« first day (2832 days earlier)      last day (2106 days later) » 

3:06 AM
@samcarter This happens when you replace in your super tikzmarmots package all \fill[ by \shade[ball color= :
user image
2
I find it actually pretty amazing... ;-)
Applied to the ducks one gets
Not quite as impressive, but I'm also biased ....
 
4:12 AM
@user2236 rip
@marmot those animals are more softer than the others <3
 
@manooooh :-)
 
4:46 AM
Hello @CarLaTeX
How are you?
 
@manooooh Hallo, fine thanks!
 
Gooooooooooooood
I'm afraid we can not talk today because I have to do some homework
Don't miss me
 
@manooooh OK, we'll talk another day :) bye!
three ducks in chat :)
 
Yes
 
4:52 AM
All of them in differents parts of the world
Hi
Ouch, bye*
@CarLaTeX do you use Opencast?
Well, I think I'm clogging
Everyone have a nice day
 
@manooooh No, I don't know what it is, have a good night!
 
@Mu30murugans2katgmail he'll read the message tomorrow :)
 
5:18 AM
sigh — xepersian.sty uses internals from fontspec, and I just changed some of the internals of fontspec, so xepersian no longer works :(
And xepersian doesn't appear to have a github repo any more, so there's nowhere to log the issue
@marmot I think just shading the larger shapes might look even better!
 
6:02 AM
@WillRobertson Karl set up a persian-tex list at tug.org for Vafa I think?
 
@WillRobertson Could it be that gitlab.com/persian-tex/xepersian is the new home?
 
@marmot all I need is a pin and BANG!
 
6:19 AM
@TeXnician ahhhhh looks like it, yes. thanks!
 
 
2 hours later…
7:58 AM
@marmot You're new marmot looks gorgeous!
 
 
1 hour later…
9:15 AM
user image
2
 
@samcarter @DavidCarlisle will want more pins.
@samcarter yesterday I saw my university in the icm video ;-)
 
@UlrikeFischer The plan is to distract @DavidCarlisle with the balloon while the marmot survives :)
@UlrikeFischer Oh! The building -- especially the spiral staircase -- looked nice!
 
@samcarter You could add more balloons with markers like a bull's eye.
@samcarter yes, it is a quite beautiful university, but the mathematicians aren't actually in the main building (at least they weren't when I studied there).
 
9:30 AM
@UlrikeFischer Curious. Which one? Bonn? ETH?
 
@mickep Bonn.
 
@UlrikeFischer Cool. Were you there at the same time as Scholze?
 
@mickep He is bit younger than me ;-).
 
@UlrikeFischer I see. I only had an upper bound for him, no info on your age :)
 
@mickep well you could get the info if you look up my profile ;-).
 
9:46 AM
@UlrikeFischer Oh, I'm obviously too lazy. Sorry. I'm right now a bit fascinated that one can use modes in ConTeXt to use the same file to compile to screen/paper. It is just very cool. (Just to get ontopic a bit)
 
yo'
10:45 AM
@PauloCereda and his dad are sure that by now the medal is lost forever as it's probably melted (it's 14k gold)
 
11:25 AM
user image
3
@marmot ^^^
 
11:50 AM
@samcarter Oh WOW! ;-)
 
yo'
12:17 PM
Do you know whether someone was taking photos at TUG? @PauloCereda @JosephWright @UlrikeFischer
 
12:39 PM
@yo' I did sent Paulo a message.
 
yo'
@UlrikeFischer ok, I'll tell him :-)
 
@samcarter nice ;-)
 
1:16 PM
@samcarter How did you achieve this nice duck?
 
1:45 PM
@UlrikeFischer I hope no pin or sharp fork will come near it :)
@marmot I'm afraid the code is not so nice: github.com/samcarter8/tikzducks/blob/master/duckpond/3Dduck.tex
 
@samcarter It looks quite solid, it probably has a (david-)pin repellent coating.
 
@samcarter I think that it looks really good! (except of course ballshading should be replaced by duckshading ;-)
 
2:02 PM
@UlrikeFischer Yay, a tagpdf answer!
 
