« first day (3821 days earlier)      last day (1098 days later) » 

8:30 AM
@DavidCarlisle oh no
 
@PauloCereda who else would I blame for bugs in the Spanish learnlatex?
 
@DavidCarlisle oh
@HenriMenke it's actually an old Mac that I revived, but I am about to turn it into a Linux box again. :)
 
9:37 AM
@Plergux quack
 
@PauloCereda baaa!
 
@Plergux ooh hi sheep person!
 
@PauloCereda Hello duck person! :D
 
@PauloCereda Petit-déjeuner
 
@DavidCarlisle oh no
 
9:38 AM
@PauloCereda Salutations de Bordeaux
 
@DavidCarlisle ooh bonjour
 
@PauloCereda I wonder if anyone ever learns to ask for beer in French...
 
@Plergux oh
 
10:20 AM
user image
2
Just listened to a very nice talk by @UlrikeFischer !
 
@samcarter_looks_forward_TUG'21 Uff. Half of the time the english word came first to my mind ;-)
 
@UlrikeFischer but very good to understand, even with my limited french!
 
@samcarter_looks_forward_TUG'21 perhaps she was speaking English and you didn't notice.
 
@DavidCarlisle you would have been interested in the talk before: org mode by someone who uses cygwin emacs and tweaks it since more than 20 years.
 
@UlrikeFischer doesn't everyone?
@UlrikeFischer but I was in Bordeaux, for a wedding.
 
10:35 AM
@DavidCarlisle Oh, you can travel? You are there now or back again?
 
@UlrikeFischer google meet :-)
 
@DavidCarlisle online wedding?
 
@UlrikeFischer I think the bride and groom (and my sister, and two or three others) were really there, but we were not
 
11:02 AM
@DavidCarlisle But how do you get cake in an online wedding?
2
 
@samcarter_looks_forward_TUG'21 THIS
@samcarter_looks_forward_TUG'21 I was confused for a moment, "A Macbook running Windows 10?!"
 
@PauloCereda We got a few of those in the University Computer Labs XD
 
@Plergux oh my
 
 
2 hours later…
12:44 PM
@samcarter_looks_forward_TUG'21 The highlight of highlights ^^^
 
@samcarter_looks_forward_TUG'21 as the Bär sees it ...
 
yo'
@JosephWright Hi, please, may I have a clarinet question?
 
@yo' Yes?
 
yo'
How do you play the higher B nat? I have found a fingering chart and it says you ha e to press both the F hole and the key above it with you left thumb, is that right?
(and you also press a lot of the keys)
 
12:57 PM
ooh
@yo' clarinet adventures?
 
yo'
@PauloCereda yeah, getting out of the comfort zone that ends with A... ;)
 
@yo' ooh
 
yo'
Still haven't mastered flats and sharps other than low Bb, but I feel like understanding the high notes is more important
 
@yo' Yup
 
yo'
@JosephWright I'm afraid my thumb is not flexible enough :D
 
1:00 PM
@yo' You need to work on opening your throat as you go up the register: that's important as you go for the overtones
@yo' I assume Boehm system?
 
yo'
@JosephWright probably -- it's the cheapo Yamaha B45 (borrowed from friends)
@JosephWright yeah, that's the other tough part -- smooth transition between the lower and upper register
 
@yo' Yes then
@yo' :)
 
yo'
It's like with chest vs head resonance in singing :)
 
@yo' You'd have to have a European (German/Austrian) clarinette to be not Boehm
@yo' Something like that, yes
@yo' You have to flex your thumb, yes
 
yo'
@JosephWright well, I didn't have to look for some weirdness to download fingering charts, so I assume it's the most common thingy :)
Thanks a lot, I think might decide to practice songs within the low E--middle A region for now ;)
 
1:16 PM
Watching the BBC
 
yo'
@JosephWright also I thjnk I'll need to switch to a harder reed to play the overtones well
 
2:01 PM
@yo' What are you playing at the moment?
@yo' Sounds good - what pieces have you gone for?
 
yo'
@JosephWright Vandoren 1 1/2
@JosephWright I'm going mostly for traditional hymns. The good thing is that I know them by heart, so I can play them in any key I know the fingerings for :)
(I starte I think with 2 1/2 as that's what my friend used, but I felt it's too hard for me)
 
@yo' ooh transpose :)
 
2:20 PM
@yo' Good (best) brand, but that's pretty soft - you'll need to go harder to get up the register
 
yo'
@JosephWright yeah, I know...
 
2:43 PM
@DavidCarlisle do you have a short but complete list of all names that belong to the mathml3 namespace described by the ID "http://www.w3.org/1998/Math/MathML"? I only need the names, not their meaning etc.
 
