6:54 AM
\input{tex/introduction}\newpage\cleardoublepage
\input{tex/material}\newpage\cleardoublepage
\input{tex/boring}\newpage\cleardoublepage
\input{tex/discussion}\newpage\cleardoublepage
\input{tex/conclusion}\newpage\cleardoublepage
\input{tex/acknowledgement}\newpage\cleardoublepage
\addtocounter{rom}{1}\setcounter{page}{2}~
\newpage\thispagestyle{plain}\setcounter{page}{3}

7:16 AM

@TorbjørnT. Thanks :-)
I am searching for posts containing the class thesis.cls or Thesis.cls and it is absolutely hopeless.

@Johannes_B You do like to torment yourself.

@TorbjørnT. Yes.
Try Thesis. If that works, replace Thesis by book. — Johannes_B Oct 20 '15 at 11:53

7:54 AM
@TorbjørnT.Thanks for pointing out my stupid mistake...

8:43 AM
@Johannes_B: Another 'Revival' for you ;-)

@ChristianHupfer I can't help it. It just keeps happening.

@Johannes_B When I answer an older question, it's hardly noticed :D

@ChristianHupfer 66/3687

@ChristianHupfer Yes. @egreg has 370. That is ten percent of all the revival badges.

8:49 AM
@Johannes_B I've got only 28 of them. And only 4 Necromancer ... that's the same sad story

@ChristianHupfer See @egreg: enlightened 810/8736; nice answer 1709/20934

9:08 AM
domi is back.

@Johannes_B Really, really, really make sure we start a new right-handed page!

@egreg :-)

9:29 AM
When I try it on a tiny document, xelatex is about twice slower than pdflatex. I get 0.5 s vs 0.25 s. Is there anything I can do to close the performance gap?
I mean a really tiny document, such as
\documentclass{standalone}
\begin{document}
$x^2$
\end{document}

@Szabolcs try with a bigger document as mostly there you are just timing a fixed larger startup time

@DavidCarlisle I know, but I have to run latex on many such tiny documents. So it is indeed the startup time that seems to matter. I tried mylatexformat, but in this situation it has minimal impact. Is there anything one can do about the startup time?

@Szabolcs No

Alternatively, is it possible to process multiple documents (all with the same preamble) in a single run (single startup), even if it is necessary to wait between receiving documents and the output is expecetd immediately after an input is received?

@Szabolcs It's not format reading here, it's the engine itself

9:42 AM
@Johannes_B But maybe another \clearpage would be safer

@Szabolcs In DVI mode you can do that: split the DVI after production
@Szabolcs There was a talk on exactly this as TUG2015

My application is similar in a way to LaTeXiT and I want to make it more responsive.

@Johannes_B: domi and masi ... our heros ;-)

I'm looking at the TUG2015 proceedings.

9:46 AM
@JosephWright Thanks! I found it here but it was password protected.

@Szabolcs You have to be a TUG member to get the latest TUGboat

10:19 AM
how can i draw a diagram like this with tikz

10:59 AM
Watch for your life! @DavidCarlisle recommending one of his packages!
3
the correct markup would be \company\ is ... \company: .. or you could look at xspace package. — David Carlisle 17 mins ago

@egreg not recommending, just mentioning.

11:19 AM
@egreg Run for life rather ;-)

3 hours later…
2:07 PM
I feel dirty.

@Johannes_B What happened?

@ChristianHupfer He used one of his own templates.

@egreg Not quite, but related. :-/

@ChristianHupfer Ah, so that's why @David doesn't use LaTeX.

2:17 PM
@Johannes_B He uses plain TeX and pretends to use LaTeX only. Or MS Word....

2:42 PM
@egreg, @DavidCarlisle: I do believe there might be a disturbance in your reputations. :)

@PauloCereda Not that I see

@egreg soon. :)

@PauloCereda Did one of the Powers appear with a stick in their hand?

@egreg Oh I meant in the good sense. :) You guys are probably rep capped. :)

@PauloCereda Of course I am! Don't know for @DavidCarlisle

2:46 PM
@egreg :)

@PauloCereda What are you going to do for the big feast tomorrow?

