth personally for clarification, that would be most wonderful :) -- EJ 01:04, 22 November 2005 (UTC)
Chances of this happening are extremely slim... I am still going to see his talk tomorrow, but there will be enough other people who will want to see or talk to him without me trying to ask legalese questions ... ;-) Schutz 01:58, 22 November 2005 (UTC)

[edit] Babel

I've added TeX to the Wikipedia:Babel project. Feel free to put it in your babelbox! --Smjg 12:36, 28 October 2005 (UTC)


[edit] Compatible Versions

It seems like fpTeX no longer exists. After some poking about, I found that the author has changed his project to XemTeX, rolling Xemacs and TeX into one package. But I am unsure of the status of the project, as it seems to have hopped between various hosts. --165.155.128.134 14:26, 3 November 2005 (UTC)

Er, that was me - stupid proxies broke my login. --Skomae 14:28, 3 November 2005 (UTC)

[edit] The name and its pronunciation(s)

I think this section should be included into the lead; it is relatively short, very introductive, and we have too many sections in the article already IMHO. Opinion ? Schutz 22:37, 15 November 2005 (UTC)

If too many sections is the problem, what about merging some of the sections? "The name and its pronunciations" and/or "Quality" could be combined with "History", and "Derived works" could be merged with "Compatible tools" (especially since the current division of the latter two is nonsystematic at best; e.g., why is BibTeX "derived work", but teTeX "compatible tool"?) -- EJ 01:20, 22 November 2005 (UTC)
Entirely agree. The derived works, compatible tools and other software links at the end of the article are a bit of a mess at the moment. My next priority is the History section; if noone else looks into these sections, I will try to do it afterwards. Schutz 01:51, 22 November 2005 (UTC)
Section has now been merged with the lead. I have not done (yet) any other merging as suggested by EJ. Schutz 23:25, 14 December 2005 (UTC)
Done in part now. At least, the sections are better structured, even if there is still work to do. Schutz 21:01, 5 January 2006 (UTC)

I removed the comparison to the Swedish sj sound. This phoneme is pronounced in an astonishing variety of different ways; formally, it has many allophones, including the unique [ɧ], which is itself realised in many ways, as well as [ʂ], [ɕ] and occasionally [χ], but rarely [x]. See Swedish language for more information. -Ahruman 11:54, 30 June 2006 (UTC)

[edit] Origin of the name

The article says: A homage to Caltech, where Knuth received his doctorate, the name TeX is intended to be pronounced "tekh". I have read many of Knuth's writings about Tex, as well as quite a few interviews cited in reference on his Wikipedia page, but I have not managed so far to find a reference to this homage. Quite the opposite, actually: in the book Digital Typography, p. 635, Knuth admits that he pronounced the name as teks for the first month or two ! Does someone know of any source for this ? If we do not find anything, I'll remove the first page of this sentence (and merge the section into the lead, as proposed above). Schutz 00:20, 7 December 2005 (UTC)

Done. Schutz 23:25, 14 December 2005 (UTC)

[edit] ConTeXt

I was thinking of removing the examples given for ConTeXt. Firstly, It looks to me like ConTeXt is not that well known; at least, it is far behind LaTeX in terms of notoriety. In addition, it looks to me like the examples given do not add much to the article, especially given the previous LaTeX example. But it may be only a bias on my side, so I prefer to check before removing anything. Of course, the mention of ConTeXt in the "Derived Works" section would stay, with the wikilink. Schutz 00:57, 9 December 2005 (UTC)

Done. Schutz 22:14, 19 December 2005 (UTC)

[edit] Computer Modern

Can anyone provide pointers to back up this statement?

  "Metafont has not been accepted by professional type designers, and fonts produced with it like Computer Modern have been harshly criticized."

I'm aware that some complaints have been lodged against specific aspects of Computer Modern, but I personally find CM very pleasant and easy to read. What, specifically, is wrong with it from a typographical standpoint? Who (by name) has criticized it, and why?

Or, rather, is it the Metafont program itself that has been "harshly criticized"?

