« first day (3585 days earlier)      last day (1542 days later) » 

12:02 AM
@samcarter_is_at_topanswers.xyz And to move diagonally, is it possible?
@nbro add a yshift in addition to the xshift.
@TorbjørnT. Is it possible to increase the scale too?
@nbro scale=2
@TorbjørnT. Will this scale all the drawings inside the scope?
@nbro Was about to say, it won't by default change the size of nodes, try adding transform shape for that.
12:14 AM
@AlanMunn -- Regarding the question I asked about comparing two texts to determine the percentage difference, my husband forwarded this article to me: technologyreview.com/2020/08/22/1007539/… . I fear it's hopeless at present.
@TorbjørnT. I am drawing a diagram of convolutiona neural network, but it takes a lot of time, because I need to tweak a lot of parameters. Do you have any suggestion to improve this process?
12:24 AM
@nbro I don't know the prosess you've used, so hard to suggest improvements ...
@TorbjørnT. I just wrote the drawings by looking the results and adjusting, i.e. trial and error
ugly process
I need to speed up the process
e.g. is there a tool that given "\draw[fill=violet,opacity=0.2, draw=black] (4.25, 0) -- (5.25,0) -- (5.25,1) -- (4.25,1) -- (4.25, 0) ", shifts everything to the left, right, top, diagonally, etc
I don't want to use scopre
I want to change the numbers directly
Is there such a script or do I need to code it myself?
I don't have time for this!!
12:44 AM
@barbarabeeton That's a great article. Gary Marcus is a very well known psychologist and AI-skeptic. He's also a great writer.
@nbro Don't know of such script, but if you're drawing many similar rectangles like that I'd probably use a pic (chapter 18 in the manual) or a define a style for a node.
1:09 AM
@AlanMunn -- As a lover of cranberry juice (straight up -- fresh squeezed cranberries, no sugar!) I was amused by the thought that a spoonful of grape juice would render it lethal. Better to try oleandrin, another potion that's been in the news recently. Something to serve to the chloroquine maven.
@barbarabeeton The cranberry is truly one of the great North American fruits.
@AlanMunn -- When asked why I moved from Baltimore to New England, my stock answer is "to be closer to the source of supply'. Of course, now that Wisconsin produces more than Cape Cod, one has to wonder.
@barbarabeeton Michigan produces some, but I think the it's a very small production. It's the third biggest blueberry producer though.
1:24 AM
@AlanMunn -- Blueberries good too, I still remember the year my sister came home from college with two suitcases full of wild blueberries she had picked. Yum!
 