@JosephWright yes and quite surprisingly not directly related to "accessibility" ;-) And I will have to investigate the hyphenation char (hard/soft) business. With lualatex it seems to work out-of-the-box.
 
@UlrikeFischer :)
@UlrikeFischer I'm looking at the uncompressed PDF stream it gives ...
 
@JosephWright the output of pdflatex is easier, you normally see some words and not only hex numbers ;-).
 
@UlrikeFischer Ah right
@UlrikeFischer Yup: probably I'll need more normalisation for LuaTeX to strip out binary data. Probably will go for 'check each line for chars <32'
 
@JosephWright you could probably do something like "if there is a char <32" strip everything from the stream before to the next endstream.
 
2:15 PM
@UlrikeFischer Ah yes, a reasonable logic. I can collect up each stream and examine it, then dump either the stream itself or a marker, say [BINARY STREAM], so it's clear what's happened. I'll look to adjust that before a release. (I'm guessing everyone is happy enough with the PDF testing as-is.)
 
Is "kindergarten-style" considered "not nice"?
 
@JosephWright looks as if one needs a "soft hyphen char" with lualatex too, e.g. \prehyphenchar\^^ad` so that real hyphens and hyphens from hyphenation can be handled differently.
 
@MdAyquassar -- it usually implies "dumbed down", but depends on the context.
 
2:31 PM
@barbarabeeton The poster of such an expression apparently took my downvote personally, so, various expressions of similar type appear ...
 
@MdAyquassar -- sigh ...
 
3:05 PM
@barbarabeeton Luckily, the OP (or the mods) have removed the comment. Whoever did it, thats for keeping the language professional.
 
3:47 PM
@samcarter Please undelete your nice answer! (I'm happy to delete mine.)
 
@marmot Please keep yours, it is much better as it points out all the problems with ... etc.
 
@samcarter You could simply add these to yours. (I am reputation capped anyway ;-)
 
@marmot Undeleting would destroy the 42 with which my reputation currently ends. You surly understand that I cannot do that :)
 
@samcarter Soon enough it will hit another 42 ;-)
 
@marmot An unfortunate timing of an accepted answer could easily prevent this. I'll play safe and stay with this 42
 
3:58 PM
@samcarter Well, you'll need that for 77777 anyway, don't you?
 
@marmot There is plenty of time to solve this problem until then. You probably will have to worry first how to switch to a 5/0 cycle if you want to hit 55555
 
@samcarter Hmm, yes, got a downvote from someone who just copied my answer... maybe that happens again, that would fix it ... there is a lot of time for that to happen ...
 
@marmot Oh no! That's not very nice!
 
4:15 PM
@samcarter On the bright side, the fraction of such users is very small.... (Soon the 30 day period will be over and my logo can go 3d ;-)
 
@marmot I'm looking forward to your new avatar! Btw: my prediction for 55555 is in 40.5 days
 
@samcarter What do you mean by "your"? It is yours... ;-) (Any chance the 3D ducks and marmots will go to the packages? ;-)
 
4:31 PM
@marmot For the marmot yes, the ducks needs some simplification (using fadings instead of shadings will probably help). However there is a fundamental problem with adding it to the package: I cannot think of a good name for the keyword that does not contain numbers.
 
@samcarter why do you have a problem with numbers?
 
@UlrikeFischer I normally use the same word to create a newif (to make it easier to remember it) and \newif\ifmarmot@3D fails
 
@samcarter You could change catcodes for a while ;)
 
4:46 PM
@samcarter you could start with expl3 ;-) Then you could use \bool_new:c{g__tikzmarmot_3D_bool}.
 
@UlrikeFischer expl3 and TikZ? That may cause some trouble…
 
@TeXnician You need to learn to use ~ but beside this it normally works okay.
 
@samcarter Hmmh, I would just have set a style that differs depending on whether or not 3d is chosen.
@samcarter You could do something like \tikzset{marmotfill={shade,ball color=#1}} in the 3D case and \tikzset{marmotfill={fill=#1}} in the 2D case and just use \path[marmotfill=\marmot@body], say, to do the filling.
 
