« first day (2056 days earlier)      last day (1926 days later) » 

3:02 AM
@nasser Try loading the luatex85 package to solve your KOMA class problems.
 
 
2 hours later…
5:21 AM
@StrongBad Don't know, but there is an email address to the maintainer in the documentation,
 
 
2 hours later…
7:36 AM
@ChristianHupfer I am so sad tex.stackexchange.com/q/315099/37907
 
8:14 AM
@egreg, one of your more helpful comments:
One of the silliest redefinitions I've seen, you have all my sympathy 😄 — egreg 1 min ago
 
Can I have a sanity check? So, right now I use a chunk of python that takes a CSV, and generates a very complex latex file out of it, which is then \include'd into a more normal latex document. I would like to take the python out of it, and use lualatex to end up with something other people can use through ShareLaTeX or Overleaf. Is this a sane plan?
 
@Oxinabox Certainly doable
@Oxinabox I take it the .csv can't be handled by datatool or pgfplots or ...?
@DavidCarlisle I see your color changes pop up in the LaTeX3 tests (OK, under the contrib part)
 
@JosephWright not surprised, they showed up in the 2e tests as well
 
@DavidCarlisle To be expected there
 
I haven't use datatool, but I can say for sure it is out of scope of pgfplots it is not a table/plot. Infact let me see if I have one one line and I can show you
 
8:21 AM
@DavidCarlisle For L3, it's only (u)pTeX that shows anything
 
@Oxinabox well it depends of course what the python is, most core python could be coded up in lua easily enough but if you are using a thousand lines of scipy logic or something, doing the same in lua might be tricky.
 
@Oxinabox Would be helpful
 
@JosephWright they may yet prove to be too brave, we'll see. Note I broke xcolor :(
 
@DavidCarlisle Why?
@DavidCarlisle Oh, that was as graphics and color might not both be loaded
 
@JosephWright because it turns out it emulates rather than includes color option handling so got undefined command when it loaded the new def
@JosephWright I mailed Uwe Kern last night and suggested xcolor adds the option at some point but it seemed polite to update {dvips,luatex,pdftex}.def with an ifundefined test
 
8:26 AM
So this is a bit out of date, I think I am generating nicer TeX now, maybe. Sorry the CSV doesn't line up with the output, but I hope this gets point across. Noted that after an about 6 pages, it changes from the timetable, to set of descriptions.
CSV: https://github.com/oxinabox/Livecon-Printer/blob/master/quick_con_program_book/con_data/sc2016Program5.csv
LaTeX: https://github.com/oxinabox/Livecon-Printer/blob/master/quick_con_program_book/example_output/example_pretty/sc2015_d02.tex
PDF:https://github.com/oxinabox/Livecon-Printer/blob/master/quick_con_program_book/example_output/example_
 
@Oxinabox where do the lengths come from 2.15cm}(10.75cm, 16.15cm apart from that looks like you could just \input the csv file to tex?
 
@DavidCarlisle Caclulating those lengths is where most of the actual logic in the python code is. It is mostly in github.com/oxinabox/Livecon-Printer/blob/master/…, and the timetable_metric_solver in cell 53 of github.com/oxinabox/Livecon-Printer/blob/master/…
 
@DavidCarlisle It's even sillier than all macro definitions in commath.sty!
 
@JosephWright (and others): what makes a command be only x expandable but not f expandable in terms of expl3. And why can't / don't we avoid it?
 
8:41 AM
@Manuel f-type expansion stops at the first non-expandable token and doesn't continue further, whereas x-type will expand all of the argument (simply 'moving on' past non-expandable tokens)
 
@JosephWright I've recently been thinking about this, and why is it a problem?
 
@Manuel To avoid 'output' stopping expansion, you can shuffle things about. For example, if you look at how \tl_lower_case:n works, it keeps moving the 'output' letters after an 'end' marker such that f-type expansion will continue
 
I mean, isn't it possible to always do a fully expandable version.
 
@Manuel It generally is but at the cost of needing to work harder (you have to keep rearranging tokens)
 
@JosephWright Yes, exactly. That's what I've come up with.
 