5 hours later…
6:51 AM
@DavidCarlisle took a quick glance at your LearnLaTeX demo with the design, for me the logo is cut off
@Skillmonlikestopanswers.xyz Please log an issue
@JosephWright will do :)
@JosephWright logged it in Jonas' PR
@Skillmonlikestopanswers.xyz Thanks
7:14 AM
How do one see the demo?
@Skillmonlikestopanswers.xyz that's weird not so much the cutting off but the lack of space aftera learn as the whole thing is a single svg davidcarlisle.github.io/ll/assets/images/logo.svg looks like this for me:
@UlrikeFischer I copied to davidcarlisle.github.io/ll
yo'
yo'
@DavidCarlisle you rely on system fonts; it would be better to curvify it.
@DavidCarlisle Is the font in the svg a font or curves?
@UlrikeFischer cmu serif .... (view source) hmm
@yo' not me:-)
maybe we should close tex.stackexchange.com/questions/559763/… as "needs more focus"?
it has become a Hot Network Question, but this is not the kind of question to "showcase" I think?
yo'
yo'
7:30 AM
@DavidCarlisle aha -- you got someone to blame? :)
7:45 AM
@yo' Ulri
@yo' Ulrike, naturally.
@DavidCarlisle I acquired various names this summer, I already found Urile Utrike Urlike and now Ulri ;-)
8:12 AM
Another one from the strange book we discussed last week: \textcolor{blue}{$\mathfrak{Colored~Fraktur~fonts}$}
According to the book, in the classes book and report you can't possibly have numbered \subsubsections...
I more and more realise where all these bad-practices are coming from:
> The direct command for creating a new line is `\newline`. The texts preceded
> by a `\newline` command are printed in a new line. A new line can also be
> created using a line break command (`\linebreak`, `\\`, `\\\\`, or one or more
> blank lines) at the end of the previous line[footnote: LaTeX does not create a
> new line set manually by pressing the `enter` button of the keyboard, rather
> the same can be obtained by inserting preferably `\newline` or `\\` at the end
> of the previous line.]
I should stop skimming over this book...
2
8:42 AM
@Skillmonlikestopanswers.xyz print it and ceremoniously burn it, in sign of respect to the LaTeX $DEITYs. Maybe will earn you karma... ;-)
@Rmano the poor trees which died for the paper...
@Rmano that book is 309 pages long...
Well, you can wait until the next BBQ and give it a respectful role in igniting coals ;-)
...and you can use a sample ;-).
@Rmano nah, igniting with paper results in too much ash.
@Skillmonlikestopanswers.xyz ;-)
@Skillmonlikestopanswers.xyz otherwise... instructables.com/id/Print-on-Toilet-Paper ;-P
@Rmano better use birch bark or flying seeds. My father always cheats with a hot air gun.
8:50 AM
@Skillmonlikestopanswers.xyz yes, you're right. But I am quite used to prepare firewood-only BBQs, so a bit of paper to start it off is normally ok.
@Rmano firewood-only is still best ignited with flying seeds. Oh, or a tampon, those are great. They are prepacked water-proof and if you pick them to pieces they are a great igniter.
I know several scouts (me included) who always have one or two with them while hiking to start the campfire.
9:15 AM
@Skillmonlikestopanswers.xyz hmmm. I have to admit ignorance, I don't know what a tampon is (in this context at least!)
@Rmano I think it's some sort of ear thingy that attempts to isolate noises :)
@PauloCereda well, the ones I know are not exactly put in ears... ;-P But maybe...
@Rmano oopsie
9:54 AM
@Rmano I think we mean the same thing...
@Skillmonlikestopanswers.xyz Ah, really interesting application! I will try it in the next BBQ. Spanish medical insurance covers basic psichiatric attention, so the risk is low ;-P ;-P ;-P
10:09 AM
user image
6
10:33 AM
@Gaussler Funny you asked about box stuff: I'm looking at xbox for other reasons :)
@JosephWright ah I'll try that later (any plans to get xbox on to ctan?)
@DavidCarlisle better ask Microsoft
@PauloCereda that's already been referenced.
@DavidCarlisle oh
@PauloCereda Will you present some games for xbox at the next TUG meeting?
10:59 AM
@samcarter_is_at_topanswers.xyz I actually have a few. :)
@samcarter_is_at_topanswers.xyz that would endanger his Nintendo commission
what is the effect that scale=2 has on the values of \draw?
In other words, how can I manually compute scale 2?
by modifying the values of \draw
for example, how do I manually scale this \draw (9.55, -0.2) -- (9.8, -0.2) -- (11.55, 1.55) -- (11.3, 1.55) -- (9.55, -0.2)?
@samcarter_is_at_topanswers.xyz 35 games installed and 148 in the cloud. :)
@samcarter_is_at_topanswers.xyz I could talk about Shovel Knight and how the TeX community simply buries itself in the sand and pretends nothing is happening in the outside.
@DavidCarlisle Is the commission exclusive? Otherwise sell both and get two times the profit participation?
@PauloCereda Sound like an excellent plan!
@samcarter_is_at_topanswers.xyz :)
11:10 AM
\documentclass{standalone}
\usepackage{tikzlings}

