TeX, LaTeX and Friends

12:08 AM

@UlrikeFischer Is it my fault if I get
(/usr/local/texlive/2018/texmf-dist/tex/latex/l3packages/l3keys2e/l3keys2e.sty)
! Undefined control sequence.
<argument> \g_xsb_dvisvgm_bool
?

Ulrike Fischer

@marmot to late to test now, also you are not showing the file which contains the code.

marmot

@UlrikeFischer Yes, because it is a very old file that compiled fine until recently. I can try to isolate the culprit but before that I wanted to know if this is a solved problem.

@UlrikeFischer MWE:
\documentclass{beamer}
\usepackage{animate}
\begin{document}
\begin{frame}[t]
\frametitle{Test}
\end{frame}
\end{document}
throws an error when compiled with xelatex (pdflatex and lualatex are fine)

Ulrike Fischer

@marmot no i meant that you are not showing the sty. The error is not in l3keys.

marmot

@UlrikeFischer animate.sty

barbara beeton

2:24 AM

I'm contemplating whether to post a question on meta, but wanted first to see if anyone else has the same experience and finds it irksome. Consider the title of this question. It's very long, and on the questions list it occupies several lines. If I wish to connect to it, if the cursor is positioned between the lines, there's no link; the cursor has to be positioned on one of the words in the title. (cont'd)

(cont'd) Shouldn't the implementor of the site design be skilled enough to make the entire "block" occupied by the title active?

marmot

@barbarabeeton Yes. I wish this was the biggest concern with the design of our web pages.

@UlrikeFischer Please disregard my previous message. I updated my TeXLive system once more. Even though animate was not in the list of updated packages, the problem is solved.

barbara beeton

@marmot -- Oh, indeed, there is lots more that I find sub-standard, but at least this has a concrete example to cite. I suppose the complaint should go on the main meta site and not on the tex.meta.

marmot

@barbarabeeton I agree. And we can't blame @DavidCarlisle for that either. ;-)

barbara beeton

@marmot -- Or me (which dpc has been trying to do more often recently).

marmot

@barbarabeeton That's right. It could be a good idea to bring this up on the main site. Maybe, if they see that there is something they need to fix no matter what, they are willing to fix all the other things which they may not deem necessary.

barbara beeton

2:37 AM

@marmot -- You may be amused (or possibly outraged) by the comment I left with this answer. That is something I probably 8can* be blamed for. (Don't tell David.)

marmot

@barbarabeeton I thought Weierstrass was no longer among us. Still fiddling with unicode, is he?

barbara beeton

@marmot -- Knuth included $\wp$ in the cm fonts, so of course it had to be submitted for addition to Unicode. The UTC was adamant that additions wouldn't be made unless there was a distinct meaning for a symbol, and they said that there was already a code for that "p". I told them (with suitable documentation) that, in that case, the Weierstrass p is definitely lowercase. (If they had been smart enough to specify that the cap P was for the power set, there wouldn't have been a problem.)

marmot

3:02 AM

@barbarabeeton I am really a dummy, especially when it comes to fonts. Yet I am very happy with the default fonts....

barbara beeton

@marmot -- I'm certainly not unhappy with the default fonts. (I rather like Computer Modern.) There are others that I like too, but have learned that sometimes a particular project demands something other than the default. Typesetting a math paper in Fraktur would be inappropriate, but it's wonderful when used for a Rilke poem.

marmot

3:40 AM

@barbarabeeton Yes, if I would not have been tortured with Rilke as a young marmot, I might have a stronger desire to typeset his poems. There is a rather nice Rilke path in Trieste region, I might try typesetting this with TikZ 😉

Faheem Mitha

6:35 AM

@marmot What form does Rilke torture take?

marmot

@FaheemMitha You get forced to memorize poems.

Faheem Mitha

@marmot That sounds unpleasant. Here we got Shakespeare and Dickens.

Though no memorization was required.

marmot

@FaheemMitha Probably it is a good trend that "learning by heart" gets less emphasized in education.

Faheem Mitha

@marmot Did you mean "learning by heart"?

