« first day (3544 days earlier)      last day (25 days later) » 

12:23 AM
ha, right, I hadn't thought about it, lol
thanks
but it would be nice to have a rank of the citations
I would like to know which paper I am citing the most
 
 
6 hours later…
6:11 AM
@DavidCarlisle One for Jonas?
 
6:24 AM
@DavidCarlisle about learnlatex.org/en/lesson-10 - maybe there should be something about the dreaded $\LARGE x$ thing there? It's a quite common error...
 
 
1 hour later…
7:27 AM
@nbro Use biblatex
51
A: Can I count the number of times each reference is cited?

lockstepIf you're willing to switch to biblatex, you may use the citecounter feature that was added in v1.3. \documentclass{article} \usepackage[citecounter=true]{biblatex} \renewcommand{\finentrypunct}{% \addperiod\space (Cited \arabic{citecounter}~time\ifnumequal{\value{citecounter}}{1}{}{s})% }...

5
Q: Using citecounter for sorting in biblatex

ThorstenI would like to define a \DeclareSortingScheme for biblatex such that \printbibliography sorts the citations according to the citecounter: Reference which is cited most in the document on top. Is there a way to do this. Everything I tried does not work.

 
7:39 AM
@JosephWright possibly although we can decide if we want it, it is more or less working at davidcarlisle.github.io/ll/en/lesson-10 (but the buttons aren't fully hooked up yet) ace comes from a CDN so no hosting required but of course it's a bigger payload for the user, but the syntax highlighting is better, you get latex environment folding, search/replace (control-h) line numbers etc. No edit or copy buttons needed
@Rmano possibly although other than "don't do that" not sure what we could say?
 
@DavidCarlisle what is ace?
@DavidCarlisle "less working" means that one currently get errors?
 
@DavidCarlisle :-) well, maybe that if you really need a big $x$ you can {\Huge $x$}...
 
8:01 AM
@UlrikeFischer ace is an online editor thing (I think it's what overleaf use as well) see ace.c9.io
@UlrikeFischer yes the buttons are still taking the text content (so see line numbers, and wouldn't see folded environments etc), need to change them to call the editor .value() function ran out of time last night, it'll be in place today at some point
 
@UlrikeFischer Indeed that line in mltex.ini was garbage: github.com/MiKTeX/miktex/issues/584
 
@DavidCarlisle the experience is quite good.
@ComFreek yes I saw your issue report yesterday.
 
ah :)
@MarcelKrüger What's the rationale behind luametalatex? (read the starred message)
 
@DavidCarlisle Looks good to me: go for it?
@ComFreek Well Hans et al. are creating LuaMetaTeX for the future of ConTeXt, stripping out anything from LuaTeX they don't need. As they actually do the engine support, seems sensible to have a working LaTeX format for that engine
 
8:16 AM
ah I guess, then my question is what the rationale of ConTeXt is
 
@JosephWright yes it took a bit of fiddling to get it integrated but I was quite happy with the way it looked, If it's all working on ll I may switch learnlatex this evening
 
@DavidCarlisle, @UlrikeFischer Editor looks good, I think having a 'standard' one is sensible, don't really see a downside
 
but that should certainly be mentioned on this site
 
@DavidCarlisle Cool
 
gotta search
 
8:17 AM
@DavidCarlisle You should be giving the talk, not me: a lot of what I say will be 'David did XXX'
@ComFreek ConTeXt users tend to stick to the mailing list, which some of us (@UlrikeFischer, @MarcelKrüger, me, for example) are on
 
@JosephWright nah I'm just backroom stuff:-)
 
@ComFreek You could post a question on the main site, probably @MarcelKrüger or Aditya would answer
@DavidCarlisle You seem to be enjoying it :)
 
@JosephWright lockdown hobby?
 
@DavidCarlisle I need to write my slides this week, expect some drafts (although they'll be pretty sparse)
@DavidCarlisle :)
@DavidCarlisle Timing has worked out for us
 
9:04 AM
quack
 
9:54 AM
@Skillmonlikestopanswers.xyz /duck hug
 
@MarcelKrüger I'm curious about part II ;-).
 