\begin{document}

\begin{tikzpicture}
\moles[shovel]
\end{tikzpicture}

\end{document}
@samcarter_is_at_topanswers.xyz talking about commissions... I talk about Animal Crossing in July. Then in August... kotaku.com/…
@samcarter_is_at_topanswers.xyz ooh Shovel Mole
@samcarter_is_at_topanswers.xyz the funny thing is that there's a boss in that game known as Mole Knight. :)
@samcarter_is_at_topanswers.xyz shovelknight.fandom.com/wiki/Mole_Knight
yo'
yo'
Wow! We've got a user asking the following question: Is there a good explanation to the motivation behind the development of LaTeX that I could read somewhere? I want to get a better understanding of the software system.
3
@PauloCereda You are highly influential :)
yo'
yo'
@barbarabeeton @JosephWright ^^ I wonder if there's a TB article covering that?
@samcarter_is_at_topanswers.xyz I should get an IKEA sponsorship. :)
11:13 AM
@PauloCereda Oh, looks (nearly) like the shovel mole
@samcarter_is_at_topanswers.xyz :)
@PauloCereda A couple of years ago, my sports team had tracksuits with yellow jacket and blue trousers - I heard Ikea jokes the whole year :)
@samcarter_is_at_topanswers.xyz ooh :D
11:25 AM
@DavidCarlisle Not at the moment, it's only really a test of xparse, although it's more-or-less a straight re-implementation is is safe (although look at how I've done framing)
@yo' At Overleaf?
yo'
yo'
@JosephWright yep
@yo' It's a bit of a vague question, not?
yo'
yo'
@JosephWright well, you get what you get :)
@yo' Sure, I just mean I think you'll have to reply with a question trying to pin down what they areally want to know
@yo' sure there must be some interviews with Leslie you could point to if the question is "why did it start", if the question is why it is still developed now, then harder to say:-)
yo'
yo'
11:29 AM
@JosephWright well, my first idea was to point the user to the History section of the TUGboat, only to realize there's no History section of the TUGboat :-)
@yo' Ah, yes, sounds good
yo'
yo'
@DavidCarlisle great, thanks! I also found tug.org/whatis.html for Knuth's TeX
Can you replace the LaTeX2e command \m@th (which is equivalent to \mathsurround=0pt) by \dim_set:Nn \mathsurround { 0pt }, or will something go wrong?
@Gaussler That will work, although we perhaps should have some interface for it
@DavidCarlisle On xbox, just realised I forgot \parbox
@DavidCarlisle Also, would appreciate thoughts on my lrbox approach: probably one for FMi, as it's very hard to set up without relying on internals (so maybe no xparse there)
11:37 AM
@JosephWright \dim_that_ugly_command_previously_known_as_m_at_th:? ;-)
After getting used to LaTeX3, I have developed @-phobia
@Gaussler Probably something like \math_no_surround:n { <tokens> }
@Gaussler The bigger issue is likely that \m@th almost always means something is wrong: semantically, we should try to avoid math mode for non-math content
yo'
yo'
Btw, I also found this deeper talk from Rio that I find highly introspective and entertaining:
@Gaussler E.g. for \textsuperscript, I'd like to use the ConTeXt approach and not have any math mode switching
12:18 PM
What is the right way to do \DeclareDocumentCommand\mycmd { e{^_} } { ... } within expl3 syntax? Substituting _ by \sb or \c_underscore_str did not have the desired effect. The only thing that worked for me was to turn Expl syntax off
@Gaussler \use:x{\DeclareDocumentCommand\exp_not:N\mycmd{e{^\char_generate:nn{`\_}{8}}}}
@JosephWright OMG
@Gaussler _ is awkward, there's no way around it
@Gaussler You could also use a 'factory'
Factory?
\cs_set_protected:Npn \factory #1
  {
    \DeclareDocumentCommand \foo { e{^#1} }
  }
\ExplSyntaxOff
\factory{_}{Code here}
12:22 PM
@JosephWright why doing this, if you use \ExplSyntaxOff?
@UlrikeFischer I was about to extend ;)
@UlrikeFischer My point was we can arrange to grab a token of the right catcode, it;s just exactly how, there are several ways
@UlrikeFischer One might of course have the code in the set up, where it might need expl syntax
@DavidCarlisle, @UlrikeFischer Who is best placed to answer to question on LaTeX-L? I was thinking Bruno on trees, I could say a little about coffin handles but really it should be David/Frank/Chris
12:48 PM
`\documentclass{article}

