« first day (3080 days earlier)      last day (1856 days later) » 

Kurt
Kurt
1:12 AM
Please vote to delete this spam answer: tex.stackexchange.com/a/483602/16550
 
barbara beeton
barbara beeton
1:27 AM
@Kurt -- Didn't vote; flagged for moderator attention. That's usually faster and more thorough.
 
 
7 hours later…
Sebastiano
Sebastiano
8:44 AM
@JonasStein Good morning
 
Jonas Stein
Jonas Stein
9:36 AM
@Sebastiano Hi Sebastiano. What is up?
 
Sebastiano
Sebastiano
9:53 AM
@JonasStein Bad but I'm with bronchitis. And then everything happened to me on Friday. I had parked my car at work and found it damaged. I hope in the GPS installed by my insurance.
 
Jonas Stein
Jonas Stein
I am confused. Do we know each other?
Did you ping me to talk about bronchitis in the tex chat?
 
Sebastiano
Sebastiano
@JonasStein no:-( I just waved goodbye. When users are in chat connected greetings for respect.
 
David Carlisle
David Carlisle
@Sebastiano you have been asked so many times to stop doing that....
4
 
Jonas Stein
Jonas Stein
@Sebastiano when you ping someone with `@` she/he gets a call for your message. That is like ringing someones door bell to say Hi and run away.
Reminds me a bit to Family Guy, when Stewie calls "Mom Mum Mommy"
 
Sebastiano
Sebastiano
@DavidCarlisle is stronger than me. It seems discourteous not to greet. :-(
 
Jonas Stein
Jonas Stein
10:06 AM
@Sebastiano Please do not use @ with my username, if you do not have anything important texrelated for me.
 
David Carlisle
David Carlisle
@Sebastiano no it is discourteous to ping people for no reason and to flood the chat with such messages every time you join.
8
 
Sebastiano
Sebastiano
ok
 
Paulo Cereda
Paulo Cereda
@JonasStein ah I remember that episode :)
 
Joseph Wright
Joseph Wright
10:23 AM
@Sebastiano @DavidCarlisle is right: in general, it's simply not sustainable for everyone to say 'hello' to everyone else in chat. I occasionally say hello to one or two people, but I know them 'in real life' and even then it's only very seldom I say things
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle found a very good PDF viewer: apvlv. I am sure you will love it. :)
 
David Carlisle
David Carlisle
@PauloCereda I'm warned off by the vs
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle You are a very wise man. :) 
apvlv — PDF/DJVU/UMD/HTML/TXT viewer with vim-like behaviour
@DavidCarlisle ^^ look, they write behaviour the right way. :)
 
David Carlisle
David Carlisle
a proper pdf viewer:
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle humpf
@DavidCarlisle still in bantham? I thought you got a new, powerful laptop. :)
Meanwhile, @JosephWright is probably in palladium. :)
 
David Carlisle
David Carlisle
10:29 AM
@PauloCereda nope:-) this is a core i7 but probably the cheapest dell made at the time,it works well enough though
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle oh I can relate with cheapest Dell models. :) Thankfully, when I got mine, I manage to dig their store looking for Linux-based systems, as it reduced the final price close to 20%!
 
David Carlisle
David Carlisle
@PauloCereda cheaper but then you miss out on that full featured windows 10 experience, as shown above by emacs running in an x server, the typical Windows use case:-)
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle no thanks. :)
@DavidCarlisle the most powerful Windows-powered machine I have is my XBox One. :)
 
Joseph Wright
Joseph Wright
11:03 AM
@UlrikeFischer I'm reading over the PDF stuff now. Are you happy for me to edit and/or pull stuff into master?
 
Ulrike Fischer
Ulrike Fischer
@JosephWright sure. You could also create a new branch if you want or think it would be better (I'm using the stuff to tag the tagpdf document to try out things so it is already "in (messy) use" in some respect).
 