8:44 AM
@Manuel Note that LuaTeX offers \expanded, which is expandable and does x-type expansion. One could imagine a LuaTeX-only approach based on expl3 where x-type expansion would be expandable: there would then be very little need for f-type
 
@JosephWright Okey, so it's always “doable” but it is just not important? I don't think it's too much work to add one or two auxiliar macros, by the way :)
 
\expanded was originally intended for pdfTeX 1.50, but that never appeared
 
@Manuel x expansion (traditionally) is done using \edef which makes the construct not expandable, doing it expandably in lua is easy doing it expandably in classical tex is probably possible but possibly not worth it (much more complicated much slower and not clearly more useful in most cases)
 
@JosephWright It does? I just recently asked about something like that.
 
@Manuel The work is not just for you :-) TeX is constantly shuffling tokens so the code is slower
@DavidCarlisle \expanded means TeX expansion is easier too: useful if you are working with tokens of mixed catcode (tricky from the Lua side)
 
8:46 AM
@JosephWright but also means it's not "classical tex" :-)
 
@Manuel Been in there from the start: Hans took the pdfTeX v1.50 code as a starting point, it's just it's never been released
 
@JosephWright I saw some time ago the source of \expanded (I might have looked at the wrong source in MkII may be) and saw `\edef\x{\noexpand#1}\x
 
@DavidCarlisle Fair enough
@Manuel ConTeXt's had a few rearrangements in this area as they had names for things that ended up as primitives (see e.g. \unexpanded, which in the ConTeXt sources is the \protected primititve)
 
@Manuel that would be a (non expandable) fake of the primitive for unextended engines that don't have the real thing
 
@DavidCarlisle Yes, I imagined. I just asked a few days ago, if in LuaTeX couldn't we do x expansion through \directlua hehe
 
8:48 AM
@Manuel actually probably more as joseph suggested since the #1 usage is different in that case
 
@Manuel The point of f-type expansion is it lets us do 'functional like' work but stay expandable
@Manuel Working via Lua is tricky as everything is detokenized and then passed back with (currently) one catcodetable active. That's fine if you know all of the material starts off with the same catcodes but rapidly becomes tricky (I've been working on this lately)
 
Okey, so: f expansion is always doable, but too much hassle for probably unnecessary benefit; and LuaTeX permits expandable x expansion and ConTeXt has already implemented it. Good to know. I wish I could learn ConTeXt.
Another thing that has passed my mind (completely unrelated) is an expandable check to see if the next char is an opening brace (or just any token with that catcode). Wouldn't that make the life much more easier and flexible?
 
@Manuel Not sure I'd say 'always', but 'commonly'
@Manuel Can you let me know your e-mail address? I'm joseph.wight@morningstar2.co.uk
@Manuel Good luck with that
@Manuel What for? There are pretty limited cases where you really need to know this
@Manuel ConTeXt is easy enough to follow, at least at the basic level: \starttext Some things here\stoptext :-)
 
@JosephWright I can send you a mail.
 
@Manuel @DavidCarlisle will tell you I'm sometime suspected of being a ConTeXt agent in the team, as I read quite a bit of their sources
@Manuel Cool
@Manuel @DavidCarlisle may guess why I've asked :-)
 
8:58 AM
@JosephWright I don't know, it's just that many times I've imagined that things would go smoother. Some primitive boolean \ifbracenext and then \def\foo{\futurecheckbrace\baz} \def\baz{\ifbracenext blah \else blih \fi}. Or something like that.
 
@Manuel Like I say, the concept is clear enough (you could just make \futurelet expandable), it's the use cases I'm not sure of
 
@JosephWright It's true that it's just to have more flexibility in the input approach, like \foo{asdf} and \foo\asdf meaning different things expandably. True, that's not LaTeX behaviour :)
@JosephWright But I would like to be flexible as I'm with LaTeX (or even TeX). But there are no example documents to look at the sources (this is what mostly interests me) and there's no documentation.
 
@Manuel Again, LuaTeX allows braces here and with Lua we can write an expandable version of \IfFileExists, so you can have the LaTeX behaviour but keep \input expandable
@Manuel There are some resources but most help is on their mailing list, certainly
 
@JosephWright Yes :(
@JosephWright I've sent you a mail.
 
@DavidCarlisle did my explanation of where the numbers come from clarify?
 