@egreg I am finishing my tredicina. :)

@PauloCereda Today's newspaper says over 200000 people are expected to come.

@egreg Oh my! I want to visit Padova some day.

@PauloCereda Waiting for you. :)

2:57 PM
@egreg <3
@egreg: ^^ we had cake. :)
@egreg: I took the photo from a beautiful image of Saint Anthony we have in our parish, then asked a bakery to print it in rice paper. :)

@PauloCereda A yummy cake!

@egreg It is indeed! I ate a couple of slices. :)

Who ate Jesus? The birthday child?
The figure looks baroque, but the background looks 60s.

@MaestroGlanz It's quite complex to explain. The image itself is in front of an altar, which has a nice background, hence the effect. :)

@PauloCereda So baroque painting, oil on shaped wood, isnt it?

3:09 PM
@MaestroGlanz yep, sort of. :)

@ChristianHupfer You'll never trust me tex.stackexchange.com/a/314388/37907

@Johannes_B That's no proof ;-)

3:55 PM
@DavidCarlisle Hope you did not drown at the picnic!

4:27 PM
@JosephWright Oh no!

4:56 PM
@JosephWright no rain at all sun came out whole time!

@DavidCarlisle :-)

@egreg :
@David Carlisle: unfortunately, classicthesisdoes not work together with your solution. Do you know what I should do to use \nameref referring to \part{}\label{} when using classicthesis (texlive 2015)? The relevant code section in classicthesis.sty (lines 600-632) does not look very promising. — Hotschke 1 hour ago
@egreg line 617 seems to fall in the middle of that range ^^^^

@DavidCarlisle Don't use classicthesis?

Nice green dress for the Queen.

@PauloCereda The day-glow one?
@PauloCereda Looked less fluorescent in a picture of it in one of the papers

5:07 PM
@JosephWright Guess so. :) It reminds me of Kermit the frog. :)
@JosephWright ooh

@Johannes_B it seems like that section has very unreliable author

@DavidCarlisle :-D LOL
@DavidCarlisle When the document has to printed in two columns, ... Is that correct?

5:27 PM
@Johannes_B has to be printed or better will be printed (depends on context)

@Johannes_B sorry that was a grammar question? I thought it was a reference to classicthesis, which I didn't understand:-)

@JosephWright Inside interface3.pdf the command \exp_stop_f: appears to be fully expandable (the red star). Shouldn't it be marked as non-expandable?

@DavidCarlisle @yo' It is at the begining of classes.pdf and sounds strange.

@Johannes_B it's ungrammatical in the current form.

@Manuel Hmm, could be: I'll think about it

5:31 PM
@Johannes_B oh I could fix that now (literally building a ctan base zip file now...)
@JosephWright ^^^

@DavidCarlisle :-)

