« first day (1339 days earlier)      last day (3676 days later) » 

4:38 AM
is there a thmtools-way of changing the \topsep of amsthm's proof env?
\addtotheorem{pre,post}headhook[proof]{\topsep=something} do not seem to do anything
(I can redefine the env, of course...)
5:03 AM
another cute thing would be a way to automagically have a command defined by \DeclarePairedDelimiter use \big when in a displayed equation.
1 hour later…
7:08 AM
@HeikoOberdiek Destroyed
7:30 AM
@willrobertson Hello
(as you are passing)
8:02 AM
Hello Friends
Searching for the selnolig package on Google leads to interesting results: imgur.com/0QnJP9K
@JosephWright Hello :) Sorry I never catch you here. Now I'm going to Portland I'm rather, er, behind in my organisation / work to get done before I leave.
@WillRobertson :-)
@JosephWright I didn't mention it yet but I've been playing more with the idea of dealing with auxiliary data. Catcodes are annoying :)
@WillRobertson Almost certainly
@JosephWright Years ago I remember reading about various TeX-based encoders that took text input and mangled it in various ways (David Carlisle would approve) — I don't think those approaches preserved catcodes, but what I was thinking was that we could store auxiliary data in a format that preserved catcodes in a "safe" way… the things Bruno has been working on help with this
8:09 AM
@WillRobertson You mean something like the regex approach to marking up catcode, I assume
@JosephWright Yeah, something like that. The encoding idea came from texdoc aroundthebend chapter 11. The big question is efficiency and how much data we might be expecting.
@WillRobertson one approach is to ban catcode changes.
@DavidCarlisle Yeah, just have everyone write their documents with \ExplSyntaxOn :)
@DavidCarlisle You're probably right though — if auxiliary data is only ever written with document catcodes you save this particular variety of hassle.
8:24 AM
@WillRobertson Honestly, I think catcodes should not be something to perpetuate. Keep syntax on the character level, allow different catcode regimes for parsing, and maybe use \scantokens.
@StephanLehmke Yes — I wouldn't want to try and preserve arbitrary catcode changes, only those between document syntax and expl3 syntax.
@WillRobertson no but just as aux file is always written assuming it will be read back when @ is a letter (even if it is not at the point the item is written) You can have one or more fixed (possibly named) catcode regimes, I don't think it's necessary to support arbitrary packages making arbitrary characters active (that may mean that all ascii punctuation will need to be active always)
@DavidCarlisle I agree, I don't think it's necessary either. The "edge" cases for me are when you might be taking user input and combining it with expl3 macros — but then I don't know why you'd want to do that anyway. Oops, gotta run.
@WillRobertson @JosephWright yea this is a serious issue. Maybe one could "rewrite" expl3 syntax to user level syntax?
2 hours later…
10:32 AM
I will not upvote this question, it is scratched. :PPaulo Cereda 29 secs ago
@egreg: ^^
Watch @Joseph appear out of nowhere quoting more sentences from the dirty Hungarian phrasebook. :)
@PauloCereda You're mean.
@egreg I'm kidding, I upvoted it, of course. :)
@egreg: there were more sentences, but they would be flagged as inappropriated. :)
@PauloCereda :)
@egreg I blame David, of course. :)
Q: How to split a long table into pages with multicolumn in each page?

MaryamI have a very long table (64 rows) with 14 columns. I split it with lontable package, but it keeps error for multicolumn syntax. Any suggestion please? part of the table is here: And I couldn't figure out how to put "continued from the previous page" in between? \documentclass{article} \usepac...