Joseph Wright
Joseph Wright
@UlrikeFischer Right OK: perhaps I'll do that
@UlrikeFischer I have two major concerns, really. The first is whether we should be exposing e.g. startlink as-is, or instead should do what pdfbase does and provide a single 'link' function that grabs the linked material as an argument.
@UlrikeFischer The second is the hook business. I think I'm still more of the mindset that we should be looking at having fixed data structures that we are clear on the semantics for ... so we don't end up with other people 'managing' data
@UlrikeFischer I'll start on something later on, time allowing
 
Ulrike Fischer
Ulrike Fischer
11:19 AM
@JosephWright the problem is the interface to hyperref and also the hooks for tagging. You simply have different types of argument, and need sometimes to insert stuff in the attr part, or use user or goto name. One argument is imho not enough. And e.g. in the toc you have links having independant start and end commands.
@JosephWright I'm not sure about this too, I mostly lead out the options. But I think that one need both properties and seq/tl structures.
@JosephWright I have to do lunch now ... ;-)
 
Ulrike Fischer
Ulrike Fischer
11:46 AM
@JosephWright btw: Markus Kohm has been working on hooks too, and he is introducing a sort of cascading hook system "heading/preinit/subsection" komascript.de/releasesvn. But imho that is at different level.
 
Faheem Mitha
Faheem Mitha
12:05 PM
What options are there for using config files with TeX? Specifically, does TeX have anything to read from YAML files?
 
TeXnician
TeXnician
@FaheemMitha What do you want to configure?
 
Faheem Mitha
Faheem Mitha
Found tex.stackexchange.com/q/445292/3406 so probably the answer is no.
@TeXnician Just want to read some data/fields from a YAML file. Nothing very exciting.
Doesn't have to be YAML, but I've used it before, and at the time it was as good as anything, and better than most.
 
TeXnician
TeXnician
@FaheemMitha Well, either use Lua or use something else for your data.
CSV does work pretty well, XML can be used quite easily with Lua (ConTeXt does this pretty well), …
 
Faheem Mitha
Faheem Mitha
@TeXnician luatex? Don't know anything about it. And something else, as in?
CSV? Hmm.
 
TeXnician
TeXnician
@FaheemMitha It's really a question about the structure of your data.
 
Faheem Mitha
Faheem Mitha
12:11 PM
@TeXnician In this specific case, it is pretty simple. But I'm interested in more general answers. It sounds like a YAML parser for TeX does not exist.
 
TeXnician
TeXnician
@FaheemMitha Well, no, it does not. And with TeX being a little bit special about spaces it is quite hard to write a full YAML parser…
 
Faheem Mitha
Faheem Mitha
@TeXnician I see.
So could one use something like github.com/gvvaughan/lyaml with luatex? Purely hypothetical curiosity - I have no plans to do so.
 
Joseph Wright
Joseph Wright
12:29 PM
@UlrikeFischer I'll make some notes in the source, probably on master, and have a think about stuff in the driver level in a branch I'll use ...
@TeXnician It's doable, of course ...
 
Ulrike Fischer
Ulrike Fischer
@JosephWright fine. I'm currently trying to clean up more in the driver so (hopefully) won't interfere. An open question in the hook is also how to pass "arguments", e.g. about the current section level to a hook. I don't think that I would like to use #2 from the "outside" command in its code.
 
David Carlisle
David Carlisle
1:11 PM
@HenriMenke have you seen the pgf issues raised on texlive list?
 
Ulrike Fischer
Ulrike Fischer
@DavidCarlisle it would be easier to check with an example ;-(. Oh, or is this an example file he mentions?
 
Joseph Wright
Joseph Wright
@DavidCarlisle Yup :)
@UlrikeFischer Presumably pythontex doesn't use pdfTeX but tex
 
Ulrike Fischer
Ulrike Fischer
@JosephWright i just run the example file he mentioned and it really breaks.
@JosephWright in miktex (where I haven't the update yet) it works if I remove the latin1-inputenc.
on my system it is defined even in tex: \pgfutil@scantokens=\long macro:
#1->#1
 
Joseph Wright
Joseph Wright
1:40 PM
@UlrikeFischer It's not defined here
 
Paulo Cereda
Paulo Cereda
@FaheemMitha Well, even a general approach has to be bound to the data structure, IMHO.
 