9:18 AM
@Oxinabox yes but also too much to take in in one go (at work, doing something else:-) probably could do it more easily in luatex as you coul djust typeset each of teh boxes once into some node list then use lua to directly position those nodes without having to write out a tex syntax for the positioning and re-typeset
 
How do I select random Exercises with the answers package?
 
@Johannes_B You can't fight any rubbish
 
 
1 hour later…
10:48 AM
log file just posted on main site: (C:\Users\Enrico\AppData\Roaming\MiKTeX\2.9\tex\latex\etoolbox\etoolbox.sty I always knew @egreg was a miktex user really.
 
@DavidCarlisle Sure! I'm so fond of MiKTeX I bought 100 copies of Windoze just to be really safe.
 
anyway I know nothing about biblatex, the error is (it turns out) `! Use of \blx@bbl@verbadd@i doesn't match its definition.
l.482 \verb` does that mean the user has `\verb` in the bibliography?
@DavidCarlisle you are definitely right. Just edited my question with the new errors and the url I am using — Enrico Ribelli 25 mins ago
 
@DavidCarlisle This one is to do with indentation in the bbl file: fixed in the dev version
 
@JosephWright you can answer that then:-)
 
6
Q: Bibliography error: "Use of \blx@bbl@verbadd@i doesn't match its definition. \verb"

SimenCurrently having problems with compiling my Master's thesis after downloading the newest version of TeXStudio and MiKTeX. The error message is as follows: Use of \blx@bbl@verbadd@i doesn't match its definition. \verb The only other place I found a similar question regarding this error message ...

@DavidCarlisle I already did before ^^^ :-)
 
10:52 AM
@JosephWright you have powers to close as dup then
 
11:17 AM
Laughing at you is like drop-kicking a wounded humming bird.

[paulo@cambridge ~] $
What is the problem with my terminal...
 
@PauloCereda Nothing. Offensive fortune cookies are awesome. :D
 
@wilx o.O
 
@PauloCereda don't speak of the latex3 project that way!
 
This channel has the weirdest ping sound.
 
11:34 AM
Recently I got an unexpectedly high number of votes for something that egreg said not to do (and many agreed with egreg's comment): tex.stackexchange.com/questions/315147/…. There seems to be a phenomenon, kind of like looking at accident scenes, whereby the more grotesque the result, the more interest it spurs. So I started checking upvote totals for answers where people said "Don't do this". Some interesting results...
What is the psychology here?
 
@StevenB.Segletes Correct solution to the wrong problem. ;-)
 
@egreg That may be part of it, but I have a lot of dull boring correct answers that garner a vote or two.
@egreg Grotesque sells!
 
@StevenB.Segletes I have lots of answers with one or two votes, and even several accepted answers with zero votes.
 
@egreg. That phenomenon I understand. It is the "don't do this" answers that get high votes that I am trying to decipher: tex.stackexchange.com/questions/135097/…
 
@StevenB.Segletes Taste for kitsch?
 
11:46 AM
@egreg. That is true. Architecture has a comparable phenomenon. Where is that by the way? Here is one I am familiar with: philly.curbed.com/2012/5/7/10373524/…
 
@DavidCarlisle sorry. :)
 
12:39 PM
@StevenB.Segletes It's Pena palace at Sintra, Portugal
@StevenB.Segletes Not to mention the possibly most kitsch building in the world
@StevenB.Segletes I visited the castle in Sintra: the exterior is nothing in comparison to the interiors.
 
If Neuschwanstein is kitsch (and I have been there), it is only thanks to Disney. The Pena Palace, however, looks like a rainbow version of it, especially here traveltoeat.com/wp-content/uploads/2013/01/…
 
@StevenB.Segletes Yes, Pena Palace is incredible! No account of it can fully render the idea.
 
 
1 hour later…
2:01 PM
Quaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaack
 
yo'
@PauloCereda woof
 
yo'
2:24 PM
@StrongBad Yo!
 
@yo' Yo Yo
 
yo'
@StrongBad how do you do?
 
@yo' Academic year is over and summer is here. Time to get work done.
I have been fighting with pgfplots
 
@StrongBad Oooh
@StrongBad See my starred comment as to what I'm fighting with
 
yo'
@JosephWright :)
 
2:26 PM
@yo' Meow
 
yo'
@StrongBad I'm not looking forward to the day when I need plots...
 