@UlrikeFischer ooh part 2
Germans are awesome
 
@PauloCereda it is a secret ;-)
 
@UlrikeFischer ooh a secret
We ducks are very good at keeping secrets
@UlrikeFischer is it a normal secret or a super secret? :)
 
@PauloCereda I found it by looking at a public commit message, so it is a super secret (nobody reads commit messages ...)
 
10:00 AM
@UlrikeFischer ooh maybe is for @MarcelKrüger's talk :)
I didn't prepare mine yet
oh no
/quacks in despair
 
@UlrikeFischer Which one?
@PauloCereda Ooh, BL
 
@JosephWright <3
@JosephWright ooh more secrets
@JosephWright BL is awesome! :)
 
@PauloCereda /rabbit bite ehh, I mean hug, /rabbit hug
 
@Skillmonlikestopanswers.xyz ooh
 
Active chars strike again: github.com/latex3/latex3/pull/758
 
10:11 AM
@JosephWright I'm not following, what's the question?
 
@MarcelKrüger @ComFreek was asking about the 'point' of LuaMetaLaTeX
 
10:22 AM
@JosephWright Oh that question. It's new, there are breaking changes with many releases which randomly change core TeX behaviour and it requires lots of custom Lua code. So what could possibly be more awesome to play with?
 
@MarcelKrüger :)
@MarcelKrüger Reminds me of LuaTeX development in the past
 
@MarcelKrüger ooh
LuaLuaMetaMetaLaTeX
3
ooh
 
@JosephWright It will be interesting to see how active LuaTeX development continues to be.
 
@MarcelKrüger True: probably much less so than it has been
@MarcelKrüger Were you following things when @DavidCarlisle and I kept reporting random breakages to Hans?
 
@JosephWright poking :)
 
10:29 AM
@JosephWright Not really, but it sounds similar to my recent interactions with him.
 
@MarcelKrüger :)
@MarcelKrüger This was when we were extending the test suite to LuaTeX (it used to be e-TeX-only): there were lots of 'things' we found
@MarcelKrüger I've never worked out what Hans does for testing
 
@JosephWright ooh findings
Ducks are good at finding things
@JosephWright what typical programmers do: one trivial test case -- it works for n = 1, then use the inductive step and thus it works for n = k, k + 1 and so on. :)
 
@JosephWright When I tried running the tests with LuaMetaLaTeX I also found quite some issues, but an even better test was just trying to run latex.ltx...
 
"Beware of bugs in the above code; I have only proved it correct, not tried it." -- DEK
 
@MarcelKrüger Sounds good to me
@MarcelKrüger We should get LuaMetaTeX into TL, then we can add it to Travis-CI
 
10:34 AM
@JosephWright Probably running some example ConTeXt document. There seems to be a correlation: If some functionality is used in ConTeXt, it is mostly stable. Otherwise, it is likely to be "more interesting" :)
@JosephWright Sadly there still is the whole licensing mess, I guess TL wouldn't be too happy about that.
 
@MarcelKrüger Ah,yes, that
@MarcelKrüger Quite
 
@MarcelKrüger oh :(
 
10:57 AM
@MarcelKrüger apart from being a back-up plan in case luatex support runs out, do you think it's a useful evolution of luatex or going in a wrong direction?
 
@DavidCarlisle use Word?
 
@PauloCereda better than vim
 
@DavidCarlisle bah
 
@DavidCarlisle you mean like Emacs went into the wrong direction just like Windows in becoming a bloated unusable OS?
 
@Skillmonlikestopanswers.xyz your English is failing you, I can't understand that comment at all, sorry.
 
11:02 AM
@DavidCarlisle I doubt that :)
 
@UlrikeFischer I think I'll be able to add CIE-Lab support today
 
@DavidCarlisle I think it's a good next step. The more LuaTeX's features are used, the more code does stuff similar to the backend anyway, so moving the backend into Lua is not that much mmore work and decouples it form engine upgrades. Also it allows to implement features there without needing agreement across formats beforehand, allowing for much more experiments than having a fixed binary. Especially this allows lower-level changes should they be required for accessibility.
3
 
@MarcelKrüger <3
 
@MarcelKrüger I wonder how Hans will react if LuaMetaLaTeX takes off!
 