The number of K's is driving me crazy!
11:07 AM
@PauloCereda One should smuggle one K_{\ell} into the answer and see whether it gets published.
@StephanLehmke ooh I like the plan! :)
11:43 AM
Hello everybody!
@HenriMenke 'ello! :)
I have a short non-TeX-related question, but I'm confident someone can answer it :) I'm currently writing my thesis and I have an appendix which is currently named "Auxillary Calculations". Now I also want to include conversion tables for units and other stuff into it such that the title "Calculations" doesn't quite fit anymore. I'm tempted to name it "Auxillary Stuff" but that sounds awful. Can anybody devise a better name?
(I won't split it up in two appendices, as both would turn out to be only two pages or so)
@HenriMenke I'm not a native, let alone a good speaker at all, but how does complementary material sound to you?
@PauloCereda That's a really nice suggestion. Although the word "Auxillary" has the advantage that it matches "Appendix A" like in the TeXbook :)
@HenriMenke That's very naughty. :)
11:53 AM
@PauloCereda I already have "Appendix B: Basis transformation"
12:15 PM
@PauloCereda :-)
@JosephWright In times like this, we need you. :)
@HenriMenke How about "Auxilia"?
@TorbjørnT. Zapped
@JosephWright I don't like spam! :)
12:40 PM
Spam is annoying, but it makes me laugh.
1:12 PM
@PauloCereda I read on the “International New York Times” I got on the train an article about a football school in São Paulo, specifically for goalkeepers: opening a school for goalies “seemed about as feasible as opening a school for sunbathing in Siberia”. The school is guided by Zetti and it seems to be having success.
1:36 PM
@egreg Ah I heard about it! :) Do you know Zetti? One of the best goalie I've seen playing.
@PauloCereda Not a frangueiro?
@egreg LOL sometimes. :) But he was great. :) fechandoogol.com.br
@PauloCereda That's it!
@egreg :)
@PauloCereda I see he also played for Palmeiras.
1:42 PM
@egreg He did. :) Not a stellar moment, but he did very good. :)
The first sentence tells me, this has to be a german guy: tex.stackexchange.com/a/187467/37907
Location Germany. YEAH!
2:15 PM
What tools are there to convert PDF to EPS on a Mac? I know pdftops -eps (from xpdf suite), which messes up my file. I know pdf2ps and ps2eps which also mess up my file: they recalculate my carefully constructed bounding box and rasterize some text.
What other options do I have?
I do have access to Acrobat, but again: it messes up fonts (original text gone, looks like garbage)
Bloody journal only accepts EPS, as usual ... I really don't want to rasterize everything.
@Szabolcs I tried converting a PDF to PS and back, fonts were preserved.
Hm, it looks like not all of these EPS are messed up. Mac Preview will convert back to PDF and that's when fonts get messed up, I think.
@egreg Usually some of these tools work for me. But sometimes they don't. It depends on the PDF input. It seems today I was particularly unlucky with this PDF. BTW do you know if there's a GhostScript based tool that will convert to EPS (not PS) and preserves the bounding box?
@Szabolcs Also the bounding box is preserved.
@egreg Which tool are you talking about?
@Szabolcs I used pdf2ps
2:26 PM
@egreg What is the syntax to produce an EPS (not PS) and preserve the bounding box? I used pdf2ps fig1.pdf which produced fig1.ps. Then I used ps2eps fig1.ps which produced fig1.eps. But now the bounding box was slightly shrunk, leaving zero space on the sides of the figure.
@Szabolcs What I get from pdf2ps is already EPS, with correct comments.
@egreg You are right. It is indeed an EPS, even though the extension is .ps. And the bounding box is correct.
Unfortunately it rasterizes the fonts. I'll search specifically for that problem.
@Szabolcs ps2eps has the bad habit of shrinking the bounding box, trimming white space. My test file had been produced with pdfcrop -margin 2, and ps2eps gobbled 2bp on each side. I don't get font rasterization, the font resources are in the PS file and, as I said, converting back to PDF shows no rasterization.
@cgnieder: nothing new so far, but there's an announcement: github.com/cereda/arara :)
2:46 PM
@egreg I'm sure it depends on the particular file and the particular fonts. It only rasterizes some of them. But finally it turned out that the EPS generated by Acrobat was good (it was just not displayed well by Mac Preview). So my problem is solved.
3:03 PM
Q: Why lua and not python?

user2751530I mean instead of lualatex. I came across the python package, so I am not asking about python support. What was so special about lua that people spent so much time on that instead of doing something with python (and which would be arguably be more useful)?

