Not a palindrome. Reduplication. :)

@AlanMunn Congrats! :)

@Werner That's exactly what I was wondering (I downvoted it, btw)

@DavidCarlisle Seen the TeX Live mailing list issue re. \undefined and pdftex.def?

@AlanMunn a square :-)

@yo' On the contrary, it is square free! Prime decomposition: 7·11·13·137. That 1001=7·11·17 is a curious fact, little known.

@HaraldHanche-Olsen a square-free square :-)

@AlanMunn In a novel by L. Sciascia (Il giorno della civetta, The day of the owl), the mafia boss calls quaquaraquà one of the categories he divides men into (ducks are involved, of course).

@JosephWright Reading the old thread I wonder if it isn't time that the code of supp-pdf.mkii wanders properly into latex. I doubt that Hans wouldn't allow it. You could replace every \ifx\XX\undefined by \ifundefined\XX and load it if etex is used ;-)

@JosephWright @UlrikeFischer Ever tried looking for \undefined in LaTeX packages?

Umm, that should be 1001=7·11·13, of course … five messages back.

@egreg In the old thread Heiko did it and reported around 530 cases. I got 648 but there are wrong matches of the type \undefinedsomething included. In tools and base I found only 6 and 8. (One real hit is in in longtable and one in bm and in shellesc @DavidCarlisle)

@UlrikeFischer I get 1004 for findstr /s /r "\\undefined[^a-z]" *.* | wc -l in texmf-dist/tex/latex

@JosephWright I counted the files. With the [^a-z] I get 448.

@UlrikeFischer Ah

@UlrikeFischer Possibly though it's a non-trivial amount of code to copy (and presumably actually read over)

@CarLaTeX Oh no

@DavidCarlisle I've added the fix: do we want to release? (I notice the 'multi-dots' business is outstanding, but presumably depends on kernel stuff)

@UlrikeFischer I probably will pick the ideas in an expl3 implementation ...

@PauloCereda but they didn't turn off the lights, so it was difficult to see you :):):)

@CarLaTeX woo :)

@CarLaTeX: we ducks know stealth mode. :)

@PauloCereda :)

Oh, that's why SP's traffic was terrible yesterday! DAMN YOU U2!

@samcarter, @CarLaTeX: developpez.net/forums/blogs/657637-nikopol/b3520/…

@JosephWright not yet, bit distracted at present

@DavidCarlisle Cool

@JosephWright yes defer the dots and other changes to kernel update so we could update graphcs anyway (or os it gragohics-def? (Ill look in a bit:-)

@JosephWright how's Cuphead?

@DavidCarlisle Gave me a chance to try out some GitKraken features, spot that the tags for graphics-def needed tidying up, learn a bit of nano config, ...

@PauloCereda Have beaten one level (Botanic Panic): thus far that is it

@JosephWright ooh psychic carrot

@DavidCarlisle It's graphics-def and with two (or more) branches that's fine: I can release today if necessary

@PauloCereda Yup: I'm not 100% so was at home yesterday, though I did have exam papers to get ready too

@JosephWright oh

@PauloCereda @samcarter Cette extension va révolutionner vos documents LaTeX LOL

@PauloCereda Also watched quite a lot of Netflix

@JosephWright speaking of kernel stuff is it OK to branch in git or would that complicate making that public later (not actually got any code in a state to check in)

Hi, to everybody

@DavidCarlisle Should work fine: gets translated to an SVN branch auto-magically

Can I ask an help?

How do with the my offer bounty? I have not understood.

@DavidCarlisle I think we probably should make the switch, but it's down to others

@DavidCarlisle I'm just waiting for sign-off from FMi

@JosephWright, @DavidCarlisle, @PauloCereda, @CarLaTeX, @ChristianHupfer, @egreg good morning

@Sebastiano Good morning!

@DavidCarlisle See for example the one 'mess up' I made in checking in to Git early-on: SVN saw it as a branch-of-sorts (what Subgit call a 'shelf')

@Sebastiano Ciao!