@JosephWright Once forced to use Word, I tend not to fight it. I am content with really badly formated documents
 
yo'
@StrongBad don't say that here or people will beat you to death using the TeXBooks
 
@yo' no, only in Word. I use LaTeX so I can write efficiently (and procrastinate making plots) and have nicely formated documents that I can revision control
@JosephWright I just started using siunitx. I really like it. I don't know why I put it off so long.
 
yo'
@StrongBad my version control:
[tohecz@toheshiba tohesis]$ ls -d tohesis-1*
tohesis-151021  tohesis-151026  tohesis-151029-submitted  tohesis-160111-printed-IED  tohesis-160118-printed-binding  tohesis-160121-online
 
2:30 PM
@StrongBad poke me if you try arara. :D
@yo' gotta love your machine name. :D
 
yo'
@PauloCereda no parrots or we don't want the computers repeating what we say :D
 
@StrongBad Not an option if your write a grant proposal and some wrapped figures wreck the text when you upload it to the funders website
@StrongBad :-)
 
yo'
@PauloCereda and have you seen the thesis folder name? :)
 
@yo' LOLOLOL
 
yo'
@PauloCereda I think that then the package name yoin is self-explained :)
 
2:31 PM
@yo' you should look into version control. I find it helpful with reviewer comments. Nice anming.
 
@JosephWright Welcome to my world: I put figures, people complain about lack of theory. I put theory, people complain about lack of figures. I add figures, Word mess with positioning. I add mathematics, Word does not recognize the symbols. Oh joy.
Unless... I add theory inside figures! wooooooooooooo
 
yo'
@StrongBad well, I don't really need version control I think for documents. (It's nice for projects though). The point is, Dropbox is taking care of backups, so this part of VC is covered.
btw, I got one of this Qi mobile charging solutions, and it's supercool :)
 
@JosephWright I have had students make each Word page a full page image to beat plagiarism detection software. You could try that.
 
@StrongBad: add an entry to the duck giveaway, you can win a duck. :)
@JosephWright: you too ^^
 
@PauloCereda I don't need a duck. I need arara 4.0 :)
 
2:34 PM
@StrongBad ouch that hurts. <3
I will try to accelerate the development. :)
 
@PauloCereda Now that you defended (you did defend???) my guess is development time is over.
 
yo'
@StrongBad @Paulo defended? Really?
 
@yo' Is it you who defended?
 
yo'
@StrongBad 4 months ago
 
@StrongBad Not yet. :)
 
2:38 PM
Opps, congratns Yo', Sorry Paulo. You will get there.
 
yo'
@StrongBad thanks
 
@StrongBad <3
 
yo'
@PauloCereda in 10 days ... fingers crossed, pal!
 
@yo' Thank you! #ducksinprague
 
yo'
user image
3
@PauloCereda ^^
 
2:51 PM
@yo' awwwwwwwwwww <3
 
@PauloCereda 1:0 :)
@PauloCereda Do you see that Brazilians can score?
 
 
2 hours later…
5:20 PM
Does anyone have Minion Pro installed on their system who can test a document? \textbf{Foo} is giving me italic bold on TL 2016 luatex
 
@egreg :D
 
\documentclass{article}
%\usepackage{luatex85} % Makes no difference
\usepackage{fontspec}
\setmainfont{Minion Pro}
\begin{document}
\textbf{This is some text} % produces italic bold with luatex
\end{document}
 
@AlanMunn nope works fine from here on both OSX and linux
 
5:36 PM
@ArTourter Weird. I wonder what could be causing it. I'm updating now to get the last set of TL updates. Maybe that will fix it.
 
I had that problem a while back. usually clearing the luaotffonts cache files did the trick
or maybe I am not getting the issue because my cache has not been recreated after the update... let me try...
 
@ArTourter Interestingly, this is what my fonst cache contains:
minionpro-boldcnit.lua
minionpro-boldcnit.luc
minionpro-boldit.lua
minionpro-boldit.luc
minionpro-it.lua
minionpro-it.luc
minionpro-regular.lua
minionpro-regular.luc
So no minionpro-bold which I assume should be there.
 