No, probably not.

Though I think that is, or was a phrase.

marmot

@FaheemMitha Yes.

Faheem Mitha

6:40 AM

@marmot Ok.

@barbarabeeton That chap has a point about usage examples.

That's an area in which TeX documentation is often deficient. But they are very diligent about showing the code in the manual. I'm not sure why.

And that's actually a common issue. At least with free software.

@egreg If you have a minute, please take a look at tex.stackexchange.com/q/476645/3406

I would not normally bother you, but you've answered questions that are very similar (if not identical), so it would take you maybe 30 seconds to check if I'm doing something dumb. Thanks in advance.

This is a case where I think I have a solution. I just want to make sure it is sane.

Ulrike Fischer

7:27 AM

@FaheemMitha that doesn't make sense. You can't keep a cake and eat it at the same time.

Faheem Mitha

@UlrikeFischer Sorry, what?

Ulrike Fischer

@FaheemMitha you are redefining boxquoting. This overwrites the old definition,

Faheem Mitha

@UlrikeFischer Yes, I know. The old definition being a copy of quoting.

David Carlisle

7:48 AM

@barbarabeeton oh that's interesting, the site has the natural markup here, <h2><a href=...">long title</a></h2> so the fact that ther is no link between the lines is a built in browser feature not something the site is controlling, I think it is more pronounced as the lineheight (baselineskip) is enlarged for the heading.

Ulrike Fischer

@FaheemMitha but why giving boxquoting a definition at all if you want to overwrite it anyway? simply use \newtcolorbox to define it.

David Carlisle

@barbarabeeton the thing is, David can read.

Harald Hanche-Olsen

@DavidCarlisle In order to get what @barbarabeeton suggests, all you have to do is to bind a suitable piece of javascript to the button down event on the <h2> element. There is plenty of javascript on the site already, so this shouldn't be a big problem.

David Carlisle

@HaraldHanche-Olsen yes I'm just slightly surprised that the default is what it is

Ulrike Fischer

@DavidCarlisle she should have hidden the remark in a manual.

David Carlisle

8:00 AM

@UlrikeFischer are there any tex related manuals?

Paulo Cereda

8:33 AM

@DavidCarlisle ooh

Paulo Cereda

9:04 AM

@Skillmon hi mr. rabbit!

Skillmon

9:32 AM

@PauloCereda Hi, Mr. Duck!

Paulo Cereda

@Skillmon <3

Faheem Mitha

@UlrikeFischer Oh. I didn't know that option existed. I'll take a look. Thanks.

Skillmon

10:15 AM

@PauloCereda proofreading is hard for rabbits!

Curio

10:33 AM

Why can't I find \cong? I'm using \usepackage{amssymb}

Curio

10:53 AM

anyone?

Ulrike Fischer

23

A: Chat's etiquette regarding questions

It's OK to ask on chat but you should not expect an answer (although you may get one anyway). The Q&A site is designed for asynchronous help, you ask a question and it stays visible and can be answered by anyone at any time that they feel convenient. If you ask in chat then although the archive...

Curio

sorry

morbusg

11:22 AM

@Curio if you can, try \usepackage{amsmath} instead

Curio

11:36 AM

@morbusg I have that package too

TeXnician

Just for anyone interested: A few days ago Google released a new Open Source tool to remove comments from a LaTeX file and scale included images: arXiv LaTeX cleaner.

Ulrike Fischer

@Curio works fine for me:

\documentclass{article}
\usepackage{amsmath}

\begin{document}

$A\cong B$

\end{document}

Curio

@UlrikeFischer maybe do I have too many packages?

\usepackage[utf8]{inputenc}
\usepackage{amsmath}
\usepackage{graphicx}
\usepackage{media9}
\usepackage{float}
\usepackage[labelformat=empty]{caption}
\usepackage{graphicx,wrapfig,lipsum}
\usepackage{lscape}
\usepackage{color}
\usepackage{mathabx}
\usepackage{amssymb}
\usepackage{gensymb}
\usepackage[makeroom]{cancel}

Ulrike Fischer

@Curio easy to test. Put your document in some test folder and then start to remove stuff until it works or you end at my document - if this fails too, you can show the log-file.