3:21 PM
@UlrikeFischer yes
 
@DavidCarlisle please sent it to me ;-) I'm trying to setup the definitions for the pdf2.0 namespaces and the reference only says that it know mathml3 but not which tags this means.
 
@UlrikeFischer you have mail (I was just extracting a text list from github.com/mathml-refresh/mathml-schema/tree/main/rnc
 
3:38 PM
@DavidCarlisle thanks ;-) (Not that I can see how you got it from the github)
 
@UlrikeFischer grep -h -o 'element[ ] *[a-zA-Z0-9]*' *rnc | sort | uniq more or less.
@UlrikeFischer but I'm surprised pdf needs to know, can't it be classed as "any element in the w3.org/1998/Math/MathML namespace"?
 
@DavidCarlisle the question is if tagpdf has to know it. I'm not sure yet. The problem is that if someone invents a new tag, say "myvar" they have to role map it to a known standard tag. And the known standard tags are the tags provided by pdf like P or H1 and the mathml tags. I'm not sure yet if I should test that the requirement is fullfilled or should simply document this.
@DavidCarlisle but beside this: if someone use e.g. mo as a structure tag, I have to mark up that it is from the mathml namespace, so I need the list here too.
 
@UlrikeFischer I was just going to look up, but easier to ask you, so the actual tagging isn't namespace aware? an xml parser doesn't have to "know" mathml or svg or xhtml" the syntax specifies which element is in which namespace (even if it is not defined). <zzz xmlns="http://www.w3.org/1998/Math/MathML"/> is a MathML element by definition, even though it is not defined by the MathML spec,
 
3:55 PM
@DavidCarlisle pdf 2.0 does introduce namespace. But if I add an element, say << /Type /StructElem /S /math then I have to add also /NS <reference to mathml namespace dict>, which means that I have know that math belongs to mathml.
@DavidCarlisle but I don't have to include your list in the pdf, that is only for the tagpdf code.
 
@UlrikeFischer the whole idea of namespaces was that it removed the worry about name clashes so <a xmlns="htt... xhtml"></a> is different to <a xmlns="http... svg"></a> but if all applications decide they need to know in advance what <a> means it's lost the point somewhere:-)
 
@DavidCarlisle well the specification does allow /S /math /NS <mathml> and /S/math /NS <svg>. The question is, if I want to force the user to add the name space all the time they tag a structure, even more as the pdf, pdf2 and mathml set are special in pdf, they are the sets declared by default.
 
4:27 PM
@DavidCarlisle If I wanted to define a "namespace of local user tag names", what would be a good URI to use as identifier?
 
@UlrikeFischer data:,zzz is what I use (it's a valid URI try it in a browser)
 
@DavidCarlisle interesting, my browser shows a black triangle. I never saw that:
 
ah I thought you meant in the main window, you meant in the location bar
 
But even if it is URI, it wouldn't be unique if used in more than one pdf, wouldn't it? Can one add some random number? something like the pdf ID?
@DavidCarlisle yes. Btw: I also see an nice favicon in your github.io, that makes the texlive.net tests easy to recognize ;-)
 
@UlrikeFischer well no, I use data: urls quite often when something wants a absolute url but I just want to keep it quiet. a data url is any string (defaulting to text/plain mime type but you can put a base64 encoded image of a bear there if you specify that before the comma
 
4:44 PM
@DavidCarlisle but data:,currenttime would be okay too?
 
@UlrikeFischer yes all the tests were giving 404 errors in the console looking for a favicon so yesterday I opened a new davidcarlisle.github.io repo just so I could have a github page at the top level and a favicon and that seemed to be the obvious one to use
@UlrikeFischer yes
 
@DavidCarlisle then I will use that. A dedicated namespace for new tags is imho a good idea, it should simplify the code. The main question is how to avoid that the testfiles breaks if the namespace uri changes all the time ;-)
 
@UlrikeFischer if you use a very standard form with the same length, l3build ought to be able to normalise them?
or can you say (in the style of car manufacturers passing emission tests) "if loaded regression-test.tex use zzz else use unique id"
 
@DavidCarlisle yes, something like this should work.
 
 
1 hour later…
6:12 PM
Today I found some curious performance thingy in expl3: \seq_set_from_clist:Nn is slower than \seq_set_split:Nnn.
 
6:40 PM
I have a one page article (using scrartcl). When I include that one page inside a biggest document, it breaks very differently. The one page in the original article now goes over two pages.
How should I start to begin to understand what is going on here?
Breaking is something I rarely pay any attention to.
 