If I knew the answers I would supply them. Alas, I only have the questions. -- User:12.65.192.77

I agree 100% with the questions and comments of User:12.65.192.77. This is what I can add:
  • The first part of the statement, as it currently stands, is incorrect; Hermann Zapf, for one, is a professional type designer who has worked in collaboration with Don Knuth to create fonts in Metafont (see AMS Euler for an example), and thus, we can be confident that he has "accepted" the system.
  • However, Don Knuth admits that Metafont has not been used as much as he hoped it would be (similar to the fact that TeX has not been modified as much as he hoped); he indicates something about this on page 612 of Digital Typography (and, I think, at other places in the book, but I don't know the references by heart). In particular, he says that asking an artist to become enough of a mathematician to understand how to write a font with 60 parameters is too much. In that sense, we can say that Metafont has not been accepted (as in: used routinely), but this does not mean that...
  • ... The system itself has been harshly critized: I don't think it is the case; never heard that anyway. And go back to first bullet point for a counter-example of a professional type designer.
  • Computer Modern has been harshly criticized: everyone has fonts that he likes and dislikes; for CM to be considered harshly criticized, it would mean that a specially large number of people have criticized it; again, I have not seen this.
In summary, I think this sentence should be removed, except if very convincing sources are provided; what corresponds to my second bullet point above could probably be added to the Metafont article instead. Schutz 00:02, 12 December 2005 (UTC)
Done. Schutz 22:14, 19 December 2005 (UTC)

[edit] Free software or Freeware?

The article says that TeX "is free and is popular in academia..." It links to Free Software, but the usage of the word makes it sound like free as in no cost. Which Free does it mean? —The preceding unsigned comment was added by 68.121.164.2 (talk • contribs).

Both. --maru (talk) contribs 07:21, 30 March 2006 (UTC)
There is an explanation in the last paragraph of the Software section. It was a separate section before, maybe it should be a bit more prominent. Schutz 07:35, 30 March 2006 (UTC)


[edit] Spurious Reference ?

The last reference, to "Beebe, p. 14" is ridiculous. What is "Beebe"? I'll remove this as well as the sentence it supports (about dollar signs) unless there's an outcry --Storkk 14:11, 29 July 2006 (UTC)

That's obviously "Nelson H.F. Beebe. 25 Years of TeX and METAFONT etc" from the bottom of the references section. Nevertheless, I've replaced it with a direct reference to Knuth. The remaining problem is that the sentence is out of place: this is an article on TeX, not plain TeX. The typesetting engine knows nothing about dollars, and the reformulation "characters with category code 3 delimit mathematics because typesetting mathematics is supposedly expensive" is no more funny. -- EJ 17:32, 29 July 2006 (UTC)
I've never seen the direct reference in the TeXbook, thanks for the ref. The joke is not that important in the article, even though I like seeing it mentioned. More important is the question of knowing if "plain TeX" stuff deserves a separate article; I don't think it does, since TeX and plain TeX are always considered together (although the difference should be clearly mentioned). And Knuth did not write a separate "plain TeX" book. Schutz 12:12, 2 August 2006 (UTC)
I see. When I wrote that this is not an article on plain TeX, it was only meant as an empirical observation of the current state of affairs:
  • There's no plain-specific information in the article, except the examples (and the dollar joke).
  • There's a red link to plain TeX, suggesting that plain stuff is supposed to go there.
Having said that, I actually do think it would be better to have a separate plain TeX article, if/when somebody writes a nontrivial information on it, simply because this article is already rather long.
Speaking of the example, it is quite misleading. It purports to demonstrate the difference between inline and displayed math, but the equations are typeset using the builtin <math> markup, which always renders in display style (and may switch away from TeX entirely in the future). I'll try to fix it; a PNG would be the best option, but I don't know how to make one. -- EJ 17:51, 5 August 2006 (UTC)
You're absolutely right about the examples. I should be able to make the PNGs (I think latex2html creates small images for each equation). Schutz 18:19, 5 August 2006 (UTC)

[edit] teTeX, TeX Live

If I understand the teTeX home page correctly, it is going to be replaced with TeX live. Can anyone add this to the article? -- Felix Wiemann 12:18, 17 August 2006 (UTC)

[edit] About TeX macros

"Description of a third-party preprocessor which has nothing to do with the TeX macro language"?