Harald Hanche-Olsen

@Curio What happens to your \cong then? Does it complain about undefined control sequence, or does the symbol just not show up in the output, or …?

Curio

11:46 AM

ooh I've just noticed that overleaf doesn't give me the hint, but it works fine if I write it

Thanks

egreg

@Curio Remove amssymb (already covered by mathabx) and gensymb (that' quite useless generally). This removes all errors you got. But \cong printed correctly nonetheless even before removing the two packages (and ignoring the error messages).

Curio

Thanks

morbusg

It would be interesting to test how LaTeX cleaner would react to xii.tex :D

oh nevermind, it doesn't do what I thought it would

Faheem Mitha

12:06 PM

@UlrikeFischer I looked at this option in the manual, but I don't see how it would do what I wanted.

Though now that I think more about it, it's possible that basing a box on quoting doesn't really make sense, since tcolorbox has similar builtin functionality. I'm not sure. Opinions welcome.

Ulrike Fischer

@FaheemMitha I have no idea what you want. But \LetLtxMacro\boxquoting\quoting doesn't make any sense, if you overwrite boxquoting in the next line again. If you overwrite it, your overwrite it. A command can only have one definition, not one active and some old versions lurking in the background.

Faheem Mitha

Actually, it looks like it uses listings though.

@UlrikeFischer Oh, I define boxquoting twice? Maybe a typo. Let me check.

@UlrikeFischer The second time it is \endboxquoting, though.

Ulrike Fischer

@FaheemMitha \tcolorboxenvironment{boxquoting} redefines \boxquoting.

Faheem Mitha

@UlrikeFischer Yes, it does. First I made a copy of the quoting environment. Then I modified it with \tcolorboxenvironment.

I thought I explained the context in my question. Let me double check.

I originally modified the quoting environment using \tcolorboxenvironment.

So for some time, I've been using a modified version of the quoting environment.

I want to change that, but I can't call the new version quoting, because it will clash with my previous usage. Though I really should not have called a modified version quoting, anyway.

So I want to create another, different, modified version of quoting, but have to call it something other than quoting.

I don't see of another way to do so, using tcolorbox.

Of course, it's also possible that quoting does not bring much to the table, and I could just use a tcolorbox directly. I'm not sure.

Ulrike Fischer

@FaheemMitha sorry I just realized that I misunderstood what \tcolorboxenvironment is doing.

GermanShepherd

12:21 PM

@egreg I have a query on formatting of pseudocode. Doing it in verbatim doesn't render it well on the document. Do you have any ideas on how I could do it? tex.stackexchange.com/questions/476745/…

Faheem Mitha

@UlrikeFischer It seems counterintuitive that \tcolorboxenvironment does expect a name change by default. Because redefining something in place is, in hindsight, not a good idea.

user0

12:54 PM

@StefanKottwitz Someone at texwelt asked to restore his/her answer to my question on bibliographies and indexes. I remember perceiving that answer as not particularly useful: it dealt only with one half of the matter (which is easy and I know how to do it). So, of course you may do what you wish, but, if you'd ask me, the restored one would be simply distracting, in my subjective opinion. So, feel free to relax and don't bother for quite a long time. I apologize for being direct.

egreg

1:04 PM

@GermanShepherd I'd add nothing to TeXnician's answer

barbara beeton

@TeXnician -- Interesting, but the announcement isn't really specific about what is scrubbed. For example, what about the content between \begin{comment} ... \end{comment} (from the comment package) or the plain approach \iffalse ... \fi?

@DavidCarlisle -- Yes, the code (with an explanation) is given in the answer to my question. But what I find interesting is the assumption by presumably everyone else reading that thread is that anyone participating is intimately familiar with html and css code. Also that the question gained two downvotes with no explanation. Discouraging.

@Curio -- Are you aware that mathabx changes the shape of all symbols? Thus, use of other symbol packages is usually not only futile, but likely to result in errors for multiple definitions using the same name.