@JosephWright I suppose I could anyway check in my experiments in my clone even if I don't push, can decide later whether to push that or just check them into a clean clone later

@CarLaTeX can you help me, please?

@DavidCarlisle Exactly: as you observed, there's a workflow where everyone does more-or-less that and everything to the main repo is done using pull requests ...

@Sebastiano 'ello

@PauloCereda :-)..'ello to you from Sicily

@JosephWright probably I'm going home this evening so may get some time to look at this later in the week

@Sebastiano You should see some button next to the answers that you can use to award the bounty.

@egreg Peraphs I have understood. I have clic the button +50 (near check green) and I have seen a window up the screen. Is it correct? Thanks a lot lot

@egreg Ok. It is correct. Thanks a lot for your kind. Ciao. Good work

Hola' to everybody

@Sebastiano Yes, it's correct

@PauloCereda @CarLaTeX L'article est très aimable!

@samcarter we canards are very popular. :)

@samcarter: I just commented on CTAN. :)

@PauloCereda You should buy an H at the Wheel of Fortune ;-)

@ChristianHupfer ooh

@PauloCereda I will found the League to increase the use of H ;-)

@ChristianHupfer ^^ I am afraid of buying vowels. :)

@ChristianHupfer ooh a league of gentlemen

@PauloCereda Well, and ladies too...

@ChristianHupfer ooh

Vanna White?

@PauloCereda No, Strawberry Fields ;-)

@ChristianHupfer :D

@PauloCereda: Or Honey Ryder, P. Galore...

@PauloCereda Ohhhh! Thanks for the kind comment <3

@samcarter we ducks are very good at comments. :)

@PauloCereda From experience I can confirm that canards are popular in France -- but in most cases with no good end for the duck :(

@samcarter oh no

They are mean

@PauloCereda At the last conference in France I had to pretend to be vegetarian to escape the duck in two out of three courses of the conference dinner.

@samcarter Sacred bleu!

@PauloCereda :-P

OH NO

@PauloCereda Les pauvre canards! However the location was great - a dinner cruise on the Seine!

@samcarter oh

I have a TikZ-style set by foo/.style={...}. Somewhere in my document, I want it to have no effect. However, \tikzset{foo/.style={}} yields no result. Why?

@Bubaya Impossible to say without more info I think, what you're saying works just fine.

@TorbjørnT. OK, so styles are supposed to be canceled this way? Then I'll try to figure out where I messed it up.

@Bubaya Redefined, not cancelled, but this for example works fine:

\documentclass[margin=10pt]{standalone}
\usepackage{tikz}
\tikzset{foo/.style={draw=blue,fill=red}}
\begin{document}
\begin{tikzpicture}
\node[foo]{a};
\end{tikzpicture}
\begin{tikzpicture}[foo/.style={}]
\node[foo]{a};
\end{tikzpicture}
\begin{tikzpicture}
\node[foo]{a};
\end{tikzpicture}
\end{document}

@TorbjørnT. I see. Then I have to messed it up somewhere. Thanks.

@PauloCereda @samcarter: I stole the comment:

@UlrikeFischer oooooh <3

@UlrikeFischer Yay!!!

@UlrikeFischer <3 (advertisement: The next tikzducks version will have its own speech and though bubbles github.com/samcarter8/tikzducks/commit/…)

@samcarter Oooh more and more features!

@samcarter Good ;-) I wanted a double pointer and so stole code from tex.stackexchange.com/a/109506/2388. Btw: It would be quite useful if one could name the ducks. Currently it is a bit awkward to handle more than one duck in a picture:

@UlrikeFischer Would a new option \duck[name={blub}] do which creates an coordinate of said name in the centre of the duck?

or maybe adds blub as a prefix to the name of the existing coordinates?

@CarLaTeX kind of addictive :)

@samcarter I think the second is better. Perhaps even better blub. (with a dot) as the prefix (if it works). Then the names look more like the normal anchor names of shapes like a.south.

@UlrikeFischer Don't think blub. would work, at least if you have a node named foo.bar, when you try using foo.bar as a coordinate, you get an error saying no shape foo is known.

