7:19 AM
@DavidCarlisle ooh

8:03 AM
Q: How much space will Brexit free up in the European Union?
A: 1 GB
7
@DavidCarlisle ^^

@PauloCereda Goodness, this made me laugh way more than it should XD

@Plergux same here.

@Plergux :D

@barbarabeeton The TV exaggerates, here in Milan the situation is quiet, even if the sanitary situation is getting worse

It's like a joke my son told me the other day:
Q: What to fishes smoke?
A: Seaweed.
I swear, I spent the rest of the day giggling XD

8:12 AM
@Plergux LOL

And of course, my favourite:
Q: Why can you not hear the pterodactyl in the toilet?
A: The p is silent.

@Plergux LOL
@Plergux good one :)

@PauloCereda :p I could carry on all day, but I'll quit while I'm a head. :p

@Plergux ooh :)

8:34 AM
@PauloCereda clearly the wrong answer. You lose all of the UK not just GB

@DavidCarlisle :(
@DavidCarlisle Trivia: whenever I send stuff to you, although I clearly state the parcels are going to UK, the postal service still treats UK \equiv GB.

@PauloCereda so do lots of things but it's a useful distinction to make to put down foreigners making jokes...

@DavidCarlisle It's pronounced BUK! XD

@DavidCarlisle hey I actually am well versed on the subject, thanks to this video: youtube.com/watch?v=rNu8XDBSn10
ooh Venn diagrams

8:58 AM
@PauloCereda If Italy left, it would be only a b-It

@CarLaTeX o.O

9:35 AM
Yesterday, I edited tex.stackexchange.com/a/439690 without understanding too much what I am doing such that the answer works for me and replaced \c_zero with 0. If anyone would like to comment on this or suggest a better solution, please feel free.

@GeekestGeek \c_zero should be \c_zero_int nowadays (\c_zero was deprecated).

@DavidCarlisle Well you could argue that with the backstop and open border to Ireland, only GB will really be lost.

@GeekestGeek wait, now that I took a look, forget about that.
@GeekestGeek it would work with \c_zero_int (I haven't tried but see no reason why it shouldn't), but in this context 0 is fine as well.

2 hours later…
11:20 AM
@MarcelKrüger tell that to B.J.

@DavidCarlisle Oh no, real politics

@JosephWright sorry should I return to "don't worry it'll soon be lunchtime"

@DavidCarlisle I think it's inevitable as we approach the Business End
@DavidCarlisle Did you see that the International Trade dept has been celebrating that we will pay 0% tariff on Japanese soy sauce under our new deal ... all great until you find out that's already the case with teh EU agreement ... and most soy sauce is not made in Japan

@DavidCarlisle oh no

@JosephWright what is "Great British Bake Off"?? Something in TV?

11:36 AM
@UlrikeFischer Yes, a cooking show

@JosephWright sounds famous, if it triggers a tweet from a dept.

@UlrikeFischer It's on Channel 4, one of the main terrestrial channels, and it's pretty popular
@UlrikeFischer Opportunism of course; 'ooh, they are doing Japanese things, let's show off our treaty with Japan'

@JosephWright yes, I understand that, but I can't think about a show, even less a cooking show, in Germany that is so popular that you could refer to it like this in a tweet. 20 years ago it was different, but now everything is quite spread out.

@JosephWright WOO

@PauloCereda You're missing the BOTs and Crown dependencies.

11:47 AM
@yo' oh it's later in the video, it gets REALLY complicated. :)

@PauloCereda oh I haven't seen it; the diagram misses them :)

2
@yo' ^^ :)
This is deliciously complicated

@PauloCereda That's better :D

@yo' LOL

11:50 AM
@yo' OOH

^^ "That's better!"