\usepackage{xparse}

\DeclareDocumentCommand\cmdone { o }
{
\IfValueTF { #1 } { true } { false }
}

\begin{document}

$ \cmdone \neq \cmdone[1] $

\end{document}`
I get true \neq true
I don’t know when this started happening
@Gaussler Don't add spaces: \IfValueTF {#1}. You are outside expl3 syntax.
@JosephWright will look later (but I spent the weekend sorting mediabox and pagecolor settings in all engines ...)
@DavidCarlisle, @Gaussler On \@begin@tempboxa, in expl3 terms all boxes contain a colour group, so it really does come down to the 'making document commands' business
@DavidCarlisle, @Gaussler Ironically, I notice I might need to extend the version in xbox as it doesn't cover \parbox (which I now remember why I skipped)
@DavidCarlisle Do you think xbox is actually useful? I've a few discussion points from it, that perhaps are, but I'm nots sure it adds much (we'd need to agree on my \fbox changes: there is a difference there)
@UlrikeFischer Sounds fun!
1:03 PM
@JosephWright not sure ;-). But now I can tell you which setting will win in xelatex if you change \pdfpagewidth and use a \special and try add to a mediabox entry to the page object ...
@UlrikeFischer :)
@JosephWright and I think we should not make page color backend independant. Better only correct the pdfmode code (and perhaps a bit dvips) and leave the rest as it is now.
@UlrikeFischer If that's the best plan, sure
1:24 PM
@UlrikeFischer Just imported the new fontloader, it seems to work fine. Supposedly this version should add support for variable fonts in generic, but I didn't test that yet. (It might need support from the feature parsing code)
@MarcelKrüger Oh good. Did you imported also in lualibs?
@UlrikeFischer No.
@MarcelKrüger I will look. We should try to get the unicode-math tests working again. And remove a few branches if they are not longer needed.
@UlrikeFischer I might have been overly optimistic. One of the latex-TU tests for arabic kerning inside of haligns fails. (A kern is applied at the wrong side of a glyph?!?
1:44 PM
@JosephWright not seen it yet, will look...
@UlrikeFischer Frank says you fixed amsmath, did you check that in, I don't see any change?
@JosephWright oh the coffin one from last week
@DavidCarlisle I only made a pull request. And I found because the unicode-math tests break -- they use \shipout\box12 ...
@UlrikeFischer box 12 why 12?
@DavidCarlisle I have no idea, but the code (we stole this from unicode-math) has
\newcommand\SHIPOUT[2][12]{%
  \OMIT
  \SETBOX{#1}{#2}%
  \TIMO
  \shipout\box#1\relax
  \OMIT
}
@DavidCarlisle Yup
@UlrikeFischer oh it's \@tempboxa (unless someone like @PhelypeOleinik is brave enough to re-order latex.ltx) but why access it by number
1:50 PM
@DavidCarlisle Ask Will ;-). But perhaps it works now with lthooks2 as Frank corrected the shipout code.
@MarcelKrüger with the newest lthooks2 code the unicode-math tests behaves again, if one adds filehook-ltx to the preamble.tex
2:40 PM
@MarcelKrüger Imho the additional glue is not from the fontloader.
@UlrikeFischer Which additional glue? I was referring to arabkernsfs. There the glue doesn't change, but a kern moves around.
I am pretty sure that this is caused by the fontloader because it doesn't happen if I load the old one.
@MarcelKrüger I meant a kern, and I get it in the build folder when I compile with lualatex-dev but not with lualatex. The luaotfload files should be the same then.
Zoom is down!
bwawawawawawawa
@AlanMunn ooh logic
More island secrets
2:56 PM
@UlrikeFischer That's odd, but I found the issue in the fontloader and fixed it by adjusting a call.
@samcarter_is_at_topanswers.xyz more updates! mynintendonews.com/2020/08/23/…
@PauloCereda Must be @DavidCarlisle playing it
@samcarter_is_at_topanswers.xyz and the Queen :)
@PauloCereda What else was she supposed to do during lockdown :)
@samcarter_is_at_topanswers.xyz ooh the Falklands indeed!
3:10 PM
@AlanMunn is that how you run your tutor sessions?
@DavidCarlisle It's highly effective.
@AlanMunn only if you have good students capable of logical deduction
@DavidCarlisle On the other hand, teaching the logical if is actually really hard.
@samcarter_is_at_topanswers.xyz vvvv
 Rise and shine, Your Majesty!