Ulrike Fischer
Ulrike Fischer
2:30 PM
 
Faheem Mitha
Faheem Mitha
2:43 PM
@PauloCereda I'm not sure what you mean. A YAML parser would confirm to the YAML specificiations, for example.
But I have no clue what this would involve, exactly.
I meant to ask, what caused the PGF/TikZ move from SF to GH?
And why not use a more community oriented site?
 
Paulo Cereda
Paulo Cereda
@FaheemMitha Not all parsers involve full coverage of YAML specs, that's why the general approach forcibly is more ad hoc than desired, but that's what's available. That's why data structures matter, at the end of the day.
@FaheemMitha then ask Henri Menke.
 
Faheem Mitha
Faheem Mitha
Or even set up a GitLab instance? Assuming the resources for that exist.
@PauloCereda Ok.
@PauloCereda I thought someone here might know.
 
Joseph Wright
Joseph Wright
@FaheemMitha What's not 'community oriented' about GitHub? lots of code is hosted there
 
Faheem Mitha
Faheem Mitha
@JosephWright The platform is proprietary, as is the code that runs it. Rather like SE, really.
A community platform would be something like the old Alioth. Though I believe that has been shut down now.
I guess the new thing is salsa. Which indeed seems to be a GitLab instance. At least, it says GitLab at the top.
 
Joseph Wright
Joseph Wright
@FaheemMitha Yeah, that's true but ... so? The data, in particular the Git repos, are freely copyable. And it's extremely unlikely that the API for issue export would just vanish. I don't think many people have the time or resource to run a full hosting solution, so being open is not a major thing. It's about where the 'users' are.
 
Faheem Mitha
Faheem Mitha
2:51 PM
@JosephWright Can the issues be easily exported? What about other metadata?
 
Paulo Cereda
Paulo Cereda
@FaheemMitha Only a few corner cases are not covered, but the export/import procedure is quite straightforward...
 
Faheem Mitha
Faheem Mitha
@PauloCereda Oh
 
Paulo Cereda
Paulo Cereda
@FaheemMitha There's also an option for mirroring, which might be interesting as a ahem backup plan. :)
 
Faheem Mitha
Faheem Mitha
GH seems to make heavy use of tags (or what look like tags). The general effect is quite... colorful.
 
Sebastiano
Sebastiano
3:32 PM
@StefanKottwitz @JosephWright I am present, I would like if it were possible to write to you too. Yesterday afternoon I wrote to Joseph. Thank you. I can't find the other moderator with ping.
 
Sebastiano
Sebastiano
3:43 PM
@egreg I just voted for your answers and the question that is an analog of the help provided by @DavidCarlisle on hyphenation. \showhyphens{matematica}
\usepackage[greek.ancient,italian]{babel}% Italian+Greek language/hyphenation
\showhyphens{matematica}
There are parts of the book where, despite the hyphenation present, some words are more detached than necessary. An example is the formula in red where the t is detached from t'. the hyphenation hyphens in my book are very few. Maybe two or three if I remember correctly.
Here is another example where words are very detached from each other. Like the "SE".
Is everything normal or is there something that can be changed? Thank you
 
egreg
egreg
4:16 PM
@Sebastiano The bigger space between the period and “Se” seems due to \nonfrenchspacing. The others on the line look equal.
Add \addto\extrasitalian{\frenchspacing} to the preamble.
 
Sebastiano
Sebastiano
Wait now I try it.
I have an old now laptop from 2004 with a 256 MB graphics memory. It takes some time to process also because there are many errors. 
! Undefined control sequence.
l.111 \addto
            \extrasitalian{\frenchspacing}
?
@egreg In my structure I have put this into the preamble 
%----------------------------------------------------------------------------------
%
%----------------------------------------------------------------------------------
\addto\extrasitalian{\frenchspacing}
\showhyphens{matematica}
\usepackage[greek.ancient,italian]{babel}% Italian+Greek language/hyphenation
\showhyphens{matematica}
 
Ulrike Fischer
Ulrike Fischer
@Sebastiano put the command after loading babel. And correct your errors (and better comment the \showhyphens-lines, they will stop the compilation too).
 