@FaheemMitha -- First thing, check the type size. Is it the same in both cases? (Larger type, more pages.) Next, check the dimensions of the type blocks. Same? Top margin of first page? Any of those can make a big difference.
 
@barbarabeeton Yes, the same type size, 12 pt. What are type blocks? Top margin might be very different. Is there an easy way to output page settings for a given page?
 
@FaheemMitha -- Here, I meant "type block" to be the main page content. Width of a full line by length of the type on the page from first through last line on a full page. (In other words, do both have the same number of lines on the page?)
 
@barbarabeeton I think the line heights and spacing are the same here. No reason why they shouldn't be. But the margins are all clearly different.
Supposedly the layout package does this. Or something like this. Let's try it.
 
@FaheemMitha -- Well, if the margins are different, then almost certainly the line length is different. That is, as long as the paper size is the same for both.
 
6:54 PM
@barbarabeeton Both A4. Font and font size the same. Line spacing should be the same too, unless different margins alters that somehow. I'm fuzzy how line spacing is determined.
 
@FaheemMitha -- Wider margins mean that there's less width available for the type, so fewer words will fit on one line. Result, more lines.
 
BTW, how are people doing with the pandemic? Things are blowing up here.
@barbarabeeton Sure. I was talking about vertical line spacing, to be clear.
 
@FaheemMitha -- Well, if top and bottom margins are wider, then fewer lines will fit on the page. Side margins affect the line width. I think that you should be able to insert a command \show\textwidth in the file, and it will report that dimension, stopping at that point in the compilation. (But I always compile from the command line, and I don't know how "fancier" guis behave.)
 
@FaheemMitha font size, text width, parskip? anything could be different
 
@JosephWright I hate to say it. But in one of my more complicated (and quite messy) documents using pgfplotstable, longtable, tikz, etc I got an input stack exceeded error which disappeared after I remove siunitx 3.0 from the search path ...
 
7:03 PM
@barbarabeeton @DavidCarlisle the layout package seems helpful. Though I've never used it before.
What TeX package you find depends on what question you ask.
@DavidCarlisle Font and font size are the same.
Isn't text width determined by font and font size?
 
@FaheemMitha no. not at all: if you go \textwidth=2in then it is 2in.
 
@FaheemMitha -- In Rhode Island, where I live, more and more people are getting vaccinated, but some don't want to. When people gather in big groups and don't wear their masks, bad things happen.
 
@FaheemMitha did you do as Barbara suggested and add \show\textwidth to see what value it has?
 
@DavidCarlisle Oh. Well, I didn't change the text width from the default. Whatever that is.
@DavidCarlisle No, I enabled layout, which gives one all parameters. Or so it seems.
 
@FaheemMitha so which are different?
 
7:08 PM
418 vs 448 pts, for \textwidth, apparently.
 
@FaheemMitha well there you are then, naturally all line and page breaking will be different
 
@DavidCarlisle Lots of other difference too.
@barbarabeeton I think there are lots of bad things happening here. Just got my first vaccine shot on Monday. Not much fun.
 
@FaheemMitha perhaps but since already all line and page breaking will be different, other parameters being different doesn't change a lot more
 
The Oxford/ZA vaccine.
@DavidCarlisle Presumably many parameters affect where a page breaks?
layout seems handy. I should try to remember it exists.
Well, good night folks. Off to bed. Eventually.
 
@FaheemMitha yes but if textwidth is different then page breaking will be different, looking at other parameters won't tell you much more about that.
 
7:12 PM
@FaheemMitha good night
 
@FaheemMitha -- Remembering is a good idea. I know it exists, and have even used it, but still forgot. Good night.
 
7:44 PM
@PauloCereda youtu.be/FF9EmIVuneM
 
8:05 PM
@PabloGonzálezL Ducky Manero
 
@PabloGonzálezL Epic! (except that the walking is eeeever so slightly out of sync :p)
 
@PabloGonzálezL @JosephWright celebrating the release of siunitx3
 
8:41 PM
Hello @DavidCarlisle and everybody here.
 
9:34 PM
@JosephWright Travis is failing because of TL'20 vs TL'21 on the latex3 repo. I'm guessing I have to clear the cache at Travis to have it install TL'21. For which branch? (or all of them?)
 
@PhelypeOleinik Just zap them all
 
@JosephWright OK. Thanks
 
@UlrikeFischer Can you send me the file or an anonymised version? I guess it's an issue \AtBeginDocument
 
@JosephWright I think it has to do with the issue I opened.
 
@UlrikeFischer Cool - I'll take a look
 
9:59 PM
 
@AlanMunn LOL
 
10:29 PM
When you're bored and decide to design webpages with tikz XD
3
 

« first day (3821 days earlier)      last day (1098 days later) »