@UlrikeFischer It's the original version of sat1.de/tv/das-grosse-backen

12:09 PM
The German version has ducks: i3-img.sat1.de/pis/ezone/…

@PauloCereda New background

@JosephWright Haven't tried it yet

@PauloCereda :)

@JosephWright Have to finish my slides for the GuIT meeting

12:27 PM
@PauloCereda Updating my VM too

1:11 PM
@PauloCereda I cannot find the EU in that picture. ;)

1:22 PM
@samcarter_is_at_topanswers.xyz Oh, thanks. As if being obsessed with the Great British Bake off, The Kiwi Bake Off, the Australian Bake off, The Irish Bake oFf and the Canadian Bake off wasn't enough, now I have to watch this too! XD

@mickep oh

@Plergux :D Your missing the French version! In one episode they baked on board of a boat cruising on the Seine - it was lovely to see all the nice buildings of Paris in the background :)

@JosephWright seems happy today ^^

1:38 PM
@PauloCereda lua-ul package docu ;-)

@UlrikeFischer ooooooooooooooooooh

@PauloCereda ohhhhh

@MarcelKrüger I WUB YOU

@PauloCereda didn't you know the package? it came from here tex.stackexchange.com/a/446488/2388

1:56 PM
@UlrikeFischer ooh

2:25 PM

@Plergux ooh
I tried DuckDuckGo... to look for Google results. :D

Good morning/afternoon/evening.

@LaTeXereXeTaL 'ello!

@LaTeXereXeTaL no "good night"?

@samcarter_is_at_topanswers.xyz Oops. Good night to the Aussies.

2:36 PM
@samcarter_is_at_topanswers.xyz oooh ding ding ding the MP sketch

It's duck weather here. Pouring rain.

@LaTeXereXeTaL ooh quack

2:48 PM
@samcarter_is_at_topanswers.xyz Is that a Tikzling? :p

@Plergux Oui :)

@samcarter_is_at_topanswers.xyz Ah, maginfique! I've downloaded the documentation for them and skimmed it, but I've not made anything with them yet. I've got a strong urge to try to sneak one into my doctoral thesis XD

ooh French
Je suis a canard
Le pomme de terre

I want to freeze the current version of my package in anticipation of the next version, which will have a new version number. What are the thoughts on having numbers/digits in a package name to distinguish it from other versions? I see it done, but is there any reason not to do it?
Changes between the old and new versions are sufficiently great that the old version will effectively be broken and I want to keep it around for people who use it.

2:59 PM
@PauloCereda You have a Twitter account, right?

@LaTeXereXeTaL we use semver.org for our softwares, perhaps it could be used for a package. :)
@MarcelKrüger I do!

@PauloCereda Could you ask the Japanese guy if s/he can send me an example document. Then I can try to fix the underlining.

@MarcelKrüger hold on!

@PauloCereda @MarcelKrüger I have to teach a class for the next two hours but I will look at this. Looks similar to github, only different.

@MarcelKrüger Done, now we wait. :)

3:03 PM
@PauloCereda Thanks!

@MarcelKrüger My pleasure, I didn't notice it had a problem. I am a simple duck, I see ducks, I vote. :)
@MarcelKrüger by the way, this is how texdoc-online handles 404 pages: texdoc.org/notfound :)

@PauloCereda TeX.SX should adopt that page too!

@MarcelKrüger <3

@ pgfmath
What does dim calculate?
\documentclass[a4paper]{article}
\usepackage{tikz}
\begin{document}
\def\List{1, 2, 3}
List: \List  \par
Dim of List: \pgfmathparse{dim(\List)}\pgfmathresult

\bigskip
\def\List{200, 1, 3}
List: \List  \par
Dim of List: \pgfmathparse{dim(\List)}\pgfmathresult

\bigskip
Dim of 200: \pgfmathparse{dim(200)}\pgfmathresult
\end{document}

3:26 PM
@JosephWright Bruno removed the (in the kernel) unused argument from \tl__act:NNNnn (now \tl__act:NNNn). That argument might not have been used, but it could be used to pass information through to the functions of the first three arguments (e.g., done in my PR #722). That commit did make \tl__act: faster, but reduced its versatility (though it's not currently used in the kernel).