Sebastiano
Sebastiano
@UlrikeFischer Wait, please, I'm a snail and I have to use the translator. First I have to understand your second part that I have not understood it.
The compilation starts now; who knows when I will correct the compilation errors that there are. Kurt and David have also recommended it to me many times. I thought that as the book will be revised and cut the parts that do not serve me will be easier to correct the errors.
It's okay. Now the spacing of the words is better. Thank you very much with my heart.
 
David Carlisle
David Carlisle
5:03 PM
@FaheemMitha yes you can move a project to gitlab for example including all issues just with a single click of an import from "github button" at gitlab.
@Sebastiano if you get an error, fix the error, don't look at the PDF.
@Sebastiano ancient Greek?
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle Wow, seriously?
 
David Carlisle
David Carlisle
5:24 PM
@FaheemMitha yes I tried a couple of projects at the time MS bought Gh just in case
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle Impressive.
Well, it's good to be prepared.
I don't see why PGF/TikZ didn't just go with GitLab in the first place, though.
At least their software is free.
 
David Carlisle
David Carlisle
@FaheemMitha I assume same reason I didn't make my projects moved to gitlab public. github has more users and more existing projects so moving only if it proves necessary, which it doesn't seem likely
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle Why does the number of users/project matter?
 
David Carlisle
David Carlisle
@FaheemMitha you need to be signed in to make an issue (on both sites)
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle Yes, I realise that. You mean people won't bother to create an account to create an issue?
That seems an odd thing to be concerned about. I make accounts for things all the time. Not enthusiastically, I must say. I have a GitHub account, even though I don't use Git.
 
David Carlisle
David Carlisle
5:31 PM
@FaheemMitha yes and it raises the bar to people doing so but so many otehr projects are already on gh it is more convenient o go with that
@FaheemMitha it seems far more natural to be concerned about inconveniencing your users than about some totally irrelevant details about who is running the site hosting
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle I wouldn't call it totally irrelevant, but then again, it's not my project.
@DavidCarlisle Raises the bar by a tiny bit, yes. And a sign up is a one time thing.
Assuming one doesn't forget the password, of course. I write mine down, at least for non-sensitive stuff.
 
David Carlisle
David Carlisle
@FaheemMitha for a while at least gitlab was running on an azure cloud as far as I understand. No matter if the software is free or not there has to be some fairly large infrastructure set up to host a site with that much use. It may be possible to build a completely free stack but chances are some big company has to supply the infrastructure at some level.
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle That might be true. There are small hosting companies, but I don't know how the support side of things works.
I mean, what goes on at the backend.
 
David Carlisle
David Carlisle
5:52 PM
@FaheemMitha that's rather the point, same as this site, it really doesn't matter how the back end is set up, so long as the data is not locked in.
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle I think we can all agree that locked in data is bad. Locked in for whatever reason and in whatever way.
 
David Carlisle
David Carlisle
@FaheemMitha sometimes, not always. I work for a software company and the fact that they pay me is a useful thing as far as I am concerned.
 
Skillmon
Skillmon
@DavidCarlisle and that's why vi/VIM are decent editors (whichs' great behaviour is copied by other programs) and Emacs is an OS...
@PauloCereda do I have to test it now? I'm very satisfied with my current one (Zathura).
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle I was talking about free software and the like.
 
egreg
egreg
@Sebastiano That should go after loading babel
 
Sebastiano
Sebastiano
6:02 PM
@egreg done from suggestion by Ulrike. See the previous comment.
@DavidCarlisle I needed ancient Greek for the Greek words. it is elegant for my opinion
They are a myriad of errors but the pdf is generated the same. You are perfectly right, and as I wrote earlier I will make the correction slowly. Although I went to work I'm not clear both for my disease and for the stress.
 
 
2 hours later…
Sebastiano
Sebastiano
7:39 PM
@StefanKottwitz I left a long message in my private room just a minute ago. Hello and good Sunday.
 
Paulo Cereda
Paulo Cereda
@Skillmon :)
 
David Carlisle
David Carlisle
8:00 PM
@Sebastiano ancient Greek not modern Greek?
@Sebastiano TeX really makes no attempt to make sensible typeset output after an error, it only carries on to allow you to check more of the the document for errors. It would be very surprising if you have usable output after an error.
 