@TorbjørnT. I feared it ...

Trying to solve this theorem-enumerate bug. What causes the slight misalignment in the following MWE?

\documentclass{article}
\usepackage{enumitem,amsthm}
\begin{document}
\newlength{\algnRef}
\settowidth{\algnRef}{\textbf{Fact.} \strut}
\addtolength{\algnRef}{1em}
\setlist{leftmargin=\algnRef}
\newtheorem*{fact*}{Fact}
\begin{fact*}
	\begin{itemize}[leftmargin=1em]
		\item Bla
	\end{itemize}
	\vspace{-\topsep}
	\vspace{\parsep}
	\vspace{\itemsep}
	\begin{itemize}
		\item Bla
		\item Bla
	\end{itemize}
\end{fact*}
\end{document}

@UlrikeFischer I could offer something like

	\duck[prefix=blub]
	\fill[black] (blub-wing) circle (0.04);

Would this notion be too strange?

@samcarter looks good. But I would use name=. Then you could write example like \duck[name=zorro] or \duck[name=knuth] (I leave out the obvious example ...).

I hate ducks

@alfred I ate ducks

@alfred I made ducks

please stop

ducks everywhere

@alfred Quack quack quaaaaaaaaaaaaaaack

@alfred I guess when you see a duck you should just duck.

@alfred direct them all here: chat.stackexchange.com/transcript/message/39233670#39233670

@alfred and so hyphenation verbosity in the chatroom, so we are even. :)

@JosephWright ooh another duck :)

@AlanMunn ooh duck puns. :)

@UlrikeFischer I pushed a new version to github: github.com/samcarter8/tikzducks/commit/…

@samcarter oh my, this is clever!

@samcarter: NAMED DUCKS

@PauloCereda The praise should go to @UlrikeFischer - It was her idea!

@samcarter YAY

@samcarter: Although @DavidCarlisle is a TikZ expert, he is mean. :)

@PauloCereda When will the dictionary add an alternative denotation to the phrase "he is mean"? From the usage it is clearly either a salutation or a compliment :)

@samcarter ooh context

@samcarter Or it is a statement that the person is average.

@samcarter It's mainly ritual, I think. Call and response.

@AlanMunn So no message at all?

:)

@samcarter I'm thinking about good names ;-) Who is pulling here whom?

@samcarter Exactly. Although the ritual obviously has some meaning.

@AlanMunn As there is a duck involved in this ritual, one should ask some biologist about the meaning.

@UlrikeFischer I'm curious which names you will come up with!

@HaraldHanche-Olsen If the average amount of tex code each and every person has contributed would equal the amount by @DavidCarlisle I wonder what the world would look like :)

@samcarter Well my first idea was Carla pulling Egreg away from tex.sx so that you get the royal flush first ;-) But then the second duck must be green.

@samcarter Buggy Feature filled?

@DavidCarlisle ooh

The levels of embedding are becoming oppressive in here.

@AlanMunn do you think so?

@UlrikeFischer Green is not a problem and from her new avatar we know what a @CarLaTeX duck looks like

@DavidCarlisle Maybe I should reconsider.

@AlanMunn To go shopping for clotheing would certainly be easier in such a world: do \usepackage{graphicx} on entering the store and if something does not fit, just scale or specify width or height

ooh

@AlanMunn: tail recursion at its finest. :)

@samcarter :) <3

@samcarter LOL

@samcarter \resizebody{height=1.8m} :)

@samcarter you could also have a wide variety of coloured stripes, ignore these "style gurus" who say vertical stripes should be avoided

@AlanMunn @DavidCarlisle TeX is so much better than the real world :)

@DavidCarlisle (@UlrikeFischer perhaps) Do you think the dvips colour names are actually that useful?

@JosephWright possibly not now, it saves a lot of macro token memory but bytes are cheaper these days...