David Carlisle

1:22 PM

@morbusg naturally it would leave it untouched, as that's the cleanest most regular, most rectangular tex source.

barbara beeton

@FaheemMitha -- If code isn't shown in a manual, how can one expect a potential user to know how to use the tool? Regarding examples, what is said is true; there are often too few. However, examples often occupy unduly large space, discouraging potential users from reading the manual. A possible approach would be to have a compact mostly verbal, manual, with a separate compilation of examples.

Faheem Mitha

@barbarabeeton I meant implementation code, not example code.

@barbarabeeton mostly verbal?

I don't think you mean an audiobook, so I'm not sure what "verbal" is meant to mean, here.

barbara beeton

@FaheemMitha -- I meant words, not pictures. I did misunderstand the use of "code", thinking it meant the code for using the options; a document that contains just the actual package implementation really isn't a user manual, in my opinion.

Faheem Mitha

@barbarabeeton That seems to be common, though - reproducing chunks of the implementation in the user manual.

I don't understand why this is does, but I'm sure I am missing something.

Sorry, that should have read "why this is done".

Ulrike Fischer

1:45 PM

@FaheemMitha the code is often documentated and that's quite often both interesting and useful. But that is not the user manual, this is for developers.

barbara beeton

@FaheemMitha -- It's a truism that documentation for software is usually the last thing to be addressed (and often is inadequate). The packages with which I'm most familiar (amsmath and the AMS document classes) have two major types of documentation: a user guide and the "annotated" implementation code. The user guide is what is accessed by texdoc, other documentation can be accessed by texdoc -l which offers a wider selection. Both are useful, but are different.

TeXnician

@barbarabeeton If one looks at the source code at Github it seems to only catch the % approach.

barbara beeton

@TeXnician -- Thanks. (I find it rather amazing how many authors use the \iffalse ... \fi approach. The \begin|end{comment} is more understandable. Nonetheless, the amount of "private" comments, and their content, in submitted manuscripts is often appalling.)

David Carlisle

2:14 PM

@FaheemMitha the documented sources are an important part of the doc process, tied to the whole Knuth idea of literate programming. As packages have got bigger it's generally a good idea to have a separate user-manual from the documented sources, but that isn't always done. (compare amsldoc.pdf and amsmath.pdf for example)

@FaheemMitha oh @barbarabeeton just said that:-)

Stefan Kottwitz

3:20 PM

@user0 Well I simply have no time to tinker with the database, as there's no undelete option in the software, so I did not undelete

Faheem Mitha

@DavidCarlisle So partly for historical reasons, then?

It's still a bit confusing/off-putting for readers, the vast majority are just looking for a user manual.

David Carlisle

@FaheemMitha for a typical 1990s package where the code part is 2 or three pages, I don't think it is too bad to have a singe file that has the user part at the front and implementation in a clearly marked section at the end, once the code part is thousands of lines and prints to hundreds of pages then it is not at all useful as a user manual I agree. Do you think this would be better as two separate documents mirror.ox.ac.uk/sites/ctan.org/macros/latex/required/tools/…

Faheem Mitha

3:44 PM

@DavidCarlisle No, clearly not. But you already know that. :-)

Ulrike Fischer

@DavidCarlisle I thought you don't know any tex related manual.

David Carlisle

@UlrikeFischer I heard a rumour that that one has a very high documentation-to-code ratio.

Faheem Mitha

4:03 PM

@DavidCarlisle Last revised 1995. I'd say it's well past time for another revision.

Paulo Cereda

@DavidCarlisle hm

David Carlisle

@FaheemMitha the code is much older than that, that was a later documentation improvement....

Faheem Mitha

@DavidCarlisle Sure. I did say revised.

user0

4:58 PM

@StefanKottwitz Got it, thanks!

mariogarcc

5:17 PM

@DavidCarlisle I ended up doing what you and other people told me, but I at the same time I also figured what was causing most of my issues when compiling (I had a command left open), and now I'm doing a combination of both solutions.