@ChristianHupfer yes, excellent package that, who wrote it? — David Carlisle 24 mins ago
 
I have just cleared my luatex-cache and recreated the pdf and still got the same results.
 
@DavidCarlisle did it again: 'Promoting' one of his packages ;-)
 
5:42 PM
I have luaotfload-tool version: "2.7" Luaotfload: "v2.7-fix-2" on my system.
 
@ArTourter Same for me. And I just erased the cache, and got the same (bad) result.
@ArTourter Can you check in your fonts cache to see if you have a minionpro-bold.* pair?
 
minionpro-bold.lua
minionpro-bold.luc
minionpro-boldit.lua
minionpro-boldit.luc
minionpro-it.lua
minionpro-it.luc
minionpro-regular.lua
minionpro-regular.luc
 
@ArTourter Ok, clearly that's the problem, but I have no idea why those files aren't being generated, since the font is definitely there.
 
what happens if you take the condensed version out of the fontpath?
also have a look in luaotfload-lookup-cache.lua to see if one of the font declares itself as the bold font
 
@ArTourter Here's what it says. I'm assuming that's the bold font (with the wrong name associated)
 ["Minion Pro/B#b##655360"]={ "MinionPro-BoldCnIt.otf", false },
 ["Minion Pro/B#b##786432"]={ "MinionPro-BoldCnIt.otf", false },
 
5:54 PM
@ChristianHupfer you were doing the promoting
 
@ArTourter How do I generate that list again? Just delete the file? (I moved the condensed bold out of /Library/Fonts, but that didn't change things, presumably because of the cached names.
 
@DavidCarlisle Yes, but you were the one praising it ;-)
 
@ChristianHupfer I know it better!
 
@PauloCereda Éder's great-grandfather was from a small town not far from here: Nove, quite famous for its ceramics.
 
@DavidCarlisle Certainly a package with a lot of ... bugs features ;-)
 
5:59 PM
@AlanMunn I just deleted the whole ~/Library/texlive/2016/texmf-var/luatex-cache and let it rebuild entirely.
 
@ArTourter Ok, so removing the names database and removing the condensed bold version of the font fixes the problem. So do you not have that font in your fonts folder?
 
\documentclass{article}

\usepackage{xcolor}
\begin{document}
Foo

\end{document}
LaTeX2e <2016/03/31> patch level 1
Babel <3.9r> and hyphenation patterns for 83 language(s) loaded.
(/usr/local/texlive/2016/texmf-dist/tex/latex/base/article.cls
Document Class: article 2014/09/29 v1.4h Standard LaTeX document class
(/usr/local/texlive/2016/texmf-dist/tex/latex/base/size10.clo))
(/usr/local/texlive/2016/texmf-dist/tex/latex/xcolor/xcolor.sty
(/usr/local/texlive/2016/texmf-dist/tex/latex/graphics-cfg/color.cfg)
(/usr/local/texlive/2016/texmf-dist/tex/latex/pdftex-def/pdftex.def
A new feature? ;-)
! Undefined control sequence.
l.444 \ifGin@setpagesize
 
@AlanMunn no I don't
 
@ChristianHupfer yes (or at least an old one, I don't get that)
 
I only have the ones that came with photoshop
 
6:04 PM
@ArTourter So I guess this is a bug in how luaotfload associates fonts to names.
 
@DavidCarlisle: I've updated latex and graphicx a few minutes ago ... now... what's the bypass?
 
@ChristianHupfer you need tomorrow's update
 
@AlanMunn hmm actually yes I do have it!
 
@ChristianHupfer The LaTeX team always wants to surprise us!
 
@ChristianHupfer just load color or graphicx before xcolor, or wait until tomorrow
 
6:06 PM
@DavidCarlisle: I don't buy this recent update, it's scratched ;-)
@egreg Feels like Monty Python's Spring surprise ;-)
 
@ChristianHupfer serves you right for not using the one true official color package. I have no sympathy.
 
@ChristianHupfer They plant bugs here and there in order to keep users up-to-date and in full attention.
 
@ArTourter I'm going to move those fonts back into /Library/Fonts to see what happens.
 