What is the collection of commands \mathord, \mathrel, \mathbin etc. called? Knuth seems to sometimes use the term “character classes”, but it hardly makes sense to call e.g. \mathbin{\mathop{\otimes}\limits^L} a single “character”. So what do we call them? Math classes?

@PauloCereda La pomme de terre ;-)

@UlrikeFischer oopsie
Die Apfel

3:42 PM
@PauloCereda Der Apfel (=1), die Äpfel (>1) ;-)

@UlrikeFischer ooh

@UlrikeFischer I never understood why plural in German looks like a bad copy of feminine ;-)
I mean, die die der der vs die die der den

@Skillmon Well \__tl_act: is as you say internal, and one can't always reuse it: for example in the case changer I have to work manually due to look ahead

@Gaussler in the TeX book they are just called "classes" and it is said that "TeX associates classes with subformulas as well as with individual characters"
@JosephWright yes, sure, I just wanted to note so that this doesn't go "unnoticed"

@cis the result of dim(123456123345) is 12.

3:57 PM
@Skillmon I guess if we need it back, it's easy enough to sort
@Skillmon You are taking a lot of interest in expl3 ... risky plan

@LaTeXereXeTaL LaTeX actually does cope with that. You can save the frozen version as mypackage-yyyy-mm-dd.sty, then whoever wants to load that specific version can do \usepackage{mypackage}[=yyyy-mm-dd] and then frozen version will be loaded instead

@MarcelKrüger I looked at all that lovely word boundary data ... but I think it's a bit too much to load in expl3 :)

@JosephWright just a random boost in interest to divert myself :)

@UlrikeFischer For Greek, I think I'm going to arrange to load data for pdfTeX from UnicodeData.txt - I'm getting a bit of a data build-up!
@Skillmon Like I said, risky ;) (Ask @egreg or @UlrikeFischer or @PhelypeOleinik)
@Skillmon (I think @MarcelKrüger's mistake was being good a Lua)

4:04 PM
@LaTeXereXeTaL there is a longish discussion at the moment complaining discussing revtex's use of numbers as in revtex4-2.cls it's a real pain avoid it f at all possible. If you really must use the package rollback mechanism so users can do \usepackage{yourpackage}[v1] to get a saved copy of the first version.

@JosephWright at least on that front I'm safe :)

@Skillmon Yeah, but Lua is so easy ... until you get to node lists

@DavidCarlisle It was precisely that discussion that prompted my question!

@DavidCarlisle Yup, or do-it-yourself like pgfplots or mhchem or ...

@LaTeXereXeTaL oh @PhelypeOleinik just said same (but then he is to blame for knows about the revtex class name

4:06 PM
@PhelypeOleinik Yes I'm aware of the versioning mechanism. I hadn't considered using it but that seems the simplest solution. (back to my class......)

@JosephWright I wrote a bit of Lua (a few widgets for my window manager), but still I have no clue about Lua :)

@DavidCarlisle revtex4-\infty
2

@JosephWright or never have bugs and need a new version like indentfirst

@Skillmon I wrote a whole build system more-or-less by accident and with loads of rubbish code (again, ask @MarcelKrüger)
@DavidCarlisle longtable

@JosephWright I feel ducksay is missing as the most well known and widely used package from this list!
2

4:07 PM
@DavidCarlisle “This file has version number v1.03, last revised 1995/11/23.”
Hm...

@PhelypeOleinik insert "Lies! Deceptions!" meme

@Skillmon Running faster with the lastest updates???

@Skillmon I didn't say where I got that line from :-)

@JosephWright ducksay got a lot faster (during load time) compared to the original version. But v2 changed the entire measurement and output routine, giving much more options. I don't think the code for a single animal got much faster.
@JosephWright or did you mean with the latest kernel updates?

@Skillmon :)
@Skillmon That

4:11 PM
@JosephWright that I didn't benchmark (but almost everything using l3tl got a tad faster, so I guess so)

@Skillmon I had hoped we'd see a benefit on Travis-CI, but nothing obvious

@JosephWright the differences are small (at most something around 15-20%) and that on macros which aren't that slow in the first place, it was just tickling the last bit of performance out of the described interfaces.