Does anybody know, related to this question: https://tex.stackexchange.com/questions/37472/spacing-before-and-after-with-newtheoremstyle/37475
how to make the spacing after the definition consistent with the text inside the definition? Because the spacing just counts after the head for the theorem (i.e. the definition word + number) and if I have more than 1 line of text, then it doesn't have the spacing I want.

Hm, nevermind, maybe it's just a trick of the light? (reference intended)

manooooh

Hello all!!

@barbarabeeton what is the meaning of "Stet" in the AMS Style Guide? I search on Google and it is disambiguated as a verb ("let it stand (used as an instruction on a printed proof to indicate that a marked alteration should be ignored)") and as a noun ("an instruction to ignore a marked alteration on a printed proof")

For example the AMS says "Stet commas before conditions beginning with text if author’s use is consistent"

What does that sentence mean?

Every time I consult the guide I always have that doubt

Do we have to treat it as a verb or as a noun?

David Carlisle

5:32 PM

@mariogarcc what did we tell you? ( I forget:-)

Paulo Cereda

@DavidCarlisle emacs brainwash?

David Carlisle

@manooooh the usage is not easy to read even for a native English speaker, assumes copy editor's jargon I think, Read it as a verb meaning "leave the author's version unchanged" (I think)

Paulo Cereda

@DavidCarlisle ay up me duck

David Carlisle

@PauloCereda can't chat now I'm working (in emacs of course)

mariogarcc

@DavidCarlisle to not separate the auxilliary files from the .tex files because it was an anti-pattern and it caused complications when building :p

Paulo Cereda

5:36 PM

@DavidCarlisle ooh

manooooh

@DavidCarlisle ohhh

David Carlisle

@mariogarcc ah that:-)

@manooooh important feature you probably don't get with texworks etc,

Paulo Cereda

@DavidCarlisle reminds of Prolog. :)

manooooh

@DavidCarlisle I do not even know what it is for, but I am glad you are always willing to consult doubts even when working :)

Paulo Cereda

@DavidCarlisle hanoi.prolog

manooooh

5:38 PM

@DavidCarlisle ok, that makes sense. Thank you!

David Carlisle

@manooooh google for "towers of hanoi"

manooooh

@DavidCarlisle yes I played that game

@DavidCarlisle ohhh are you coding a tower of hanoi?!

David Carlisle

@manooooh with emacs you don't need to play it, the editor has a built in facility to play it for you..

@manooooh it is a built in emacs function

manooooh

@DavidCarlisle are we in the era of robots?

David Carlisle

@manooooh you can sit and watch the <0> move around the buffer as it solves the puzzle.

manooooh

5:41 PM

@DavidCarlisle does it have any use other than playful?

@DavidCarlisle ohh ^-^

David Carlisle

@manooooh emacs or the hanoi function?

manooooh

@DavidCarlisle what you have shown in the image

David Carlisle

@manooooh No it is entirely pointless.

manooooh

@DavidCarlisle "It is not entirely useless" or "No, it is entirely pointless"?

David Carlisle

@PauloCereda except the useful feature of making @PauloCereda jealous that vim does not have this

@manooooh it has the use of animating a solution to the puzzle (starting from a tower of user specified height) if you consider that a use.

manooooh

5:44 PM

@DavidCarlisle it is a hobby, so...

Paulo Cereda

@DavidCarlisle humpf

Harald Hanche-Olsen

6:36 PM

@manooooh It seems that you have it figured out, but I read it as: “Leave it unchanged, even though it runs counter to this style guide.”

manooooh

@HaraldHanche-Olsen ok :)

barbara beeton

7:54 PM

@manooooh -- As @DavidCarlisle says, it's proofreader's jargon. "Let it stand" = leave it alone, don't change it. The reason for this decision is that, in the case described, the author's usage makes sense and won't be misunderstood. (The goal is consistency. Usage has changed over the years; fifty years ago, this would have been changed, but the changes would ensure (a different) consistency.)

Ulrike Fischer

@barbarabeeton today at last we got out tugboats ;-)

manooooh

@barbarabeeton thanks! It is clear for me. However, what is more preferred? The consistency of the author or the use of the AMS Style Guide?