@DavidCarlisle ... and no empathy ;-) But switching the load order works, thanks!
@egreg It's like let's tear down the house before it burns down rather ;-)
 
@ArTourter Well putting them back worked until any new recreation of the database. :(
@ArTourter I wonder if it's a problem with the font. What version do you have?
 
6:12 PM
@AlanMunn ["version"]="Version 2.112;PS 2.000;hotconv 1.0.70;makeotf.lib2.5.5900",
 
@ArTourter Mine are older: Version 2.015;PS 002.000;Core 1.0.38;makeotf.lib1.7.9032
 
@egreg Oh don't
 
@egreg anyway it's not a bug it's a changed requirement on a package loading a driver.def file that it sets up the pagesize option. I updated the def files last night so that a package not doing that is silently accepted rather than an error, but that's an an enhanced feature not (or course) a bug fix.
 
@ArTourter I wonder if there's a way to update?
 
@AlanMunn how did you get your version?
 
6:23 PM
@ArTourter They are from Creative Suite 3 (!).
 
I have adobe CC so everything is kept up to date for me
@AlanMunn Ah
 
@ArTourter I basically gave up on Adobe after it became more and more impossible to preserve backwards compatibility and I can't justify the cost of upgrading every year or so. I mainly used it for InDesign (which I still miss and there's no open source alternative that is any good.)
 
@AlanMunn well one of the percs of being university staff... I get student pricing!
 
@ArTourter I do too, but that still isn't so cheap.
 
@AlanMunn I mainly use mine for lightroom and true it is not that cheap, but we do quite a lot of photography in the house so...
 
6:28 PM
@DavidCarlisle @JosephWright with today's update there is an error when pdftex.def is loaded.
\listfiles
\documentclass{article}
\input{pdftex.def}
\begin{document}
\end{document}
Undefined control sequence.
l.444 \ifGin@setpagesize
 *File List*
 article.cls    2014/09/29 v1.4h Standard LaTeX document class
  size10.clo    2014/09/29 v1.4h Standard LaTeX file (size option)
  pdftex.def    2016/06/04 v0.06f Graphics/color for pdfTeX
infwarerr.sty    2016/05/16 v1.4 Providing info/warning/error messages (HO)
 ltxcmds.sty    2016/05/16 v1.23 LaTeX kernel commands for general use (HO)
supp-pdf.mkii
 
@clemens @DavidCarlisle's already addressed this
 
@JosephWright OK :)
 
@clemens too late :-P
 
@ChristianHupfer that's okay :)
 
@clemens scroll up this page a dozen lines;-) (load graphicx before xcolor, or wait until tomorrow)
 
6:30 PM
@DavidCarlisle not urgent for me :)
 
@clemens I thought I've done something stupid in one of my packages, but then I realized: It's the fault of some other blokes ;-)
 
@ChristianHupfer Uwe Kern ?
 
@ArTourter I don't know if the prices vary, but CC is $238 for 12 months. I really don't like software that expires...
 
@DavidCarlisle No, that's not the name I had in mind ;-)
 
@AlanMunn well there are different plans depending on what you want but I see your point.
 
6:45 PM
@ChristianHupfer I was working on some document, updated TL, turned back to the document and the error was there. Easy to find.
Gotta go. :)
 
@ArTourter And that's the 'suck you in' promotional pricing, after one year, "At the end of your offer term, your subscription will be automatically billed at the standard subscription rate, currently at US$359.88"
 
@clemens Well, it was the same proceeding for me too. Rule Number One: Never update TL while developing a package ;-)
 
@ChristianHupfer but we've been developing latex3 since 1990, should we still be running unix tex on sun 3s?
anyone got a reopen vote spare?
1
Q: \typecolon is a \mathbin; it should be \mathrel

Mark McGregorIn unimath-symbols.pdf, the open colon ⦂ (U+02982), called the Z notation type colon, is declared as \mathbin. As far as I know, you use this type colon as demonstrated in the following example. The lines x ⦂ int f ⦂ int → int declare a nullary symbol x of integer type (or integer sort, depend...

 
@DavidCarlisle Done
 
@ChristianHupfer thanks (not sure if I'm going to answer, but I might..)
 
7:03 PM
@DavidCarlisle Get on with it?
@DavidCarlisle I would say xor but ...
 
