« first day (4884 days earlier)      last day (66 days later) » 

8:41 AM
@JosephWright ooh
@JosephWright why don't you change paths in the terminal like every wacky duck :)
 
@PauloCereda Doesn't work for macOS GUIs
 
9:15 AM
@PauloCereda why change paths when you can just make sure the paths you set up ages ago uses a placeholder instead of the year part of the path, then make that placeholder a soft link and you only need to change the softlink (that is presumably what the MacTeX gui does behind the scenes)
 
9:37 AM
@daleif I was just kidding. :) I have /opt/texbin which is a symlink to my TeX installs.
➜ ls -l /opt/texbin
lrwxrwxrwx. 1 root root 53 mar 14 07:13 /opt/texbin -> /opt/paulo/applications/texlive/2024/bin/x86_64-linux
 
@PauloCereda oh no, I got /usr/local/texbin
 
@Skillmon ooh
 
(actually, I'm not 100% sure, might as well be /usr/local/bin/texbin)
(but /usr/local, that I'm sure of)
 
10:09 AM
@yo' Kudus to the guardian for getting the capitalisation of LaTeX right :)
 
yo'
@samcarter oh yes, I completely forgot about this point! (I guess that they, unlike others, are careful about "latex" having a more known meaning.
 
@yo' :P
 
10:25 AM
This appeared on my Twitter timeline:
Schrödingers documentation:
If there are docs, noone will read them.
If there are no docs, everyone will complain.
4
@DavidCarlisle ^^
 
@PauloCereda docs or ducks?
 
@DavidCarlisle ooh
 
11:09 AM
@DavidCarlisle docs, with ducks the first line must be no one will eat them
3
 
 
2 hours later…
1:17 PM
In pdftmanagement-testphase, is there a way to query / determine the existence of a certain key in a pdfannot dictionary of a certain type, such as Border or C in a link/GoTo annotation? I only found \pdfannot_dict_put:nnn and \pdfannot_dict_remove:nn for manipulating annotation dict content.
@UlrikeFischer ^^^?
 
1:49 PM
@AlexG why do you need that?
 
@UlrikeFischer I'd like to improve ocgx2 a bit and adjust the look of ocg switching links according to whether hyperref is loaded or not. (Alternatively, I could just test whether hyperref is loaded or not at the beginning of the document, but I am looking for a more elegant way.)
Hyperref adds Border and C, which are empty otherwise.
 
@AlexG sorry what I meant is, why do need to query C for this? What do you want to know?
 
@UlrikeFischer E. g. the current color setting of a link's border.
@UlrikeFischer If C is empty (because Hyperref has not been loaded), I would set Border to 0 0 0, for instance.
 
@AlexG hm, imho if you want to do something depending if hyperref is loaded or not, the most elegant is to use \AddToHook{package/hyperref/after}. Relying on some side effects from hyperref sounds a bit dubious. E.g. assume that hyperref gets some options "noborder, no colors".
 
@UlrikeFischer Knowing about those settings is exactly what I want.
@UlrikeFischer I will use the hook you suggested.
@UlrikeFischer Thank you :)
 
2:04 PM
@AlexG there is currently no interface, you would have to access the internal name of the dictionary (\pdfdict_get:nnN {l__pdfannot/link/GoTo}{C}\l_tmpa_tl). It wouldn't be difficult to add a \pdfannot_dict_get:nnN, so if you think it would be useful open a feature request and I will add it.
 
@UlrikeFischer Such an interface for querying could indeed be useful in places. For now, I think, using the package/hyperref/after hook will be sufficient.
 
@JosephWright I am trying to have a look at l3doc.cnf file to see what normally goes in there. I understand this is limited to the Team. I have changed l3 to a new name for some experiments with my colorizer ackage and would like to add some items to a config file. Sorry I am posting here, but questions now get buried easily on the site
 
@yannisl why a config rather than simply calling l3doc with the options you want?
 