Because if we let the author only be consistent with what he writes, then there will be no AMS Style Guide

barbara beeton

@UlrikeFischer -- Oh, gosh, that really took a long time! But I'm glad it arrived.

Paulo Cereda

@UlrikeFischer oh me too!

barbara beeton

@manooooh -- It would really be preferable for an author to follow the style guide, but these days, there is also the (publisher's) goal to produce the result as cheaply as possible, so some "corners are cut" as long as the math is understandable and not actually incorrect.

@manooooh -- Actually, the AMS Style Guide was written not for authors (although authors can benefit from it) but for free-lance copyeditors who are commissioned to work on manuscripts submitted for publication when the workload exceeds what the in-house copyeditors can handle.

Ulrike Fischer

8:04 PM

2

@PauloCereda ^^^^ they are glad too ;-)

manooooh

@barbarabeeton it is confusing to me, because on the one hand there is this wonderful guide, but on the other you say "Oh, but if you are consistent with what the author writes there is no problem, forget this guide" ???

Paulo Cereda

@UlrikeFischer ooh

barbara beeton

@manooooh -- As I said, the current practice is for some really desirable cleanup to be ignored to streamline the production process in order to decrease the cost. Copyediting and correcting typeset math is expensive, and always has been. (That's why it's been called "penalty copy" since math started being printed.)

thymaro

8:20 PM

quack

manooooh

@thymaro quack!

thymaro

@clemens any thoughts on tex.stackexchange.com/q/476214 ?

Ulrike Fischer

@thymaro well my thoughts were "no minimal example ..."

thymaro

hmm... I don't think Clemens will see this. Is there a way to conjure him in this chat?

@UlrikeFischer :D that is a common issue

manooooh

@thymaro if you quote using @ and the users is still connected in this room you will see his name on the top of the box message; if not, you will not see it (and it is probably that he never entered here since he seems to be a new user)

thymaro

8:30 PM

@UlrikeFischer I didn't even see it, because I know the problem and I don't know enough LaTeX yet to solve it.

David Carlisle

@thymaro I'm sure that's a faq I'm not sure why you ping Clemens, it's a feature of uppercasing, you would see the same with \ref{zz} or \cite{zz} It must be a duplicate question.

Ulrike Fischer

@thymaro what didn't you see? In any case as David says it should be rather easy to resolve, but I really don't want to spent time building a test file.

thymaro

@manooooh I figured it wouldn't work. :D The Clemens I am trying to lure into this duck pond is Clemens Niederberger. He's not a new user ;)

Harald Hanche-Olsen

@manooooh You can ping people in the chat room even if they are not present. But if they're gone for a long time (I don't know how long), you can't ping them any more. (@manooooh)

thymaro

it's not a big deal. The problem has a workaround I am happy to peruse.

David Carlisle

8:33 PM

6

Q: How to make headers uppercase without breaking references in latex?

I am trying to make section headers uppercase. I tried different packages (titlesec and sectsty) but always get broken references. \documentclass[english]{article} \usepackage[T1]{fontenc} \usepackage[latin9]{inputenc} \makeatletter \usepackage{titlesec} \titleformat{\section}% {\upper...

thymaro

@DavidCarlisle ah that makes sense

@UlrikeFischer that there was no mwe and thus people would be prone to look over the question, if they didn't find it interesting enough.

David Carlisle

@thymaro well given the OP posted and accepted a non-answer, no one is likely to look at it at all now.

Ulrike Fischer

9:14 PM

I wonder what a package does that claims that it puts "real hats on symbols" ;-) Can it do bowler hats?

Sigur

11:47 PM

Hello guys. What do you think about the command $|\det X|$ ? In my opinion there is a space before det. I'm not sure if it is right.

marmot

@Sigur There are no guys here. Only @DavidCarlisle, a marmot whom he forgot to mention, and ducks.

yesterday, by David Carlisle

@manooooh I'm the only person here, everyone else is a duck

Sigur

@marmot, they always appears soon or later.

Transcript for

$TeX - LaTeX$ TeX, LaTeX and Friends