$svn commit -m "to be or not to be" classes.dtx Sending classes.dtx Transmitting file data .done Committing transaction... Committed revision 1185. @Johannes_B fixed thanks. ^^ @JosephWright May be it's very special, since it actually “expands” under f expansion, but in \edef it stays untouched. Not really big problem, just to let you know. @DavidCarlisle :-) Das ist sehr... spooky. :) Und das ist auch... spookier. :) 5:41 PM @PauloCereda I don't suppose it's worth asking why a Portuguese speaking Brazilian is posting German language translations of Mary Poppins to an English language TeX forum? I cannot remember, why i opened classes.pdf. I bet i was looking for some information. @DavidCarlisle I am out of cheese. :) @Johannes_B perhaps the German word for Supercalifragilisticexpialidocious? @DavidCarlisle Exactly! @DavidCarlisle There's not a real German syllable or word in that German word 'Super....' ;-) 5:53 PM @ChristianHupfer but Dick Van Dyke's accent sounds better in German:-) @DavidCarlisle ooh :) @DavidCarlisle I gave advice, Miede did differently. @egreg excuses excuses @ChristianHupfer How about Überkalifragilistikexpialigetisch? :) I wonder how @cfr would say supercalifragilisticexpialidocious in Welsh. :) @DavidCarlisle: Google is amazing. I searched Supercalifragilisticexpialidocious in Welsh and the most promising answer came from a Spanish learning forum where a bloke asked this very question: spanishdict.com/answers/189737/… LOL @DavidCarlisle Why is the allocation for XeTeX char classes still up to 255 in latex.ltx? \def\newXeTeXintercharclass{% \e@alloc\XeTeXcharclass\chardef\xe@alloc@intercharclass\m@ne\@cclv} 6:03 PM @egreg because you are not running LaTeX2e <2016/03/31> patch level 1 @DavidCarlisle Oh, it's in the patch file, thanks. @egreg we don't have patch file as such any more, it's in svn and in a zip file i remade a couple of minutes ago as @Johannes_B complained about a some English grammar:-) Just waiting for an all clear to push to ctan @PauloCereda No, the rest is still nonsense :D @DavidCarlisle I was just looking for the symbolic name of the last character class % Classes allocated 1 to 4094 (or 254 on older xetex) % (In earlier XeLaTeX versions 1, 2 and 3 were pre-set for CJK). % \changes{v2.0g}{2015/06/19}{Use$-1$for first range to get contiguous allocation} % \changes{v2.0q}{2016/04/22}{XeTeX 0.99996 has 4096 char classes not 256} % \begin{macrocode} \ifx\XeTeXcharclass\@undefined \else % \end{macrocode} % \begin{macrocode} \ifdim\the\XeTeXversion\XeTeXrevision\p@>0.99993\p@ \chardef\e@alloc@intercharclass@top=4095 \else \chardef\e@alloc@intercharclass@top=255 @egreg ^^ @Johannes_B svn blame blames Johannes B :-)  4 johannes % When the document has to printed in two columns, we sometimes 6:15 PM @DavidCarlisle I am still trying to figure out why i am looking at that document. @Johannes_B apparently you are the first person ever to do so: svn revision 4 was a while ago (1993). @DavidCarlisle :-) @DavidCarlisle Do I need to load latexrelease for getting \e@alloc@intercharclass@top? @egreg no, as a rule we don't track engine changes with \IncludeInRelease guards as you can't roll forwards an old installation with latexrelease package and change the underlying engine, so the change is just there. But arguably it should be a 2016/06/01 full release rather than a patch level, I sent a note yesterday around the team just asking about that... @egreg JK caught us out and slipped this into the texlive pretest so late it came after the "final updates needed for tl2016" 2016/03/31 latex release. @egreg but now the normal cycle of texlive updates has re-started I think we should push out this update one way or another. Looking at texbook.tex and it is really hard to concentrate on clubs when somebody is trying to communicate with you. 6:41 PM texbook.tex (including corrections from 2013); Looking at line 6361, why does it mention \kern1pt and then use \vskip1pt later? Hello guys ;-) What could be the reason we cannot access a package internal method from a custom class definition? eg. in babel we can access \ifx\bbl@opt@main\@nil [...] but not \bbl@load@language{english} (undefined control sequence) (Just requiring your help, because the new question I'm currently writing is inconsistent to be published as-it and looks silly) @wget ? @Johannes_B well it makes the same output, but you could ask @barbarabeeton if you should get a cheque.... @DavidCarlisle I noticed the same output, it seemed a bit strange though. 