@DavidCarlisle I have changed the l3doc to load book rather than article if you see the code not possible with any option, unless I missed something
 
well even with a l3doc.cfg you'd need to hack \LoadClass since it is unconditionally called. but you can use colorizerdoc.cls that patches \loadClass so that \LoadClass{artice} loads book, then loads l3doc, I would guess.
@yannisl but l3doc is such a mess, and many of the good bits are in doc v3 anyway so I'd use \documentclass{book}\usepackage{doc} see the first paragraph of the l3doc class doc
 
2:19 PM
Thanks, l3doc is really in a mess and the only advantage is sorting l3 macros into groups, if this is an advantage, as they would sort themselves out in any case
Even doc now is a bit messy, try and define an indexing command that acts like a normal \index. You need to create NewElement. But convenient otherwise. It served us well for 30 years.
 
Hello. I want to improve this answer: tex.stackexchange.com/questions/33526/… I have described a flaw I found here : gitlab.com/rdrg109/snippets/-/issues/1 I have also described a question I need help with so that I can improve the answer.
 
@rdrg109 in any case I wouldn't edit 10 year old answers, post a new answer if you wish
@rdrg109 @yannisl one of yours ^^^ :-)
 
2:38 PM
@DavidCarlisle If the answer is 10 years old and the improvement is a small modification to a macro that covers a new scenario, should I post a new answer?
 
Your suggested change just changes the amount by which you are prepared to letter-space the string from an invisibally small amount to an invisibly small amount thats 3 times as big. It happens to work in that case as a isn't very wide and so there are lost of a and so by luck adding 2*82 sp gave just enough glue to justify the line, if you used a larger letter you might need more glue
@rdrg109 I would there may be millions of links to that answer on the internet by now and changing the code they link t can be confusing
@rdrg109 or leave a comment and ask the original author (who happens to be here today)
 
@rdrg109 @DavidCarlisle I think it is better to just post a new question. David's explanation in the chat I think is good enough. And it is an amazing coincidence I am here today :)
@rdrg109 What really needs to be edited is the question itself. Like put a realistic string say using DNA sequences or numbers, then one can also do grouping. For bigNumbers I usually display in fours, but I use Lua for this. CAn be done also otherwise.
@DavidCarlisle Hehehe what an amazing coincidence 3sp by chance fixed an old answer and I was here to hear it. I think I am going to buy a Lotto ticket or something.
 
@yannisl good plan when you win 50 million can i have half?
 
If the front page is flooded with edits to capitalise questions titles, can we call this capitalism? :P
4
 
@DavidCarlisle You can have 80% of it.
 
2:47 PM
@yannisl now dare you actually buy a ticket and test this...
 
@DavidCarlisle Need to find where...I can buy one. I think Spain has one
@samcarter No you need to call it Capialism! Site is flooded with a new generation of users, too many questions with 1 reputation point. Why do you think this is happening? Nobody seems to want to read docs anymore.
 
Mar 26, 2012 at 19:37, by David Carlisle
@Canageek moral of the story: never read the documentation, bad things happen
 
I think overleaf did a lot of good to bring new users, but raised expectations. Also with social media, attention spans are diminishing.
Maybe in 20 years, people will talk in snippets and communicate with emoticons, like in proto-writing:)
 
@yannisl They can then ask their latex questions at en.wikipedia.org/wiki/Emojli
 
3:03 PM
Unicode is about to add a few more emoticons, and yes they will be searching for answers at the wiki:) or worse trying their luck with ChatGPT
 
3:51 PM
5 hours ago, by Paulo Cereda
Schrödingers documentation:
If there are docs, noone will read them.
If there are no docs, everyone will complain.
 
@Rmano Nice one by Paulo.
 
In formulae: (Exhaustivity of a doc)×(Willingness to read it)=ℏ ?
 
user581918
Hello! I think about methods for bailing out of expansion driven by \the. Is there a control sequence \foobar with \the\numexpr\foobar\relax expanding to emptiness/to no token at all? Or for any of the other ...expr? Is there a control sequence \barfoo other than s.th. denoting an empty token register with \the\barfoo expanding to emptiness/to no token at all?
 
4:24 PM
@Cattleya I think you can use \the<token variable> as long as the <token variable> in question is empty. (<token variable> subsumes <toksdef token>, \toks<8-bit number> and <token parameter>. <token parameter> subsumes \output, \everypar, \everymath, \everydisp, \everyhbox, \everyvbox, \everyjob, \everycr, \errhelp. But there are packages for LaTeX which redefine these primitives.)
 
user581918
@UlrichDiez "But there are packages for LaTeX which redefine these primitives." - What do you mean?
 