7.30am

The Queen starts her day bright and early, with a pot of freshly brewed Earl Grey tea (milk, no sugar), poured from silver pots into a bone china cup and a few biscuits before her breakfast. She tunes in to BBC Radio 4's Today programme while her maid runs her morning bath. In her dressing room, she puts on her first outfit of the day, before her hairdresser arranges her hair.

8.30am

Breakfast with the Duke of Edinburgh is served by a tail-coated footman in the Queen’s private first-floor dining room, which overlooks the beautiful Palace garden. The
> If for any reason the Queen still isn’t fully awake and alert, she will be after a bout of bagpipes, played by Her Majesty’s kilted piper - a traditional start to every weekday when she is at Buckingham Palace, Windsor, Holyroodhouse Palace or Balmoral Castle.
HOLY COW
What's a high tea?
@PauloCereda mid-afternoon snack for posh people
yo'
yo'
3:16 PM
@PauloCereda a better afternoon tea
@DavidCarlisle @yo' ooh
> Any dropped crumbs are gobbled up by her corgis.
Very economical - no need to buy a vacuum cleaner!
@samcarter_is_at_topanswers.xyz ooh efficiency
@PauloCereda I wonder if the corgis prefer the crumbs from tea or the toast from breakfast ...
@samcarter_is_at_topanswers.xyz ooh or both
3:22 PM
@nbro It multiplies the coordinates by 2 ...
@MarcelKrüger I get an additional kern from the patched pgf file I'm using.
@UlrikeFischer it happened. Someone wants to add a watermark to standalone... And by the way, using \AddToHook{shipout/background}{\put(0,0){TEXT}} does somewhat break standalone in pdflatex-dev. It outputs two pages instead of one.
And I can't reproduce it...
3:40 PM
@Skillmonlikestopanswers.xyz can't the someone offer some code? Then I could check if it happens also with the newest version.
@UlrikeFischer The someone doesn't use latex-dev, I did, so no, he can't, I can't :)
@UlrikeFischer I can now reproduce it. It's some strange behaviour of standalone I guess. Compiling an empty document once generates multiple pages, adding contents then creates two pages on the first run, and only one page on consecutive runs.
Anyways, the hook doesn't seem to work (or I'm doing something horribly wrong):
\documentclass{standalone}

\usepackage{xcolor}

\AddToHook{shipout/background}{\put(-15,-15){\textcolor{red}{\rule{30in}{30in}}}}