6:59 PM @DavidCarlisle related to: http://tex.stackexchange.com/q/268885/66964 i'm trying to call the method defined in the babel internals: \bbl@load@language{english} but the compilation complain the control sequence cannot be found What I don't understand is that I can check babel internals values and this is working: \ifx\bbl@opt@main\@nil \typeout{this is nul} \fi while \bbl@load@language is not :-( It is defined when I try it: \documentclass{article} \usepackage[english]{babel} \makeatletter \show\bbl@load@language \stop Yep outside the declaration i.e. not in the preamble, this is working. But it isn't when I try the same from my own custom class file. Looks like as if the method cannot be called if used in a preamble or something like that. But in LaTeX, aside from the cat code, there isn't any "private" statement that could prevent me from accessing that method. @wget But the above example shows that it is defined in the preamble. indeed. I used the hint from David, and \showbbl@load@language is reported as bbl@load@language=undefined hum... @wget no. I can't guess what your issue is, make an example that shows an undefined command error @wget different babel versions? Babel <3.9r> and hyphenation patterns for 83 language(s) loaded. 7:09 PM 3.9r too. I have a minimal installation. I install everything from tlmgr, so completely in rolling release from ctan. @wget well in that case the code above should make the terminal output: LaTeX2e <2016/03/31> patch level 1 Babel <3.9r> and hyphenation patterns for 83 language(s) loaded. (/home/davidc/texmf/tex/latex/base/article.cls Document Class: article 2014/09/29 v1.4h Standard LaTeX document class (/home/davidc/texmf/tex/latex/base/size10.clo)) (/usr/local/texlive/2016/texmf-dist/tex/generic/babel/babel.sty (/usr/local/texlive/2016/texmf-dist/tex/generic/babel-english/english.ldf (/usr/local/texlive/2016/texmf-dist/tex/generic/babel/babel.def))) > \bbl@load@language=macro: #1->\InputIfFileExists {#1.ldf}{\edef \bbl@loaded {\CurrentOption \ifx \bbl@loa @wget talk of "methods" and "private" statements suggests that you may be writing TeX but you are thinking in Java or some such. TeX has no hiding of any sort, it just has macros and the full definition of any macro is available and can be inspected and changed at run time. @PauloCereda Which did you consider the most promising answer exactly? @DavidCarlisle and @Johannes_B -- karl is now the official tex entomologist. nonetheless, i'm checking this now. in the texbook (print) the \kern1pt' is on p.105, para.4, l.4; the \vskip1pt @DavidCarlisle This is indeed what I have. hum looks like there is a check somewhere in the babel code. I have reread my Tex theory: https://en.wikibooks.org/wiki/LaTeX/Plain_TeX#Plain_TeX_macros and my syntax seems correct. I should be allowed to call that method from my preamble. http://ctan.cs.uu.nl/macros/latex/required/babel/base/babel.dtx Don't know why that method is undefined when no language has been specified. @barbarabeeton yes I know but @ karl wouldn't work so well:-) @wget you get an error from babel in that case don't you? 7:25 PM @DavidCarlisle Nope just my compilation which is failing: ! Undefined control sequence. l.64 \bbl@load@language {english} \documentclass{article} \usepackage{babel} \makeatletter \show\bbl@load@language \stop gives  ! Package babel Error: You haven't specified a language option. ok but in my use case: \documentclass{article} \usepackage[base]{babel} \makeatletter \show\bbl@load@language \stop \bbl@load@language=undefined. l.4 \show\bbl@load@language @DavidCarlisle and @Johannes_B -- karl is now the official tex entomologist. nonetheless, i'm checking this now. in the texbook (print) the \kern1pt is on p.105, para.4, l.4; the \vskip1pt is only in the file texbook.tex. i've check the bug archives, and don't offhand see a report on this. so, @Johannes_B, i will send this on to karl in your name. won't get any attention from dek unti 2021 (i think), so remember you may need to update your address. (you guys type faster than i can look things up, or make corrections.) @barbarabeeton All right, thank you. @wget And again one can see how important it is to show the actual code. With the base option you are loading only a very small part of babel. 7:37 PM Is the CTAN-team aware that the package not found page has a layout error? If not can somebody ping one of them? ctan.org/pkg/siunitxandatest @wget well there you are that's \usepackage[don't load most of babel]{babel} @barbarabeeton arguably it isn't a bug, just a small white lie, the printed book would be the same if the source had the kern it says it has, you only see that a skip was used if you look at the source @Johannes_B I could, but I'd only mail ctan at tug.org, so you could too:-) @DavidCarlisle Maybe they already know? @Johannes_B they tend to fix things they know about:-) @DavidCarlisle Alright, i'll send a mail :-) @Johannes_B ctan at ctan.org (not tug.org) 7:47 PM @DavidCarlisle -- absolutely true, and don may answer exactly that. but this is a