@JosephWright copyright strike hmmm
How about a metal, steampunk-like moon for a logo?
Meta = metal (close), moon = lua. :)
Something along these lines. :)
 
11:08 AM
@DavidCarlisle the first tries I made looked quite fast compared to lualatex.
@JosephWright create luametametatex?
 
@PauloCereda only thing missing is the "aTeX" from the name.
 
@Skillmonlikestopanswers.xyz ooh
 
@UlrikeFischer I've not tried yet, but isn't it lacking HarfBuzz?
 
@JosephWright yes, but Marcel said it should be possible to add it.
 
@JosephWright Hardcode a test which breaks the binary if \documentclass is defined?
 
11:10 AM
@UlrikeFischer Cool
@MarcelKrüger Ah, yes, could be
@MarcelKrüger Hans is normally quite laid-back, he'll just say 'people get what they get', as for ConTeXt he can change the engine and the macros/Lua in sync
 
@JosephWright luaotfload dev
 
@JosephWright That's related to the secret project Ulrike mentioned: The latest release of LuaMetaLaTeX has HarfBuzz on Linux (My Windows dll doesn't work and I gave up), the luaotfload code is almost ready too.
 
@MarcelKrüger Ah, right
 
@MarcelKrüger I want to look at the breaking luaotfload tests in the afternoon. I guess the change to ^^^ notation like ! ....\TU/LatinModernMath(1)/m/n/10 ^^^^^^01D465 in some cases is harmless?
 
@MarcelKrüger Secret Plans, heh :)
 
11:12 AM
@MarcelKrüger ooh secret
@JosephWright ooh more secret
 
@UlrikeFischer That should be fixed now, I forgot to add a 0.
@JosephWright Also there are indications that HarfBuzz support might get into upstream LuaMetaTeX's optional extension mechanism. Then the only needed support code would be a native HarfBuzz library.
 
@MarcelKrüger Ah, that would be good
 
11:29 AM
@MarcelKrüger where did you see that?
 
@UlrikeFischer ooh more secrets
 
11:48 AM
@UlrikeFischer The Context documentation about that mechanism has a commented out line listing HarfBuzz with the comment maybe, for idris, testing uniscribe
@JosephWright Well we need a separate library for kpathsea anyway, so it doesn't make that much of a difference if that also provides HarfBuzz. (Actually the latest luaetatex release did add kpathsea support in the optional system, but they only added a small subset which isn't sufficient for us. I would take that as indication that they specifically decided against full support.)
 
Best regards to all users into chat.
@egreg Very kind egreg, good morning, I have seen your message and also that of Peter Grill. I have put the screenshots of Papeeria TeXlive 2019 where I compile with another pc where I have not installed MikTeX 2.9. I not get any errors always with a lot of sincerity.
 
12:07 PM
@MarcelKrüger testing uniscribe? Isn't that windows specific?
 
12:20 PM
Yes but it kind of makes sense to use it for testing because it's kind of a reference implementation of OpenType.
 
@DavidPurton hi mr. koala!
 
12:34 PM
@MarcelKrüger travis now fails because the status is different (256 instead o 0):
! Compilation 1 of test file completed with exit status 256
 
12:52 PM
@MarcelKrüger I think it was because latex-amsmath-dev was missing. Checking this now.
 
@UlrikeFischer It complains about \negmedskip already being defined.
 
@MarcelKrüger yes, because of the wrong amsmath version.
@MarcelKrüger it passed that test and now fails one that failed locally too. This failure are all because some 0 box is gone, but imho that is ok and I will change the test files. E.g.
***** .\BUILD\TEST-CONFIG-UNICODE-MATH\operatorname.tlg
   19:  ..\kern0.0
   20:  ..\hbox(0.0+0.0)x0.0, direction TLT
   21:  ..\TU/lmtt/m/n/10 s
***** .\BUILD\TEST-CONFIG-UNICODE-MATH\OPERATORNAME.LUATEX.LOG
   19:  ..\kern0.0
   20:  ..\TU/lmtt/m/n/10 s
*****
 
@UlrikeFischer Yes, that looks like some amsmath or unicode-math change.
 