@Skillmon Yes, I get it :)
@Skillmon I'm hoping to find a 'final' home for xparse, so we don't have to copy it for all of the L3 docs: that will make Travis-CI faster

@PauloCereda In addition, one cannot use a (moderate) lower case writing in German, as is known in English:

Try to translate this sentence correctly:

"In moskau hat er liebe genossen."
vs. "In Moskau hat er liebe Genossen."

4:26 PM
@JosephWright Thanks. It's my favorite kind of error: I can blame someone else. :P
3

@JosephWright learnlatex.org: LaTeX2e <2020-10-01> patch level 2
@MarcelKrüger is there any other kind?

4:46 PM
Please, how does LuaTeX translates font name into file name if it's a system font? It there a command line tool for that?

@PhelypeOleinik Oooooo reading that article now!

@yo' luaotfload-tool --find=..., see the manpage for details.

@MarcelKrüger ah thanks!

@yo' It's currently broken though, but you can download the fixed dev version from GitHub

@MarcelKrüger oh I bet it's not broken on Overleaf TeX Live 2020 image :)

4:53 PM

@yo' Well, it's been like this since the beginning of September. So it might go either way.

@MarcelKrüger I think we freezed mid-August or so. And with luaotfload broken, we wouldn't pass our own sanity checks I beleive, with a lot of the test projects being on LuaTeX with strange fonts and stuff.

@yo' luaotfload isn't broken (otherwise we would have done a patch release already), it's just font lookup with luaotfload-tool. (Some new lookup code didn't have the special case for the tool version implemented)

@MarcelKrüger ah I misunderstood you! Well, seems to work on Overleaf, so we're fine :-)

@yo' Great. But why do you need LuaTeX compatible font lookup on the command line in Overleaf?

5:07 PM
@MarcelKrüger to check the font location, a user needs TL2019 for a package (there were seemingly some backwards-incompatible changes), but wants a TL2020-only font.

how to force multicols to extend to one line?
I'm using this
\begin{multicols}{2}

\textbf{Linearity: }
\begin{align*}
y_1(t) &= u^2_1(t) \\
y_2(t) &= u^2_2(t) \\
u_3(t) &= c_1u_1(t) + c_2u_2(t) \\
y_3(t) &= u_3^2(t) \\
&= \left[c_1u_1(t) + c_2u_2(t)\right]^2 \\
&= c_1^2u_1^2(t) +2c_1c_2u_1(t)u_2(t)+ c_2^2u_2^2(t) \\
&\neq c_1y_1(t) + c_2y_2(t)
\end{align*}
The system is not linear.

\textbf{Time-Invariance}
\end{multicols}

@CroCo \setcounter{unbalance}{1} (probably) before the multicols

@DavidCarlisle not working

@CroCo well you didn't supply a test file so I didn't test:-) multicol is already struggling to balance that display when the first column contains the large unbreakable align. Not sure how unbalance logic works in that case (some decades since I looked at the code...) you could always add the line inside the align in an \intertext

@DavidCarlisle so there is a possibility then, right? I can double check the documentation.

5:23 PM
@CroCo sure or you could use the star form to disable column balancing, and use a forced column break I think.

@DavidCarlisle I've used the star form but unfortunately, it takes the entire page for multicols which I don't want. I want it to be inside the content.

@CroCo \columnbreak should work, but it is easier to answer such questions if there is a complete test document.

@PauloCereda Better stick to Portuguese...

@AlanMunn ooh eu ser um pato

@CroCo vv
\documentclass{article}

\usepackage{multicol}
\usepackage{amsmath}

\begin{document}

\vspace*{-3cm}

A
\begin{multicols}{2}

\textbf{Linearity: }
\begin{align*}
y_1(t) &= u^2_1(t) \\
y_2(t) &= u^2_2(t) \\
u_3(t) &= c_1u_1(t) + c_2u_2(t) \\
y_3(t) &= u_3^2(t) \\
&= \left[c_1u_1(t) + c_2u_2(t)\right]^2 \\
&= c_1^2u_1^2(t) +2c_1c_2u_1(t)u_2(t)+ c_2^2u_2^2(t) \\
&\neq c_1y_1(t) + c_2y_2(t)
\end{align*}
The system is not linear.