@JosephWright did the mathclass-14.txt parsing stuff get on to ctan in the end? (for ^^^)
 
@DavidCarlisle Yes: let me remind myself what (if anything) is using it ...
 
@JosephWright i wasn't planning to actually run it just use it as a link to that file, which actually lists it as (F)ence which seems worse than mathbin or mathrel
 
@DavidCarlisle Ah, that's OK then: yes, I included it in the set
palladium:tex-ini-files joseph$ kpsewhich MathClass.txt
/usr/local/texlive/2016/texmf-dist/tex/generic/unicode-data/MathClass.txt
@DavidCarlisle Currently I think the only think reading this 'live' is the non-distributed expl3 format-mode stuff
@DavidCarlisle You didn't give me a clue on what to do with F(ence) so load-unicode-math-classes doesn't do anything
 
7:26 PM
@JosephWright thanks, just trying to work out from this what it is supposed to be but it's not really that specific open-std.org/jtc1/sc22/open/n3187.pdf just making it do what : does .. oh got called away and I see you said the same;-)
 
7:38 PM
@DavidCarlisle -- since i'm the one sort of responsible for getting this into unicode, i went back and looked at the documentation i received from the z folks. the class of this symbol simply isn't specified anywhere i could see. so for a definitive answer, that's who should be asked. i can try to go back to the person who provided the z symbols if that would be helpful.
@DavidCarlisle -- lacking anything definitive in the documentation i was given, i'm likely the one responsible for assigning it the same class as the \colon (which appears to be misguided).
 
user174558
7:52 PM
@JosephWright and @PauloCereda Quack! Quack!
 
@DavidCarlisle I actually meant: Don't update your local TL installation while developing a package.
 
user174558
@DavidCarlisle and @egreg still look as old as ever.
 
@JasperLoy Yellow squares have no right of opinion on these matters. :P
 
yo'
Please, tell me it's not just me and the font really is ugly, so that I can stop ranting about it and get over it:
 
@yo' Very nice indeed!
 
8:06 PM
@yo' -- well, it's certainly not the most beautiful i've ever seen ...
 
user174558
@yo' It is very ugly. I like Georgia.
 
yo'
@barbarabeeton but @Paulo should be happy: the capital U looks like a duck.
@JasperLoy please, no.
 
user174558
I think Georgia is the most beautiful font for English, but too much beauty will kill you.
 
@yo' -- the whole thing looks like an ad for a 1940s airline.
 
yo'
@JasperLoy I don't agree.
@barbarabeeton thanks, I'll forward the message and see what they think :)
 
user174558
8:11 PM
I am still deciding whether to learn pstricks or pgf, LOL.
 
@JasperLoy picture mode rules the world
 
@barbarabeeton making it same as : or \colon would b ethe two obvious choices but it's down as a fence line | which just seems wrong
@ChristianHupfer yes I know, but if you spend three decades on the package when can you update TL?
 
@egreg @DavidCarlisle uses pgf/TikZ with a picture mode driver and usually posts the compiled output.
 
@StefanKottwitz Nah, he used directly MS Paint
 
@egreg I let my kids draw and scan it
 
8:21 PM
@JasperLoy I prefer Baskerville, though not the terrible version Windows comes with
 
user174558
@StefanKottwitz I saw your new LaTeX Cookbook!
 
@egreg vvvvv
 
@JasperLoy Nice to hear! Do you like that approach?
 
user174558
@Canageek I have not used Windows for years. I only install a few Microsoft fonts on Linux using the package ttf-mscorefonts-installer on Debian.
 
8:23 PM
@DavidCarlisle -- you're right. and i've just checked the table again. \colon is identified as a relation; the z \typecolon is a fence. clearly wrong. i'll make a note to fix the table.
 
@JasperLoy No idea which version you'd have then.
 
user174558
@StefanKottwitz All your books are good! Of course, the formatting is determined by the publisher, so it is not so nice. =)
 
@DavidCarlisle See? I was right!
 
yo'
@barbarabeeton just to make things clear: I like the logo, I understand the ideas behind it and I think the outcome is pretty fine. But the font is just terrible.
 