somewhat more devious white lie than most, in that there is an explicit contradiction to be found, without explanation, if you look for it. worth putting on the list, i think, but i've asked for karl's analysis before doing so. (we do have until 2020 to think about it. and it's the minimum amount for the check, even if accepted. more symbolic than anything else.) @barbarabeeton unlike the totally real badges and rep points that people fight for here:-) @DavidCarlisle -- yes, but waving that check in front of some people's faces is good bragging rights. (totally boggled one of my nephews. he grabbed, and showed it around to his friends for a year. then gave it back. good nephew.) @barbarabeeton :-) @DavidCarlisle What i want to know, why is a kern better? @Johannes_B usually a skip is better as if the page break happens there you probably don't need extra space between the two lines (a kern is not dropped at a page break) @DavidCarlisle Thanks 8:16 PM Greek numbers and english numbers? Usually i would say that are roman (or rather Roman) numbers along with (indo-) arabic numbers. — Johannes_B 2 hours ago @DavidCarlisle Yes, it's dropped, being discardable @egreg I blame the queen @Johannes_B: I learnt that there are in fact Greek numbers, recently. But in the example of course, it's the \Roman ... of course (at least for the \thepart) @Johannes_B believe @egreg (even if he has broken classicthesis) @DavidCarlisle :-D @ChristianHupfer Now looking at this, wouldn't numeral be better? What's the difference? 8:20 PM @Johannes_B As far as I know the word 'numeral' means the number as word, such as 'one','two', not the symbol, i.e. '1' etc. @DavidCarlisle If added with \vadjust it may not be a legal break point, because it is not necessarily followed by glue (if TeX inserts a penalty between lines. @ChristianHupfer Ah, ok. Nice to know. @Johannes_B Well, I am not sure, actually. At least it has this meaning in German language @egreg yes I think that's what I had in mind kern=not break point, but it is of course discardable if the break happens there anyway 9which is the usuaal case, due to baselineskip) (@Johannes_B) @barbarabeeton :-) 8:23 PM @DavidCarlisle I am tired and confused by now. @ChristianHupfer not really in English at least @DavidCarlisle It has both meanings then? @MarkMcGregor: Well, that's not my job, I am not part of the LaTeX3 team. The interface3.pdf is designed to be more of a reference file than an real user manual. Perhaps there are plans to provide document for expl3 beginners. — Christian Hupfer 11 mins ago @ChristianHupfer There's an Italian bloke here working on one :-) @yo' The 'classicthesis' guy? ;-) @ChristianHupfer that, and 'xparse' and 'eqnarray' :) 8:25 PM My vim screen is split. On the left texbook.tex; on the right tex.web. @Johannes_B really it's an insult to both those files to put them in vim, you should show more respect and use emacs 3 @DavidCarlisle :-) @barbarabeeton What editor does DEK use? @Johannes_B pico ... or ... ed .... or punch card machine ;-) @ChristianHupfer not really. We actually use all three of "numeral", "numeration" and "number" for "num... systems" (such as decimal or binary) (note that the community includes native British speakers who seem to mix the notions too) @yo' Well, that's the difference between languages then 8:31 PM @DavidCarlisle Baselineskip glue comes after possible penalties automatically inserted: here's a two line paragraph (with \clubpenalty=301 and \widowpenalty=501) and \vadjust{\kern1pt} in the first line ...\kern 1.0 ...\penalty 802 ...\glue(\baselineskip) 5.05556 @Johannes_B -- probably emacs. when i was shipped off to stanford to learn tex, that was the only editor in regular use by the people in the tex project. and i can't think of any reason why he might have changed. @barbarabeeton Oh no. Emacs. 2 :-) @Johannes_B -- well, i wouldn't have learned tex if i hadn't learned emacs first. 9:20 PM @cfr None, actually. :) 10:06 PM @barbarabeeton If I'd had to have learnt emacs first, I'd definitely still be using Word! @PauloCereda ;) 10:41 PM is there a way to tell texlive's install-tl script to continue the installation without redownloading what was already downloaded the first time? it takes the configuration from the previous, interrupted run but starts anew, which is somewhat absurd :-/ 11:11 PM The best thing about emacs is CTS. 11:21 PM @PauloCereda And the best thing about vim is that, with some effort, one can get out of it. @egreg Exactly. :) So it's almost 100% safe. :) @PauloCereda Except when you overfill the buffer trying every key combination for convincing it to let you leave free. @egreg ooh @PauloCereda Solution: go to another console and type killall vim && rm$(type -p vim)

@egreg Nope. :)
@egreg alias emacs=ed