\textbf{Time-Invariance}
\end{multicols}

5:32 PM
@MarcelKrüger did you get the link?

@PauloCereda Hmm. Better stick to quacking.

@AlanMunn ooh woof
:)

@DavidCarlisle, perfect thanks.

@PauloCereda You quack me up.

5:34 PM
@UlrikeFischer thanks. I will try it.

@UlrikeFischer yes but in English, @PauloCereda doesn't understand.

@UlrikeFischer oh sorry

6:09 PM
How to push equation in the second column to the left
\documentclass{article}
\usepackage{geometry,amsmath}
\geometry{a4paper, margin=8mm}

\begin{document}

\begin{minipage}{0.45\linewidth}
\textbf{Linearity: }
\begin{align*}
y_1(t) &= u^2_1(t) \\
y_2(t) &= u^2_2(t) \\
u_3(t) &= c_1u_1(t) + c_2u_2(t) \\
y_3(t) &= u_3^2(t) \\
&= \left[c_1u_1(t) + c_2u_2(t)\right]^2 \\
&= c_1^2u_1^2(t) +2c_1c_2u_1(t)u_2(t)+ c_2^2u_2^2(t) \\
&\neq c_1y_1(t) + c_2y_2(t)
\end{align*}
The system is not linear.
\end{minipage}
\hfill
\begin{minipage}{0.45\linewidth}
\textbf{Time-Invariance:}

6:35 PM
@PauloCereda Oh, I accidentally answered to Joseph instead of you. But yes, I got it. Thanks.

@DavidCarlisle I see that learnlatex.org stats are ticking over nicely
@DavidCarlisle Just need to get the design finished ...

@JosephWright I'm quite curious about it, how can one see it?

@CroCo Because the align contents is narrower it moves further out due to the size of the minipage. If you make the minipage narrower the content moves further into it.

@MarcelKrüger My pleasure

6:44 PM
@JosephWright nice, the vietnamese font looks odd on my machine, some chars are bolder and smaller then others.

@JosephWright ? what does breadcrumbs mean here?

@UlrikeFischer Links to show where in the sructure you are

@JosephWright yes, one can see that things are still missing, but it looks quite good.

@UlrikeFischer Fingers-crossed

6:57 PM
@JosephWright ;-). But vietnamese looks definitly odd, now that I looked more carefully: I get a mix of sans serif and serif.

@UlrikeFischer Eeeerk

2 hours later…
8:41 PM
@UlrikeFischer yes flagged that already, as did JouleV apparently you need a newer version of the Roberto font the version there doesn't cover Vietnamese @JosephWright

8:58 PM
@UlrikeFischer hmm issue is closed but github.com/learnlatex/learnlatex.github.io/pull/…

@DavidCarlisle oh, a rather old issue ;-).

9:23 PM
I am confronted with a mystery: In one of my documents, I use tcolorboxes and want to be able to continue paragraphs after them. If I add \noindent\ignorespaces to the after hook of one invocation of the tcolorbox environment (using \AddToHookNext), that pushes following content further down the page leading to an undesirable page break. If I instead manually add \noindent\ignorespaces after the \end of the environment, that does not happen.

I have not been able to boil this down to an MWE, which is why I'm not asking a proper question. Does anyone have an idea what might be

10:19 PM
Dumb question: I'm using \graphicspath{{./chapters/ch2/figs/}} to not have to give the path to my figures every time: is there a way to do the same thing for pgfplots?

10:54 PM
@schtandard you probably need \ignorespacesafterend if I understand correctly

11:14 PM
why did we redefine \@hspace?
Oh "support calc"

11:57 PM
@Skillmon Thanks! So, what is your advice: keep 0 or change to \c_zero_int?