@MarcelKrüger I will push in a few minutes. There is also a small patch in database.lua for windows (from Akira).
 
1:14 PM
@MarcelKrüger build passed ;-).
 
 
1 hour later…
2:15 PM
@PauloCereda Oops, I was only on briefly. Evening!
 
@DavidPurton hi :)
 
2:34 PM
@nbro On counting citations: I'm curious why you think this is a useful thing to do. What would the ranking tell you and what will you do with the information?
 
2:50 PM
@AlanMunn the ranking would tell me which source my work is most based on
at least roughly
I asked a more specific and concrete question here.
0
Q: Is it possible to automatically generate a rank of the citations?

nbroI would like to know which papers or books I am citing the most. So, is it possible to automatically generate a rank of the citations? (i.e. automatically count the number of times all references are cited and then generate an ordered list)? I know I could use my editor to find the number of time...

 
@nbro Yes, I read the question, but I'm asking why you think that information will be useful to you.
 
@AlanMunn Well, that will tell me, in a certain sense, the similarity between my work and the work of others
 
@nbro My experience is that some things get cited a lot because they are foundational/important to a particular area/problem, but this doesn't necessarily mean they are important to the specific paper you're writing, which usually is connected to some more specific problem.
@nbro Similarity in what sense?
 
Similarity in the sense of relatedness
Sometimes I not only cite papers that introduced a certain concept but also papers that support my explanations
 
3:09 PM
@nbro Well to each their own. :) I'm not sure I see the value in this as a measure.
 
well, this will also tell me how important is a certain paper/book for my work
 
Customer: Do you have any two-watt, 4-volt bulbs?
Sales Rep: For what?
Customer: No, two.
Sales Rep: Two what?
Customer: Yes.
Sales Rep: No.
 
@PauloCereda In some varieties of English this wouldn't work: 'what' and 'watt' are not pronounced the same. :)
 
@AlanMunn ooh like in the "I never drink wine" line?
:)
 
@nbro Surely you know this without having to count?
@PauloCereda ?
 
3:16 PM
@AlanMunn more or less, yes, I have a rough idea, but I would like to have the ranking to be more precise, let's say :)
 
@AlanMunn YAPM sorry
 
@PauloCereda :)
 
@AlanMunn Bela Lugosi's line in the Dracula movie... sorry. :)
 
This is not extremely important, but it's something interesting in terms of statistics
I could actually write a script that parses my latex file, but maybe someone else has a better idea and already knows how to do it. I would like to avoid doing that by hand
 
@PauloCereda do you know the two ronnies 4 candles sketch?
 
3:20 PM
@DavidCarlisle no...
 
@nbro I always suspect when people ask for this that they only want to be sure that they cited their professor enough times.
 
@DavidCarlisle ooh
 
@PauloCereda one of the better ones:-)
 
@UlrikeFischer Anything less that 20 is simply unacceptable.
 
3:23 PM
@DavidCarlisle oh my
 
@UlrikeFischer ahah, no, actually, I have not cited my ex professors yet. I will eventually do it for another part of my work
 
@UlrikeFischer LOL
 
@DavidCarlisle I need subtitles ;-(
 
The Compiler Language With No Pronounceable Acronym, abbreviated INTERCAL, is an esoteric programming language that was created as a parody by Don Woods and James M. Lyon, two Princeton University students, in 1972. It satirizes aspects of the various programming languages at the time, as well as the proliferation of proposed language constructs and notations in the 1960s. There are two maintained implementations of INTERCAL dialects: C-INTERCAL, maintained by Eric S. Raymond, and CLC-INTERCAL, maintained by Claudio Calvelli. As of May 2020, both implementations were available in the Debian Software...
This part is gold:
INTERCAL has many other features designed to make it even more aesthetically unpleasing to the programmer: it uses statements such as "READ OUT", "IGNORE", "FORGET", and modifiers such as "PLEASE". This last keyword provides two reasons for the program's rejection by the compiler: if "PLEASE" does not appear often enough, the program is considered insufficiently polite, and the error message says this; if too often, the program could be rejected as excessively polite. Although this feature existed in the original INTERCAL compiler, it was undocumented.
Also, @DavidCarlisle ^^ undocumented features
 
@UlrikeFischer Slight holdup on LAB colors: I realise I've not quite got the model right
 
3:28 PM
@JosephWright should I look at something?
 