@Cattleya For example the package "everyhook" by Stephen Checkoway.
 
user581918
4:39 PM
@UlrichDiez In TeXbook, chapter 20 I just found: "\the<token variable> produces a copy of the token list that is the current value of the variable. For example, you can expand '\the\everypar' and '\the\toks5'." Is there no way of having \the-expansion stop with emptiness other than an empty <token variable>?
 
@Cattleya I don't know any other way. I don't know a way of having \the...expr\foobar\relax terminate without delivering any tokens, either. But that would be nice to have.
 
5:32 PM
I'm back. My laptop refused to boot on February 10, and although I managed to get a substitute, I lost access to all my passwords. It will take a while to catch up, and I have some other things to do first. But I've lost something more -- My husband, Gordon, passed away on February 22. He was terribly ill since early December, and, mercifully, is now at peace. Be kind; you never know.
14
 
5:48 PM
@barbarabeeton Sincere condolences
 
6:02 PM
@barbarabeeton <3
 
6:17 PM
@barbarabeeton oh, I'm really sorry. Sincere condolences. <3
 
@Cattleya it depends how you define \the expansion. \def\the#10{}\the\count0 expands to empty
 
@DavidCarlisle Cheat
 
For those people waiting decades to get non bold Greek see ^^ that Unicode have been busy adding more useful characters.
2
 
@DavidCarlisle I see you picked flying saucer without beams :(
 
@JosephWright me? I just based an editors draft on a draft UCD not cheating at all:-)
 
6:21 PM
@DavidCarlisle I meant \the ...
 
@JosephWright I didn't want to scare people too much
 
@DavidCarlisle :)
 
@JosephWright ah
 
@DavidCarlisle Or people who wanted one symbol that is well-established and has been for >100 years ...
 
@JosephWright well not so different from standard latex and \the\everymath not expanding to what people think it expands to
 
6:23 PM
@DavidCarlisle Well true yes
 
@JosephWright oh is that in 16 what's the plimsol thing called?
 
@DavidCarlisle Well depends on the people ... I'd expect the result you get
@DavidCarlisle No, will have to wait for 17 - there's a long lead time
 
@JosephWright I suppose a flying saucer would do, same basic shape
 
@DavidCarlisle They I think decided to call it 'MEDIUM CIRCLE WITH CROSS BAR' or some such - really don't know why we couldn't have 'PLIMSOL SYMBOL' or 'STANDARD STATE SYMBOL'
@DavidCarlisle :)
 
7:10 PM
@barbarabeeton My sincerest condolences.
I read docs. All the time. Well written docs are as good as well written software.
Probably harder to create too.
 
7:26 PM
@LaTeXereXeTaL yes @UlrikeFischer just referred to some particluarly excellent documentation on the main site:
you could check the documentation - it mentions on the first page how to input parentheses. — Ulrike Fischer 2 hours ago
 
@DavidCarlisle They can be tricky.
 
7:54 PM
@DavidCarlisle A documentation that mentions the important point on the first page is clearly particularly excellent!
 
@UlrikeFischer @JosephWright
https://texlive.net/run?%5Cdocumentclass%7Barticle%7D%0A%5Cbegin%7Bdocument%7D%0A%5CExplSyntaxOn%0A%5Csys_split_query:nnN%20%7Bls%7D%7B*%7D%20%5Cl_tmpa_seq%0A%5Cseq_map_inline:Nn%5Cl_tmpa_seq%20%7B%5Ctexttt%7B%231%7D%5Cpar%7D%0A%5CExplSyntaxOff%0A%5Cend%7Bdocument%7D%0A%20%0A
grr urls here are so unreliable
 
@DavidCarlisle wow. Is it already in texlive? or did you add something?
 
@UlrikeFischer the expl3 layer is in the format, and strangely enough I had the rest to hand...
 
@DavidCarlisle cheating ;-)
 
@UlrikeFischer well i was logged in and switching it to tl2024, and while I was there.....
 
8:20 PM
@DavidCarlisle Cool
@DavidCarlisle So when do we can the first question?
@UlrikeFischer Tonight I think
 
@UlrikeFischer you need to %-encode markdown characters (_ and *)
@JosephWright I wonder if I should prod that tools PR :-)
 
@DavidCarlisle Yup
 
9:04 PM
@barbarabeeton very sorry to hear about your loss.
 

« first day (4884 days earlier)      last day (66 days later) »