@DavidCarlisle My thinking also: I'm not sure they even make it into the PDF

@DavidCarlisle Pondering Chris's mail

@JosephWright I doubt anyone just coming in from the latex2e color (or even xcolor) documentation would even really notice the names are there

@DavidCarlisle :)

@JosephWright Do you mean the names like "BurntOrange" or the special handling with dvips/postscript when the prologue option is used?

@DavidCarlisle Not sure why he has an issue with 'spot' colours ...

@JosephWright I don't think they would even with dvips if you use the standard header they will get expanded to cmyk colours while distilling

@UlrikeFischer I haven't looked at prologue yet: thinking BurntOrange and the like, and more specifically handling them separately from just doing \definecolor for all of the names

@JosephWright definecolor or definenamedcolor?

@DavidCarlisle \definecolor, i.e. not worrying about saving tokens. Like I said, haven't looked at the prologue business yet, and I need to see what actually gets done by ps2pdf in that regard.

@JosephWright Who is he? I'm currently at lost where the problem is. Also isn't dvipsnam.def from David?

@UlrikeFischer He = Chris R.

@JosephWright yes but it ends up with a different user interface that way (ie by default BurntOrange` isn't defined as a color but is a value in the value space of named colours, so \color{BurntOrange} isn't defined but \color[named]{BurnedOrange} is. basically the names are mostly user-defined apart from the primary colours,

@UlrikeFischer Yes, dvipsnames is @DavidCarlisle's but from quite some time ago ...

@JosephWright so if you were to drop the predefined dvipsnames I'd tend towards not defining them at all rather than predefining a lot of names in the "user-space" of names defined by \definecolor

@DavidCarlisle I know :) I'm not imagining doing \cs_set_eq:NN \color \color_selected:nn or whatever :)

@DavidCarlisle Indeed, hence the 'are they useful' question: I don't really see that they are (particularly for pdfTeX where they just get turned into CMYK anyway)

@JosephWright I think the only real use is if you've got documents using them, so dropping by default and providing some option or add on package that provides a compatibility layer should be fine I think (same for svgnames, htmlnames from xcolor)

Anyone of the l3 guys out there that could give conceptual advice here (it already got a Tumbleweed-question)?

@DavidCarlisle I guess you are imagining I get something like xgraphicx working :) Take over all the \color, \textcolor, \includegraphics, etc. functionality!

@JosephWright well you're well over half way there aren't you?

@DavidCarlisle Yes, I guess: most of the low-level graphics stuff is done, almost all of color is working, just have to attach it all together (and move from l3trial to l3experimental)

@DavidCarlisle I actually have almost all of the 'system' layer of TikZ too: that one could be very interesting :)

@JosephWright can you make tikz and latex use the same colour models so tikz colour is pushed and popped on the same stack as latex colour?:-)

@JosephWright It is still quite unclear to me about what you are speaking but I know about prologue and named colors with dvips only because of a recent question about the acmart.cls which uses the named color model tex.stackexchange.com/a/396352/2388.

@UlrikeFischer In itself that's an answer :) You see lots of LaTeX code 'in the wild', and hadn't seen much use of the named model. So it's probably not that widely used/useful.

@DavidCarlisle I can only do so for new code :) Of course, I could set up an series of interfaces such that expl3 hijacks the TikZ system layer entirely ...

@DavidCarlisle Perhaps I should cover that in my UK-TUG talk ...

@AlanMunn ^^ :)

@PauloCereda Apart from the fact that "Faculty Lounge" has no referent in the real world, (certainly not the one I inhabit) this is also a distinctly humanities kind of joke. (Although I'm not sure that in the humanities 'operational definition' is the norm.)

@PauloCereda Does that comment make me Reviewer number 2. :)

@AlanMunn IT WAS YOU

hackaday.io/project/27359-tiny-wearable-8-bit-vt100-console

@AlanMunn ooh

Bah it was no direct link. :(

@PauloCereda Not going to click through all the bait, but this is a big topic of study in the acquisition literature. Children's development of pragmatic knowledge takes a while. For example, young children are happy to say Yes to sentences like "Some elephants are mammals" (which adults reject), or in a situation where Sam was playing soccer and broke her leg and then went to the hospital, they will say yes to sentences like "Sam went to the hospital and broke her leg".