@UlrikeFischer No, not yet: I've done some stuff, but it's not quite ready and I want to rework it a bit, likely tomorrow
 
@JosephWright fine. Bruno seems to have time - I should ask for a bitset implementation ;-).
 
@UlrikeFischer Bitset is on my list, it's a question of how much we need
 
@JosephWright I only need (apart the commands to create and add/remove a flag) a sum output (\bitsetGetDec). I also want to address the fields by names, but that can be done with properties. And my bitsets are short - no need for large integers.
 
3:48 PM
@UlrikeFischer I added a luaotfload PR for no longer normalizing symlink paths. While I thik that it makes sense, it has a certain potential to introduce subtle differences in existing behavior. Could you look over it if I missed anything?
 
3:59 PM
@nbro -- Why don't you extract the "cite" references from the .aux files, sort and count them? (Not done directly by LaTeX, but more compact to deal with.)
 
@MarcelKrüger hm, well we windows user don't really understand symlinks ;-) (I ignored your instruction to symlink the luametatex binary and created a copy ...). But it sounds okay.
 
cis
4:45 PM
I need a letter like this.

Mmmh, shall I create the gridded parts also as tcolorboxes or better plain-TikZ?
 
@JosephWright @UlrikeFischer ace editor working on ll now for overleaf and latexcgi davidcarlisle.github.io/ll/en/lesson-07
 
@DavidCarlisle That's looking great! (also no 'edit' and 'copy' buttons :-)
 
@DavidCarlisle looks nice. Did the colors change from yesterday?
 
@PhelypeOleinik yes I needed those to get rid of the github syntax highlighting as they interfere with editing, but using ace can have syntax highlighting while editing
 
@DavidCarlisle I never liked them :-)
 
4:53 PM
@UlrikeFischer not as far as I now. The colours are a bit odd math commands are red which looks like an error but that's just the defaults, if I knew how to read the documentation they are probably changeable.
 
@DavidCarlisle (though I get why they were needed)
 
@PhelypeOleinik since anyone who looks seems to prefer this version I'll move it over to learnlatex.org this evening
 
@DavidCarlisle ah, that's why I had red in my mind.
 
@DavidCarlisle Also, there's an extra m in “produce dummmy text” (a comment in the "Making images float" example)
 
@PhelypeOleinik should be replaced by a z ;-)
 
4:56 PM
@PhelypeOleinik sorry. @UlrikeFischer makes so many typos
 
@DavidCarlisle tehhehehe!
 
@DavidCarlisle Of courze :-)
 
@PhelypeOleinik she wouldn't make that one, no one can find z on a German keyboard
@PhelypeOleinik fixed thanks
 
5:33 PM
@UlrikeFischer I'll look at that once color stuff is done
 
@JosephWright it is not really pressing. Actually the only thing that currently bothers me here, is the name of the commands ;-). I currently use \pdf_flag_new:nn as wrapper around bitset and this is simply wrong, as it is not a pdf specific concept. How would we call a module for this?
 
@UlrikeFischer Well it is bit or bitflag or some such
@UlrikeFischer Team list for a module name, then I can start on the features
@DavidCarlisle Looks pretty good
 
6:45 PM
@JosephWright but today's top tip: if you want to see how to improve the highlighting don't google for "ace latex theme"
 
cis
0
Q: tcolorbox vs. TikZ vs. geometry: How to put a proper grid below a tcolorbox

cisOn a DINA4 (210 × 297 [mm]) paper I want to create the following: I need ca. 2cm margins to the left and the right. I need a tcolorbox with a total height of 6cm. Below of that tcolorbox I want a grid of 2 × 2 [cm] with as much as possible squares in the width. So I choose \usepackage[ w...

....
I mean, my explanation graphics with the meaningful evil smiley are simply incredible.
 