[edit] TeX macro language (CORRECTED SECTION)

TeX provides an unusual macro language; the definition of a macro not only includes a list of commands but also the syntax of the call. Macros are completely integrated with a full-scale interpreted compile-time language that also guides processing.

TeX's macro level of operation is lexical, and it is implemented by the TeXpp[1], the TeX preprocessor. The preprocessor reads several files, understands and evaluates macro definitions, and later the invocations are replaced by the body of the macro.

Comparing with most widely used lexical preprocessor like Cpp, TeXpp is a kind of hibrid (of lexical and syntactic), and differs slightly, as the body of a macro gets tokenized at definition time, that is, it is not completely raw text. Except from a few very special cases, this gives the same behaviour.

The TeX macro language has been successfully used to extend TeX to, for instance, LaTeX and BibTex [2].

TeX macro language refers to the built-in macro expansion facilities of TeX. It is implemented by TeX itself, not by TeXpp. TeXpp is a (rather unknown) third-party preprocessor which uses a language of its own. This page is about TeX, a description of TeXpp is here off-topic; at best it would belong to its own article, but being an esoteric piece of software, I don't think it passes the notability criteria to warrant creation of such article, a simple external link would be more than enough.
LaTeX is indeed implemented as a package of TeX macros. On the other hand, BibTeX is not a TeX macro package (it is a stand-alone application written in Pascal, or more precisely, WEB), nor does it implement any macro language itself (it interprets a minimalistic programming language to be used in style files, which however has no macro capabilities).
Why do you keep referencing a rather unrelated thesis with only a passing mention of TeX when there are many good real references on the TeX macro language (e.g. the TeXbook, for starters) is escaping me. In any case, pasting literal excerpts from the thesis, as you did, is a copyright violation, and therefore strictly prohibited. -- EJ 14:19, 11 December 2006 (UTC)
Yes, I checked, you correct (thanks!!)... TeXpp is exotic, and "TeX itself" implement macro expansion (and cited source is perhaps wrong or out-of-date about BibTex)... final sugestion to correct the text: -- Krauss 11 December 2006.

TeX provides (...) guides processing. -- same text

TeX's macro level of operation is lexical, but it is a built-in facility of TeX, that make use of syntax interpretation. Comparing with most widely used lexical preprocessor like Cpp, it differs slightly, as the body of a macro gets tokenized at definition time, that is, it is not completely raw text. Except from a few very special cases, this gives the same behaviour. -- new text

The TeX macro language has been successfully (...). -- same text

(add example) -- new

[edit] Sample sugestion

Need some sample on TeX#TeX_macro_language. Please, if you agreee, copy/adapt for article. Adapeted from Brabrand (1998):

TEX macros are defined by the \def construct, or similar variants, followed by the name of the macro preceeded by a backslash. Hereafter follows a sequence of tokens and arguments. The arguments are identified by ciphers preceeded by the token "#", bounding the number of arguments by nine. Finally comes the body of the macro, which is taken to be whatever is written between two balancing brackets. Of course, the body can also contain corresponding argument usages, the treating of which is deferred to invocation-time, as with the Cpp macros, yielding dynamic macro scope.

\def \vector #1[#2..#3]{
$({#1}_{#2},\ldots,{#1}_{#3})$
}
\vector x'[0..n-1] ! $({x'}_{0},nldots,{x'}_{n-1})$

This TeXmacro example (\vector) shows a definition of a macro called vector plus an invocation. —The preceding unsigned comment was added by Krauss (talk • contribs) 16:15, 8 December 2006 (UTC).

Retrieved from "http://en.wikipedia.org../../../t/e/x/Talk%7ETeX_0029.html"