\setlength\unitlength{1in}

\begin{document}
Hello world
\end{document}
Doesn't show any rule.
@Skillmonlikestopanswers.xyz Oops
@DavidCarlisle I've been looking at \parbox: feels like a coffin to me :)
@Skillmonlikestopanswers.xyz you should set the unitlength inside the picture.
\AddToHook{shipout/background}{\setlength\unitlength{1in}\put(-15,-15){\textcolor{red}{\rule{30in}{30in}}}}
@Skillmonlikestopanswers.xyz yes can happen. latex creates now sometimes a temporary page for the lastpage shipout code.
@UlrikeFischer ah, ok (I remember reading something along those lines a few weeks ago). So now compiling once is never safe and I have to compile everything at least twice?
Anyone know where the second and third optional arguments to \parbox are documented? They are not in Lamport
3:50 PM
@JosephWright I just know roughly what they do...
@Skillmonlikestopanswers.xyz 'Annoy Joseph by not being documented'?
@JosephWright source2e? p. 376
@JosephWright perhaps source2e
@AlanMunn snap, you beat me to it.
@AlanMunn Well ... sort-of
% \DescribeMacro\parbox
%  |\parbox[|\meta{pos}|][|\meta{height}|][|\meta{inner-pos}|]{|^^A
%                    \meta{width}|}{|\meta{text}|}| :
% Makes a box with |\hsize| \meta{width}, positioned by \meta{pos} as
% follows:
%              c : |\vcenter| (placed in |$...$| if not in math mode)
%              b : |\vbox|
%              t : |\vtop|
%         default value is c.
@AlanMunn, @Skillmonlikestopanswers.xyz You see the issue? They are mentioned in the line telling us the signature of \parbox, but not then described. So there is only the code itself.
@JosephWright BSD? The code is documentation enough!
3:54 PM
@Skillmonlikestopanswers.xyz They are not in TL2 either
@JosephWright True, it's not really user level doc at all. I don't have my copy of the LaTeX Companion. I wonder if it's documented there
@AlanMunn Nope, I've just checked
@AlanMunn One for the team list
@JosephWright I always found this giss.nasa.gov/tools/latex/ltx-294.html a good quick resource, but it doesn't even mention the other arguments.
@JosephWright it is in the latex companion.
@AlanMunn Indeed, whilst something like en.wikibooks.org/wiki/LaTeX/Boxes#parbox,_minipage,_and_pbox does have them but it's wrong
@UlrikeFischer Really? Where?
4:00 PM
@JosephWright page 866
@UlrikeFischer Ah, drat: I tried all of the other index entires
@UlrikeFischer It's not on p 863, so I thought it was not included
@JosephWright I didn't go by index, I knew where it is ;-)
@UlrikeFischer That would helpe
@UlrikeFischer OK, so it's not a coffin pole in this case, but I can sort it :)
@samcarter_is_at_topanswers.xyz Thank you very much. Edited the post a bit to show \mkbibcompletename and the name wrappers.
@samcarter_is_at_topanswers.xyz biblatex 3.15 is on the road to becoming the most buggy biblatex release yet. It's probably going to need a second bugfix release soon :-(. So if you find some bugs, now is the time to report them.
@AlanMunn Revised 8 Aug 1995. so it's more or less 2.09
@JosephWright texdoc usrguide
4:16 PM
@DavidCarlisle Too bad. It was pretty useful.
@AlanMunn but the text in usrguide is so clear (as is the source code) what else is needed? :-)
@DavidCarlisle I thought Lamport updated for 2e, so shouldn't it be there?
@DavidCarlisle I had no idea that document existed. But it actually does look like a useful thing to know.
@DavidCarlisle you wrote it?
@DavidCarlisle Section 7 is extremely useful.
4:21 PM
@DavidCarlisle Description of inner pos default doesn't seem to match the code ...
@JosephWright isn't the duplication of #1 here {\@iiiparbox{#1}{#2}[#1]}} doing that apart from s which is % \changes{v0.1d}{1993/12/15}{Changed default from 'c' to 's'} probably later than usrguide...
@DavidCarlisle Exactly :)
how do I decrease the size of \\?
@DavidCarlisle It doesn't really make a difference
[0.1em] does not seem to take effect
\\ [0.1em] does not decrease anything
4:30 PM
@nbro best not to use \\ at all, but \\[-2in] will make it go down less.
@nbro \\ isn't anything that has a size, it just specifies a line break at that point, so the following text comes baselineskip below the current text, just like an automatic line break.