5:15 PM
@marmot I'm afraid I don't really now how to switch between the two tikzsets. Would you be so kind to make a example based on the following code?
\documentclass{standalone}

\RequirePackage{tikz}

\makeatletter
\newcommand*{\marmot}[1][]{%
	\begin{scope}%
		\tikzset{/marmot/.cd,#1}%
		\marmot@draw%
	\end{scope}%
}

\newif\ifmarmot@threeD

\tikzset{
  /marmot/.search also={/tikz,/pgf},
  /marmot/.cd,
  3D/.code			 = \marmot@threeDtrue,
}

\def\marmot@draw{%
	\ifmarmot@threeD
		\shade[ball color=brown]
	\else
		\fill[brown]
	\fi
	(1.18,0.2) ellipse (0.23 and 0.09);
}
\makeatother

\begin{document}

\begin{tikzpicture}
\marmot
\end{tikzpicture}
@marmot another problem with \path[fill= is that it does not work with colour models, see tex.stackexchange.com/questions/385158/…
@UlrikeFischer I know that I say this a lot, but learning expl3 could really be useful... I'll have to move this upwards on my todo list.
 
5:32 PM
@samcarter I was thinking about something of the following type:
\documentclass{standalone}

\RequirePackage{tikz}

\makeatletter
\newcommand*{\marmot}[1][]{%
\begin{scope}%
\tikzset{/marmot/.cd,#1}%
\marmot@draw%
\end{scope}%
}

\newif\ifmarmot@threeD

\tikzset{
/marmot/.search also={/tikz,/pgf},
/marmot/.cd,
3D/.code = \marmot@threeDtrue,
}



\def\marmot@draw{%
\ifmarmot@threeD
\tikzset{marmot@style/.style={shade,ball color=##1}}
\else
\tikzset{marmot@style/.style={fill=##1}}
\fi

\path[marmot@style=brown] (1.18,0.2) ellipse (0.23 and 0.09);
@samcarter But of course if there are problems with the colors, then this might not be very good. On the other hand, this might simply mean that one cannot change the color model for 3D marmots, not necessarily a problem, who would really want to draw a 3D pink marmot? ;-)
 
@marmot Thanks a lot! I'll have a closer look later this evening -- will be offline for a bit
 
 
1 hour later…
6:42 PM
@marmot So where is the code for the \marmot macro?
 
@FaheemMitha In the tikzmarmots package, which is part of the most recent TeXLive distributions (and probably other distributions as well)
@FaheemMitha Or do you mean the 3D marmot?
 
@marmot You're kidding.
@marmot I was referring to the code in your profile.
No apparently, you weren't kidding.
 
@FaheemMitha The code in my profile works when you have @samcarter stellar tikzmarmots package installed. ;-)
 
@marmot I see.
I missed the \usepackage line, sorry.
 
7:06 PM
@marmot I cannot understand your reservations against a pink 3D marmot :)
 
@samcarter This one is purple. ;-)
 
@marmot \marmot[body=violet!50!pink, 3D,whiskers,teeth] -> we are both correct :)
@marmot The quick and dirty hack is in github.com/samcarter8/tikzmarmots I'll have to find a better way to differentiate between the flat and 3D version sometime later.
 
@samcarter Please let me know if I can do anything. (You do not seem to like the style trick...)
 
@marmot I like the style trick very much, it is really elegant -- it juts does not seem to like \selectcolormodel{gray}...
 
@samcarter I am probably just stupid, but where is the difference? It seems that \fill or \shade all have problems. Or are you saying that using styles is particularly problematic?
 
7:18 PM
@marmot you need an l3drawmarmots instead of tikzmarmots to get more consistent colour handling (and someone closer to hand to blame if it doesn't work)
 
@DavidCarlisle Ho ho
@DavidCarlisle Enjoy the package?
 