7:12 PM
@JosephWright made PR for the record. I think I copied everything over: if something is missing when its live in gh-pages I'll clean up. Help page will need some edits as well, I'll do that in place when it's live
 
 
2 hours later…
8:46 PM
Any idea why this item of an enumerate goes outside of the right margin? This only happens for this item and for the word "parametrized". If I remove the word, that disappears. Also, note that the second sentence is like cut
 
@nbro if you are using the default US english patterns tex can not hyphenate parametrized
@nbro parameterized is normal spelling (or s if you are here)
 
oh, that was really the problem
 
*\showhyphens{parametrized parameterized}

Underfull \hbox (badness 10000) detected at line 0
[] \tenrm parametrized pa-ram-e-ter-ized
 
but the second sentence is still cut
don't know why
 
@nbro cut?
 
8:54 PM
if you look at the second sentence, the top of the letters is cut
 
@nbro tex can't do that must be an artefact of your pdf viewer. what fonts are you using, are they bitmap?
 
I am using a custom documentclass that derives from book. I need to check it
 
@nbro just put \expandafter\show\the\font on that line and say what message you get
 
you were right I think. I recompiled and re-opened the pdf viewer and the issue went away
 
9:11 PM
Anyway, it's weird that latex is not able to deal with "parametrized" or "parametrised"
 
10:05 PM
@nbro why is that strange? neither match US hyphenation patterns
@nbro some dictionaries give it as an alternative spelling, if you want to add a hyphenation exception just add \hyphenation{pa-ram-etr-ized} somewhere
 
@samcarter_is_at_topanswers.xyz thanks!!
I don't know why 2. More text appears 3 times instead of 1 in this MWE:
\documentclass{beamer}

\begin{document}

\begin{frame}
\begin{block}{}
Text here

\begin{onlyenv}<2-4>
\begin{enumerate}[<+->]
\item Text
\pause
\begin{enumerate}[<+->]
\item Some text
\item More text
\end{enumerate}
\end{enumerate}
\end{onlyenv}

\begin{onlyenv}<5-17>
\begin{enumerate}[<+->]
\setcounter{enumi}{1}
\item More text
\begin{columns}[T]
\begin{column}{.5\linewidth}
\begin{itemize}[<+->]
\item $n=0$:

\uncover<+->{
Some graph here
}

\uncover<+->{$x_0$.}
\end{itemize}
\end{column}
\begin{column}{.5\linewidth}
Moreover, each itemize appears more than once. Why is this happening and how can we prevent it?
\documentclass{beamer}

\begin{document}

\begin{frame}
\begin{block}{}
Text here

\begin{onlyenv}<2-4>
    \begin{enumerate}[<+->]
        \item Text
        \pause
        \begin{enumerate}[<+->]
            \item Some text
            \item More text
        \end{enumerate}
    \end{enumerate}
\end{onlyenv}

\begin{onlyenv}<5-26>
    \begin{enumerate}[<+->]
        \setcounter{enumi}{1}
        \item More text

        \begin{columns}[T]
            \begin{column}{.5\linewidth}
                \begin{itemize}[<+->]
 
10:22 PM
@DavidCarlisle well, I thought one could also write parametrized. I don't really know why
Isn't parametrized proper English?
 
@nbro as I say some dictionaries have it as an alternative spelling so i guess it's proper but uncommon American English. But the hyphenation patterns are just patterns that can not catch all words, that's why \hyphenation command exists to add any more that you need.
 
10:48 PM
@nbro I've never seen it, but apparently it does occur, although less frequently than parameterized (which is twice as frequent).
 
11:30 PM
@DavidCarlisle -- You can't be serious about "pa-ram-etr-ized"??? Is that really what you get with the UK patterns? (I'd guess "pa-ra-me-trized"; the "-etr-" is really beyond belief, and it's not according to etymology.)
 
11:58 PM
@JosephWright -- I have just mistakenly voted to close tex.stackexchange.com/q/553460 as a duplicate. I believe it is a duplicate, but not to the proposed question. I was attempting to propose tex.stackexchange.com/a/308980 but the system "grabbed" the vote without allowing me to enter the change. My vote has been challenged in a comment that I can't read (laptop problems). Can you fix this, or at least remove my close vote, please?
 

« first day (3544 days earlier)      last day (25 days later) »