Joseph Wright
Joseph Wright
@UlrikeFischer I'm reading the pdfTe manual and the PDF reference ....
 
Ulrike Fischer
Ulrike Fischer
@JosephWright many pages ;-)
 
Joseph Wright
Joseph Wright
@UlrikeFischer Yes
@UlrikeFischer I'm trying to get a feel for where the 'line' falls in terms of what goes with startlink
 
Ulrike Fischer
Ulrike Fischer
8:15 PM
@JosephWright I have the reference open just now too. Trying to figure out what \the\pdfpageattr /Hid true is doing in the hyperref driver, but can't find it ...
@JosephWright one should use the index. /Hid was never implemented and is now obsolete - one more thing to remove.
 
Joseph Wright
Joseph Wright
@UlrikeFischer :)
 
Ulrike Fischer
Ulrike Fischer
@JosephWright there is some old stuff in these drivers ;-) I'm trying to get a feeling what is really needed.
 
Joseph Wright
Joseph Wright
@UlrikeFischer Sure: Heiko covered everything. We had the same with the graphics ones
@UlrikeFischer Where's it listed?
 
Ulrike Fischer
Ulrike Fischer
@JosephWright the /Hid key? page 1104 in my version (3.6.2, “Page Tree”). And I already removed the stuff to set the page size, and also the bookmark stuff - I will rely on the bookmark package. And we need a driver function and a hook for the page attribute ;-)
 
Joseph Wright
Joseph Wright
@UlrikeFischer I'm thinking about links. Currently, I'm not sure if at the driver level you want to separate out user, goto name, etc.
 
Ulrike Fischer
Ulrike Fischer
8:30 PM
@JosephWright at which level would you do it? How and when would you add keywords?
 
Joseph Wright
Joseph Wright
@UlrikeFischer That's what I'm wondering. The idea is that driver stuff should be the minimal abstraction. I'm not sure if it belongs there or not ... possibly. I really need to see how the links are handled in other engines. Will read divpdfx manual next ..
 
Ulrike Fischer
Ulrike Fischer
@JosephWright I thought also about a \driver_pdf_startlink:nnn where the middle argument is the user or goto name keyword. But how to fit xetex in here?
 
Joseph Wright
Joseph Wright
@UlrikeFischer I'm looking at it: may need a few days
@UlrikeFischer What does goto name do exactly? Almost everything can be implemented in PostScript ...
 
Sebastiano
Sebastiano
8:52 PM
@DavidCarlisle I'll take your advice and put the modern Greek option. Yes even if there are so many errors the pdf is produced the same.
 
Ulrike Fischer
Ulrike Fischer
@JosephWright it creates an annotation with an /A << /S /GoTo /D (yyy) >> action. I'm not quite sure if one could "fake" it with the user type.
 
David Carlisle
David Carlisle
@Sebastiano it was a question not advice, I don't know if you are using classical or modern greek (since I can't read either I am hardly an expert in these things:-) . What do you mean by "the pdf is produced the same"? Sure you will get a pdf file but it is almost certainly unusable for publication if there are tex errors. The typeset text after an error is usually nonsense.
 
Joseph Wright
Joseph Wright
@UlrikeFischer Right, so we probably do need a second driver level
@UlrikeFischer I notice hyperref uses a single macro for the link, not separate begin/end ...
 
Ulrike Fischer
Ulrike Fischer
@JosephWright there are variants. E.g. in the toc it uses \hyper@linkstart{link}{\Hy@tocdestname}{#2}\hyper@linkend. (And I still wonder if this will work with tagging as it actually puts the link structure around the label and the text and adobe doesn't like it ...)
 
Joseph Wright
Joseph Wright
@UlrikeFischer Right, yes, for internal words
@UlrikeFischer But for 'normal' links you can absorb everything I think as an argument
 
Ulrike Fischer
Ulrike Fischer
9:10 PM
@JosephWright are you thinking about the driver or the user level?
 