@AlanMunn Oh that's interesting!

@samcarter, @AlanMunn ^^ tihs was the image I wanted. :)

@AlanMunn If @DavidCarlisle is older than me, then I didn't drink a glass of wine on the train.

@PauloCereda That's beautiful. Names are completely arbitrary phonological strings, so why not!

@AlanMunn :D

@egreg This one I think is much more to do with the fact that natural language conditionals don't really work like logical conditionals. When you teach intro logic it's really hard to teach this.

@DavidCarlisle what was the italian workaround for '

@PauloCereda Too bad the square is not green, else we could check if the names are correct.

@samcarter ooh that's true :)

@AlanMunn: in other news (thesis-related), I am writing a sentence generator given a CSG grammar. :)

@PauloCereda However the given solution might be valid:

@samcarter Where's Alice? :)

Alice seems predominantly green

@PauloCereda And Charlie, Eve, and Mallory …?

@HaraldHanche-Olsen Yes. :)

@PauloCereda Only one Mallory here. He's light blue. Charlie and Eve more multicoloured.

@HaraldHanche-Olsen Interesting

@HaraldHanche-Olsen Mallory-s are typically women. :)

@PauloCereda To what end? And why context sensitive? Surely too powerful for Human.

@AlanMunn Given this grammar, I want to generate a parser too. :)

@AlanMunn Harmless experiments. :)

@PauloCereda :) Wasn't that what they said in Jurassic Park?

@AlanMunn That. :) And also Spared no expenses. :)

@DavidCarlisle nm, I think it was \lccode\'=\'

Back in 2012 I mentioned that if A=1, B=2, C=3, etc., then ARM + BEND = ELBOW and KING + CHAIR = THRONE.

Peter Dawyndt of Ghent University challenged his students to come up with more, and they found these:

WHITE (65) + HOUSE (68) = GOVERNMENT (133)
PETER (64) + PAN (31) = NEVERLAND (95)
COMIC (43) + BOOK (43) = FANTASY (86)
ABSENT (61) + MINDED (49) = FORGETFUL (110)
BLOOD (48) + BATH (31) = MASSACRE (79)
DRUG (50) + ADDICT (41) = STONER (91)
MICRO (58) + SOFT (60) = COMPUTING (118)
RED (27) + BULL (47) = COCKTAIL (74)

From futilitycloset.com/2017/10/24/more-word-sums

@PauloCereda I wonder how they came up with them. It would be interesting to do it using WordNet or equivalent.

@AlanMunn Indeed!

DAVID(40) P(16) CARLISLE(79) = MEAN(33) ENGLISHMAN(102)

@AlanMunn ^^

@PauloCereda LOL

@AlanMunn <3

@AlanMunn ^^ so close...

@PauloCereda Do you know if Greasemonkey is available for Chrome? I'm getting sick of FF (too buggy and slow). But I use Greasemonkey scripts and it doesn't show up when I search. Or is there an equivalent?

@AlanMunn I think Chrome has some sort of "default" support for it!

@PauloCereda Your test for glossaries? ;-)

@ChristianHupfer ooh :)

@PauloCereda Hmm. Ok I'll have to investigate further.

@AlanMunn I think you can install scripts directly.

@PauloCereda People seem to recommend 'Tampermonkey' for Chrome.

@AlanMunn Oh that one! I almost forgot!

you ducks are making so much noise

@alfred Ducks producke noice ;-) @PauloCereda :-P. That's a whacky/weird produckt ;-)

@ChristianHupfer LOL

@AlanMunn Indeed!

@alfred what was? (you need characters with non zero lccode in patterns but that is the same in miktex and texlive)

@DavidCarlisle I added \lccode\'=\' and it worked

@alfred yes that's why it was in that original zz.tex example I suggested but it couldn't have been working in miktex without that either

I think it should have but it caused an error earlier so I added it

trying texlive now