Too broad, offtopic.
Is there something one can do when references start turning out ugly like this?
I have 7 of them that look like this with two-column revtex4.
I made the mistake of writing the paper with pdflatex + microtype but now need to switch to latex -> dvi before submission ...
@PauloCereda cool! I like the new logo! :)
@PauloCereda Yes
(Using bibtex BTW.)
@PauloCereda Also misses the things the LuaTeX team have said many times on this: people just see the language but not the full requirements for setting it up, etc.
3:21 PM
@JosephWright Indeed. :(
@cgnieder Best part:
> Reading documentation? that will never catch on. (David Carlisle)
@JosephWright: a good video for the Lua vs. Python question: youtube.com/watch?v=D_JS5WCO31Y :)
@PauloCereda that is so true :)
4:07 PM
@Szabolcs raggedright
@PauloCereda you'll get me sacked
@DavidCarlisle oopsie. :)
@PauloCereda do you understand this comment, the very idea seems shocking to me:
Have been using this for a while now with great success. Just to mention a bug I just hit, for the archives: The \ztag command has to be placed in the last column. E.g. if you just have content in the leftmost column, you still have to do my stuff goes here && \ztag{Foo}. Otherwise, the tag is not set, and I think even equation numbers are messed up. — rainmaker 5 mins ago
@DavidCarlisle A bug? Preposterous. :)
@PauloCereda quite so
@PauloCereda I mean if you apply 31 \expandafter into some arbitrary journal package alignment code that you didn't really read, what could go wrong?
@DavidCarlisle LOL
4:26 PM
Reading documentation? that will never catch on. (sed -n 's/^%% File: longtable.dtx Copyright (C) 1990-2001 \(.*\)$/\1/p' `kpsewhich longtable.sty`)
@Szabolcs Add \raggedright somewhere?
Oh, @DavidCarlisle already said so.
3 hours later…
7:15 PM
@egreg I updated my question :-) for mtpro2.
1 hour later…
8:34 PM
@DavidCarlisle I had to add my customary overlong answer to the question about paragraph breaking. ;-)
@egreg funny when I saw there were 2 answers I expected one from you but it was a self answer from the OP (which I didn't understand) typical of you to try to gain a tick on a question asking about my code:-)
@Brent.Longborough Hi! Aren't you watching USA-Belgium?
@DavidCarlisle I even proposed an improvement to the code!
@egreg Three guesses ;-}
@Brent.Longborough No, no and no.
@egreg so I saw:-)
8:39 PM
@egreg , and I don't think so...
@egreg Brent is a Brazilian at heart, so expect some football passion. :)
@PauloCereda For a real Welsh a ball can't be round.
@egreg ooh! :)
Stop it! @egreg is correct, sometimes.
@PauloCereda Welshmen use oval balls even when playing golf.
8:48 PM
@egreg LOL
9:02 PM
Courtesy of Joel C. Solomon - a challenge ` \textsc{The Interesting Life
of Ragıp Hulûsi Özdem} `
@Brent.Longborough This definitely needs language markup
@egreg Indeed.
@Brent.Longborough They really had a bad idea when they adopted the dotless i.
The rondò of Haydn's trumpet concerto is being played: I love this piece.
@egreg At least they're consistent. Our uppercase I must be oldish-style French: "ze uppercase lettres, zey 'ave no accent"
@Brent.Longborough Well, the story is a bit different. ;-)
9:09 PM
@egreg Really? How did we get here?
The dot on the i is similar to the small sign that Germans were used to draw over their lowercase u's in handwriting, to distinguish them from the n's. The dot was used for avoiding misunderstandings and has remained also in print.
@egreg Thanks. I wasn't aware of that.
@Brent.Longborough The Latin I had no dot, of course.
Turkish should have used i and ï
@egreg Hmm, but surely that was only because of blackletter?
@Brent.Longborough It started long before blackletter: fr.wikipedia.org/wiki/Point_suscrit_dans_l%27alphabet_latin
9:19 PM
@egreg Fascinating!
@Brent.Longborough Russians draw a small line over their т in order to distinguish it from a ш. In handwriting they bear the same relationship as "n" and "u"; the т is written as we write "m".
Just to make things easier, the handwritten п (pe) is written like our "n"; and г (ge) is written like a dotless i.
The и is written like a "u" and д is our "g"
9:35 PM
@egreg I'm not seeing the problem telling a handwritten u from an n!
@JosephWright They use(d) the Sütterlin script: en.wikipedia.org/wiki/Sütterlin
@egreg I was imaging it was not like my handwriting :-)
@egreg I actually did learn how to write in Sütterlin.
@Johannes_B So it's still used?
@egreg Nooo.
@egreg It was a lecture at the university about archives in general. And you need to read Sütterlin/Kurrant to know what had been written down. Being able to wright it was an extra bonus.
9:45 PM
@Johannes_B Sometimes I wish students would write in Sütterlin: it would be easier to read than their script. ;-)
@egreg Since i am in Freiberg, much papers were documents of the Bergamt
@egreg :-D
@egreg Handwriting is old fashioned. You should only accept tex-files.
Seems like somebody should update the wikipedia page of Bergakademie Freiberg.
Drunk studens singing/shouting to youtube.com/watch?v=yOEve_yITYA

« first day (1339 days earlier)      last day (3676 days later) »