@JosephWright true though isn't it? (especially the bit in brackets)
 
@DavidCarlisle :)
 
@JosephWright Is the package out? Where can I find it?
 
@marmot I meant a physical package: @DavidCarlisle should have had a delivery
 
7:20 PM
@JosephWright not tried it yet, been a bit distracted.
 
@DavidCarlisle Sure
 
@JosephWright the one you blamed frank for, or do you mean another one?
 
@DavidCarlisle The one that's Frank's fault :)
 
yep that arrived.
 
@UlrikeFischer Minor adjustment to the PDF testing approach to be checked in shortly: would a release of l3build be useful for tagpdf?
 
7:23 PM
@JosephWright well I would certainly like to try (also the exclude tests system).
 
@JosephWright Found it. Seems to be a really great package, none of the example pics in the manual has a flaw. ;-)
 
@marmot \fill and \shade work fine, the problem only occurs for the syntax fill=<color>. See tex.stackexchange.com/a/385169/36296 for some explanation by @UlrikeFischer.
 
@JosephWright see ^^
 
@samcarter I see. Didn't appreciate the details.... ;-) (I thought that \fill[color] and \path[fill=color] are equivalent but this seems to be wrong...)
 
hey all . . . i just pasted the "busy beaver" example in the tikz manual into a beamer frame, and i'm getting a ton of errors . . . i have no idea why . . . anyone care to help me debug this?
 
7:28 PM
@marmot :) I thought the very same thing
 
all the code for the busy beaver example is on p. 93 of this document: texample.net/media/pgf/builds/pgfmanualCVS2012-11-04.pdf
 
@dbliss does \begin{frame}[fragile] help?
 
@samcarter just tried that, and no
i'll try again
 
@UlrikeFischer OK, will ship to CTAN shortly
 
@dbliss Could you make a minimal working example (MWE)?
 
7:33 PM
@dbliss That's a very old version of the manual
 
@JosephWright good call, thanks . . . looking for a similar example in a newer version.
 
Should we ask for clarification in tex.stackexchange.com/questions/444348/…?
 
\documentclass{beamer}

\usepackage{tikz}
\usetikzlibrary{arrows}
\usetikzlibrary{automata}
\usetikzlibrary{positioning}
\usetikzlibrary{shadows}

\begin{document}

\begin{frame}
\begin{tikzpicture}[->,>=stealth',shorten >=1pt,auto,node distance=2.8cm,on grid,semithick,
                    every state/.style={fill=red,draw=none,circular drop shadow,text=white}]

  \node[initial,state] (A)                    {$q_a$};
  \node[state]         (B) [above right=of A] {$q_b$};
  \node[state]         (D) [below right=of A] {$q_d$};
@dbliss The above code works fine for me
 
@samcarter Works for me, too. But tell me, why are ducks round and red, and have labels like "q_c" on them? ;-)
 
@samcarter works for me as well . . . turns out all i needed was to know which tikz libraries were needed (they weren't in the example)
thank you
 
7:47 PM
@dbliss Congratulations, you just discovered one of the biggest challenges of the tikz user guide: knowing which libraries to load :)
4
 
@samcarter Just was about to say the same. ;-)
 
is there a way to change the text "start" to something else?
and to have an arrow with different text point into another circle?
 
@marmot These are very hungry ducks with their bills opened so wide that only the inside is visible.
 
@samcarter Yup: really there is no reason Till et al. couldn't just streamline them all into one big file and load it (@DavidCarlisle)
 
hmm . . . maybe hacking this automaton to be something else is not the best way to go . . .
 
7:50 PM
@dbliss initial text=duck.
 
@dbliss Yes: ` \node[initial,state,initial text=marmot] (A) {$q_a$};`
 
@JosephWright or an option "most" like tcolorbox.
 
@JosephWright pst-all ....
 
@JosephWright There is. Without this the reputation of the TikZy users would be 20% lower. ;-)
 
@UlrikeFischer That would be incredible helpful!
 
7:55 PM
@samcarter I guess it would not be very complicated to write a library that loads all others...
 