@yo' -- depends on what it's supposed to represent. the little wings remind me of the wings on the helmet on the "mercury" figure that i think was used at one time by a telegraph company. (i.e., "dated".) i think i recognize the essence of the boy scout emblem in the graphical part of the logo.
 
yo'
8:38 PM
@barbarabeeton This is the starting point (unavoidable and unchangeable):
 
@JasperLoy This is the best one I think: stormtype.com/families/baskerville-original
 
9:02 PM
A mosquito bit me, it itces. Have to withstand the itch.
 
yo'
9:41 PM
A quick question: is there a switch to undo \raggedright? Something like \normalalignment?
 
yo'
@DavidCarlisle I know. Anything else?
 
@yo' ragged2e package has \justified or some such if I recall correctly
 
@yo' ragged2e has \justifying
 
yo'
@egreg that's it I think, thanks.
 
9:43 PM
@egreg slower but more accurate I assume
 
@yo' You see, @DavidCarlisle is getting older and doesn't remember things
 
yo'
@egreg older than you?
 
user image
2
@yo' no he's older and greyer than me ^^^
 
@JasperLoy Hi Jasper!
@yo' Quack. :)
@egreg Cool. :) Does he know how to shoot a penalty? :)
 
@PauloCereda Not sure: he has a Brazilian background, you know…
 
9:56 PM
@egreg Exactly. :)
 
 
1 hour later…
11:00 PM
Wait, you've been developing LaTeX3 since the 90s?
So, LaTeX3, which isn't out yet, is younger then some of the people in my graduate program?
Tries to look innocent
 
@Canageek I think you mean older?
 
Dammit, yes, older.
I knew it had been in the works, but didn't know for this long
So should I expect it out before or after I finish my PhD? I'm aiming at less then 2 more years
 
@Canageek Now that's funny. :D
 
@AlanMunn So, more then two years. Got it.
 
@Canageek This is on the verboten list. :) tex.stackexchange.com/q/79861/2693
@Canageek And the slightly more informative version: tex.stackexchange.com/q/953/2693
 
11:10 PM
@AlanMunn Have you considered a Kickstarter or Patreon to get some funding?
 
@Canageek I'm not part of the team, but honestly I don't think the issue really is funding. All of the team have real jobs, and so having funding won't really give them any more time than they already (don't) have to spend on the work. And TeX is not exactly the kind of language that you can outsource to some code monkeys somewhere.
 
@AlanMunn That makes sense. I've heard that getting funding from it often helps convince significant others it is a worthwhile use of time from people in other projects, but don't know if that would help.
 
@Canageek I have a personal testimony on how challenging a project can be. I am planning version 4.0 of arara for at least two years now, and I am not able to find proper time to release it in the wild.
 
@PauloCereda Oh, given how much people know and how skilled you lot are, I'm sure it is a herculean task.
 
11:28 PM
@Canageek :) The LaTeX3 people are doing a great job, as they still have to work on the 2e branch. :)
 
yo'
11:41 PM
@AlanMunn right. And actually, I think many of the people would have moral dilemma whether to make it a real job, even if they were willing to. Also, it's not something permanent, that's for sure.
 
@yo' The "outsourcing" problem is the real one, I think. For example, for many other projects I imagine that some sort of short term funding to CS students in the summer might be workable. But for TeX this is just not possible. And TeX is not a language that you just pick up like the next scripting language to appear.
 
yo'
@AlanMunn well, you can pick it up, but you can't pick it up so that you can develop it ;-)
(I think this is similar to other languages, imagine outsourcing writing the g++11 compiler
 
I also worry about what will happen to the project when the members age: Does it get many new people onboard over time, or is it static?
 
@Canageek Well Joseph and Bruno certainly aren't old. So I think there has been some new blood added to the team.
 
yo'
@Canageek there are new people into the team: Joseph, Bruno, Will, ...
Now I'm off, guys. It's been a long day, and there's more like that to come! (and only 5 hours of sleep, which sucks)
 
11:56 PM
Here is nice comparison between speed of mathJax and KaTex for math rendering on web page. KaTex is really getting along very well, and so much faster now than mathjax. I might give a try soon. intmath.com/cg5/katex-mathjax-comparison.php
 

« first day (2056 days earlier)      last day (1926 days later) »