Joseph Wright
Joseph Wright
@UlrikeFischer Driver level ...
@UlrikeFischer Have to start there
 
Ulrike Fischer
Ulrike Fischer
@JosephWright I didn't really dare to do this - there are commands in the drivers where the start and link is separated. E.g. \find@pdflink has \Hy@StartlinkName and the endlink is somewhere else. And I don't know if all the form field stuff can really be handled as argument.
 
Joseph Wright
Joseph Wright
@UlrikeFischer Sure I'm just getting a handle on it
 
Ulrike Fischer
Ulrike Fischer
9:26 PM
@JosephWright one could probably create a goto name-annotation also with the user keyword, but I'm not sure if it worth it.
 
Joseph Wright
Joseph Wright
@UlrikeFischer No, on balance I think not: I'll commit some suggestions soon
 
Ulrike Fischer
Ulrike Fischer
Every driver has a message like this: \Hy@Message{Sorry, pdftex does not support FORM gauges}% No idea what this should be ;-)
 
Joseph Wright
Joseph Wright
@UlrikeFischer Progress bar?
 
Henri Menke
Henri Menke
@DavidCarlisle @UlrikeFischer github.com/pgf-tikz/pgf/commit/…
 
Ulrike Fischer
Ulrike Fischer
@JosephWright I have absolutly no idea, google didn't give anything and all drivers say they don't support this, with the exception of the texhtm driver which says that is does not yet support it. ;-)
 
Henri Menke
Henri Menke
9:37 PM
7 hours ago, by Faheem Mitha
I meant to ask, what caused the PGF/TikZ move from SF to GH?
@FaheemMitha Because SourceForge is an unusable pile of garbage. I have experience with GitHub and know how to setup pipelines etc. and since I am the one who took over the maintainership for now, this was the natural choice. LaTeX development is also on GitHub, so most LaTeX devs already have a GitHub account and can start contributing right away.
7 hours ago, by Faheem Mitha
Or even set up a GitLab instance? Assuming the resources for that exist.
@FaheemMitha It's not only the lack of resources but also that people do not want to create yet another account.
 
Ulrike Fischer
Ulrike Fischer
@JosephWright still wondering about the version stuff. Should be setup a \sys_pdfversion_fp variable? but how then do we prevent people to change this variable and then wondering why it has no effect?
 
Joseph Wright
Joseph Wright
@UlrikeFischer I don't think we can do it that way, as we can write the minor version to the PDF. Has to be similar to the random seed.
 
David Carlisle
David Carlisle
@UlrikeFischer for the pdf version? better two integers than a decimal I think.
 
Ulrike Fischer
Ulrike Fischer
@DavidCarlisle ok, it would make comparing a bit easier but other stuff is awkward. So we only need a name ;-)
@HenriMenke ah good.
 
David Carlisle
David Carlisle
@UlrikeFischer comparing is what I had in mind, I don't think you can guarantee ordering if there are more minor versions: how do you compare versions 2.1 and 2.10 ?
 
Ulrike Fischer
Ulrike Fischer
9:46 PM
@DavidCarlisle ah yes, that too ;-) (I was only considering one-digit minor versions as this is what we have now, but there is naturally no garanty).
 
David Carlisle
David Carlisle
@PauloCereda were you supporting Oxford or Cambridge?
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle vvv 
$ uname -a
Linux cambridge 5.0.5-200.fc29.x86_64 #1 SMP Wed Mar 27 20:58:04 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
So I believe Cambridge it is. :)
Is it the boat race thingy?
 
David Carlisle
David Carlisle
@PauloCereda good choice: they won the men's and women's and the men's and women's reserves today:-)
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle ooh
 
David Carlisle
David Carlisle
Of course it helps if your "typical Cambridge Student" sitting in the boat just happens to be 43 years old with two gold Olympic medals and a few world championships ...
 
Paulo Cereda
Paulo Cereda
9:54 PM
@DavidCarlisle :)
@DavidCarlisle Was it an exciting race? :)
 
Faheem Mitha
Faheem Mitha
@HenriMenke I see. Thank you for the explanation.
 