@nbro that increases the space between the lines by 0.1em.
so I had to use the minus
@nbro as I say, if you are needing to use it, something is wrong, somewhere.
it's in tikz pic, so I guess it's not wrong there
@nbro inside a node presumably, but even there why is the baseline spacing wrong that you need to adjust it on the actual line break?
well, I don't want too much space in a specific case
4:39 PM
in other news latexcgi.xyz tlmgr update so if you need to try out all the latest biblatex bugs, now's your chance:-)
@nbro [..] is there and a documented user command so it's not wrong, just that in 99% of cases it is misused. So I'll assume that you are in the 1% and it's fine:-)
4:51 PM
I wanted to check if something happened in filehook but the repo is no longer reachable ;-(
@Skillmonlikestopanswers.xyz with the next latex-dev version you will no longer need the work around with xfp. Then this will work out of the box:
\AddToHook{shipout/background}
  {%
    \put(1in+0.5\paperwidth,-1in-0.5\paperheight)
      {$\vcenter{\makebox[0pt][c]{\mywatermark}}$}%
  }
@DavidCarlisle, @UlrikeFischer Thanks for the info: I'm closing in on a working \parbox in xbox :)
@DavidCarlisle, @UlrikeFischer Still fiddling with coffin handles (to avoid \vcenter)
5:38 PM
@MarcelKrüger what had we for numbering system for version in luaotfload? I would like to increase the number to something "before 3.15". Would 3.15-dev or 3.15-1 work?
@UlrikeFischer I don't think we had anything like that before, so we can be creative. Basically IIRC we would otherwise have used 1.14.1 etc. IMO 3.15-1 would be very confusing (That notation is used by my package manager to mean: The first packaged version of 3.15), so I would vote for 3.15-dev or 3.15-alpha1.
@MarcelKrüger if I may, semver.org (and sorry for jumping into the conversation)
3
@PauloCereda :)
@MarcelKrüger the main question is if the tagging gets confused if I'm creative. I suggest that we use 3.15-dev and regularly update the package date. Then it should be recognizable enough.
@MarcelKrüger I forgot, we have code that looks for -dev don't we?
@UlrikeFischer Oh right, the current tagging only accepts 3.15-dev. I'll look into changing that for that for the next release.
@UlrikeFischer Yes, we can use -dev.
5:51 PM
@MarcelKrüger I'm not sure if we really still need the special handling for dev, it was quite useful when we had two engines, but now it can install itself imho normally.
@PauloCereda The installation target is actually not triggered by dev, it always installs into latex-dev except if the branch starts with master (because of Travis). Feel free to drop that, I don't really mind where it installs to.
6:09 PM
@PauloCereda For many things I would agree, but sometimes semver leads to silly version numbers.
IMO luaotfload is such a case: The official API is of so little importance to the actual user that choosing version numbers based on that would just be misleading. The much more useful interpretation of backward compatibility would be output compatibility, but that technically breaks in every version, so we would end up with Chrome-ish version numbers like 45.0.0 which isn't particularly helpful to the user either.
@MarcelKrüger it is ok for now I think. I increased the version number and added a work around for filehook.
 
1 hour later…
7:28 PM
 
1 hour later…
8:49 PM
@yo' -- You mean "What is the motivation behind LaTeX?" I can't think of anything in TUGboat. I think the original motivation for Lamport was to have a nice way to prepare his papers; the motivation for AMS was AMS-TeX users wanting to make use of automatic numbering and cross-referencing and LaTeX had that (AMS-TeX didn't). And otherwise, other users wanted or needed other things. Individual bits are written about in TUGboat, but no overall coverage.
yo'
yo'
@barbarabeeton yeah, that's what I realized, that there's no single place to check. But thanks for the interesting note!
in the end I send the user some interesting links on TUG and L3 websites, and also the video of Frank's talk from Rio
*sent

« first day (3585 days earlier)      last day (1542 days later) »