@marmot No, but getting the maintainers of tikz to make an update to include it would be much more difficult.
 
@marmot what about have the arrow come in from the other side?
what about have two "start" arrows come into the same node, from slightly different angles?
 
@samcarter you can write your own tikzlibrarymost.code.tex, as you could write your tikzducks library. It doesn't need to be in pgf.
 
@samcarter In principle one only has to see what the pgfmanual.tex does (I did that some time ago to see why the pics in the pgfmanual seem to work but not here. Of course, I could not solve the puzzle.)
@dbliss What is "the" arrow? There are several?
 
@dbliss Quick hack: \draw[->] (-1,1) -- (A); \node at (-1.2,1.2) {bla};
 
8:02 PM
thank you, this is very helpful
i see that the option [loop above] to edge works only for self-connections
what if i want an edge from, say, (A) to (B) to be curved?
 
@dbliss \draw (A) to[out=20,in=160] (B);
or \draw (A) to[bend left=20] (B);
 
ah, but \draw doesn't take a text label the way \path does . . .
(A) edge [bend left=20] node {$\omega_{EE}$} (B)
there we go
in path
 
I think it should be (A) edge [bend left=20] (B) node {$\omega_{EE}$}
 
ok
for drawing an arrow . . . how do you get it to target the right side of the node instead of the left?
i see . . . B.east
ok, here's my finished code, thank you all:
\begin{center}
\begin{tikzpicture}[->,>=stealth',shorten >=1pt,auto,node distance=2cm,
on grid,semithick,
every state/.style={fill=red,draw=none,text=white}]
\node[initial,state, initial text=$I_{ext, E}$] (A) {E};
\node[state, fill=blue] (B) [right=of A] {I};
\path (A) edge [loop above] node {$\eta\omega_{EE}$} (A)
(A) edge [bend left=20] node {$\eta\omega_{IE}$} (B)
(B) edge [bend left=20] node {$\omega_{EI}$} (A)
(B) edge [loop above] node {$\omega_{II}$} (B);
\draw[->] (0,-1) -- (A); \node at (0,-1.2) {$I_{stim}$};
 
8:19 PM
@UlrikeFischer mmm, I assume this would be better than nothing - although not as convenient as \usepackage[most]{tikz}....
 
@samcarter @UlrikeFischer These are the libraries that the pgfmanual loads:
\usetikzlibrary{
arrows,
arrows.spaced,
arrows.meta,
bending,
babel,
calc,
fit,
patterns,
plotmarks,
shapes.geometric,
shapes.misc,
shapes.symbols,
shapes.arrows,
shapes.callouts,
shapes.multipart,
shapes.gates.logic.US,
shapes.gates.logic.IEC,
circuits.logic.US,
circuits.logic.IEC,
circuits.logic.CDH,
circuits.ee.IEC,
datavisualization,
datavisualization.polar,
datavisualization.formats.functions,
er,
automata,
backgrounds,
@samcarter Howe about
\documentclass{standalone}

\RequirePackage{tikz}

\makeatletter
\newcommand*{\marmot}[1][]{%
\begin{scope}%
\tikzset{/marmot/.cd,#1}%
\marmot@draw%
\end{scope}%
}

\newif\ifmarmot@threeD

\tikzset{
/marmot/.search also={/tikz,/pgf},
/marmot/.cd,
3D/.code = \marmot@threeDtrue,
}



\def\marmot@draw{%
\ifmarmot@threeD
\def\marmot@part@draw##1{\shade[ball color=##1]}
\else
\def\marmot@part@draw##1{\fill[##1]}
\fi

\marmot@part@draw{brown} (1.18,0.2) ellipse (0.23 and 0.09);
 
@marmot Oh, that looks very good! Let me test with the full package and I'll report back.
 
8:45 PM
@marmot That works great! Thank you so much!
 
@samcarter No thank you!
 
Hey I have a quick question about biblatex
can someone help me to put a comma between two authors instead of and
I can't find the question in tex.stackexchange
 

« first day (2832 days earlier)      last day (2106 days later) »