David Carlisle
David Carlisle
@PauloCereda no one sank:(
 
Faheem Mitha
Faheem Mitha
@HenriMenke I can certainly relate to that.
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle bah
 
Faheem Mitha
Faheem Mitha
@HenriMenke So the people who originally developed PGF/TikZ have stopped? All of them? I'm not even sure who they are, except for the original author, namely Till Tantau. I see that Christian Feuersänger is also mentioned on the Wikipedia page.
@DavidCarlisle Don't they have to be students to participate?
 
David Carlisle
David Carlisle
9:58 PM
@FaheemMitha yes he's returned to education, doing a masters degree.
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle Some gold medallist?
 
David Carlisle
David Carlisle
James Cracknell
James Edward Cracknell, OBE (born 5 May 1972) is a British athlete, rowing champion and double Olympic gold medalist and prospective Conservative Party politician. Married to TV and radio presenter Beverley Turner, he and his wife have three children. Cracknell was appointed OBE for "services to sport" in the 2005 New Year Honours List. == Biography == Cracknell began rowing whilst attending the independent Kingston Grammar School and rowed at the Junior World Championships in 1989 and 1990, winning a gold medal in 1990. He graduated from the University of Reading as a Bachelor of Science (BSc...
 
Faheem Mitha
Faheem Mitha
@DavidCarlisle I see.
 
Paulo Cereda
Paulo Cereda
10:13 PM
@DavidCarlisle we need more boats
 
Skillmon
Skillmon
@PauloCereda actually I tested it, it looks promising but I'll stick with Zathura for now :)
 
Paulo Cereda
Paulo Cereda
@Skillmon ooh
@DavidCarlisle ^^
Rabbits are good with PDF viewers :)
 
Faheem Mitha
Faheem Mitha
@HenriMenke I also don't disagree about SF. I've never had a good opinion of it, but I'm also not clear what problems people have had with it. To me, it just seems slow, clunky, and the interface is not up to much.
But there are still people using it.
 
Henri Menke
Henri Menke
@FaheemMitha The original developer is Till Tantau. Christian Feuersänger made a lot of contributions and took over maintainer ship when Till became inactive. Eventually Christian also became inactive due to offline commitments. So neither Till nor Christian have stopped. They are just rather inactive, but both of them are also administrators of the new GitHub repository which means that they clicked the invite I sent them, so they are definitely alive.
@FaheemMitha I have also been in email contact with Christian, especially because I am not yet authorized to publish new releases on CTAN. He is still in charge of that currently.
 
Faheem Mitha
Faheem Mitha
@HenriMenke I see. Thank you for the clarifications.
What were your big problems with SF, if I might ask?
 
Henri Menke
Henri Menke
10:35 PM
@FaheemMitha The horrible user interface and the painfully slow Git server. git push could take up to five minutes. Also no simple way to add test pipelines. When I took over maintainership the manual couldn't be built anymore because there were no automated tests.
 
Faheem Mitha
Faheem Mitha
@HenriMenke I see. You needed automated tests for the manual?
 
Henri Menke
Henri Menke
@FaheemMitha Nobody needs automated tests if everyone just runs the tests locally before committing, but automated tests are just really convenient.
@FaheemMitha Especially because testing the manual on all engines takes over 1h.
 
Skillmon
Skillmon
@HenriMenke you could reduce that, on a modern CPU you have at least 2 cores most times 4, so you can run 4 engines in parallel... Still time :)
 
Faheem Mitha
Faheem Mitha
@HenriMenke Yes, I see.
I actually just posted an issue on SF. GH is certainly a more pleasant experience.
 
Henri Menke
Henri Menke
@Skillmon Sure but let's be honest, nobody likes running tests.
 
Skillmon
Skillmon
10:41 PM
@HenriMenke that's true :)
 
Paulo Cereda
Paulo Cereda
@HenriMenke assert(true, true); and we are done. :)
 
Skillmon
Skillmon
@PauloCereda if(bug) return feature;
2
 
Paulo Cereda
Paulo Cereda
@Skillmon LOL
 
Henri Menke
Henri Menke
 
Paulo Cereda
Paulo Cereda
@HenriMenke LOL I love the VW hint!
 
Faheem Mitha
Faheem Mitha
10:49 PM
Does PHP really help to be even more dysfunctional?
 

« first day (3080 days earlier)      last day (1856 days later) »