@HarishKumar Hi

@egreg Hi

@barznjy Hello

Folks who use emacs: if all goes well, you'll have some measure of expl3 editing support in AUCTeX :)

Question: Would it be appropriate to call expl3 a different dialect of TeX (akin to Plain, LaTeX, and ConTeXt)? I'm well-aware that it's an effort of the LaTeX project, but it's not really…LaTeX-y (on purpose).

@JosephWright we need to work this out who does what there ... the title I gave to Karl would allow for it but that doesn't necessarily mean I would when using this title

@SeanAllred expl3 is by design fairly lowlevel and we made sure that it does in fact work on top of LaTeX2e, plain TeX and ConTeXt so in that respect it is a dialect of TeX, but it is also different as the other dialiects only partly work on top of the others if at all as they all attempt to provide a full universe not just a foundation - the corresponding full universe would be LaTeX3

@barznjy Ask an MWE on the site, perhaps?

@SeanAllred Do you work on AUCTeX?

@AndrewCashner I found a nice summary of available fonts at tex.stackexchange.com/a/59405/3406.

Morning all

@FrankMittelbach I'm planning to focus on Unicode stuff, so was assuming you'd tackle the bigger picture

@JosephWright does the LaTeX team have a traditional IRC channel?

@FaheemMitha No

@FaheemMitha Most things go to the team list or if public to LaTeX-L

@FaheemMitha We've ended up using TeX-sx chat as it's convenient: I'm often about, so is David, Frank off-and-on and in the past Will and Bruno (Robin has I think not come into the chat but is registered on the site)

@FaheemMitha I hope we are not disturbing others!

@JosephWright guess we work that out eventually ... good chance that @DavidCarlisle, Will and Bruno are there so we should have a chance to divide

@FrankMittelbach Yes

@FrankMittelbach Team meeting :-)

@FrankMittelbach I doubt we have much chance of getting Morten to come :-(

@JosephWright not sure he often does

@FrankMittelbach Fingers-crossed

@JosephWright Ok. It is quite common to developers to use IRC traditionally, which is why I asked. E.g. Mercurial. SE might have some advantages - better interface, history is kept indefinitely and can be accessed via url. That last might be a bug or a feature, depending on how you look at it.

@FaheemMitha The team is relatively 'low tech' in that sense, partly as things go back a very long way (early 1990s)

@JosephWright I see.

@FaheemMitha Email has tended to be popular as it's archived

@JosephWright yes, email is good.

I save all my email.

I'm going to edit that font answer. I hope someone can review it to check I didn't make silly errors.

@FaheemMitha I tend to 'shred' mine: I see it as akin to a phone call

@JosephWright doesn't that depend on the email? Suppose the email is on a technical topic?

@FaheemMitha I keep some stuff but review periodically and then destroy. For example, for LaTeX e-mails once something is implemented the thread is no longer useful so I delete it

@JosephWright Oh. That makes it harder for future TeX historians.

@FaheemMitha Anything important should be logged in some other form

@JosephWright ok

@FaheemMitha Like I say, anything to the team list is archived: I mean stuff at my end

For example, we don't know much about the early history of apt because it was mostly on IRC, in the late 90s.

@FaheemMitha I keep anything that might still be 'live', so have some e-mails that have hung around for a few years

@JosephWright Ah, Ok. The mailing list has an official archive. Got it.

@FaheemMitha Goes back to 2005 when it was converted from (I guess) a redirect

@FaheemMitha @FrankMittelbach has mails back >20 years I believe!

@JosephWright ok

@JosephWright I hope he saves them carefully.

@FaheemMitha Main ones I hold are work-related. For example, anything significant about teaching needs to be kept until the academic year ends and exam board has verified results

@JosephWright sure.

Not sure how you manage to be an academic and do this stuff too, though.

With a couple of fonts, I'm seeing that the pdf looks ok, but when printed, it is garbled. What are possible causes for this?

BTW, are people off work today? It's a public holiday in the UK

I'm guessing this will apply mainly to countries with 'Christian tradition'

This list could be much expanded -> tex.stackexchange.com/q/59403/3406

And it doesn't distinguish between pdflatex and xelatex engines.

Which I think would be useful.

@JosephWright It's a public holiday here too, sort of. Though probably not so much as in the UK.

@FaheemMitha I'm intrigued by the concept of scale of the amount of public holiday :-)

@JosephWright some stuff is closed, other stuff isn't. It's an Indian Govt holiday, at least.

timeanddate.com/holidays/india/good-friday

Holidays are a big thing here. They have a lot of them. Hindu, Muslim, Christian...

Not vouching for the accuracy of that link above.

However, India does not have anything comparable to the period around Christmas, when, as I recall, the entire UK basically shuts down for a few days.

The difference between Good Friday and a major Hindu holiday is that everything basically shuts down on the latter. In the former, much commercial stuff is still working.

This is just bad. It's Good Friday and I'm teaching today :-(

@yo' Commiserations.

Isn't your uni on holiday?

@FaheemMitha ummm not really. But it is on Monday, since that's the holiday in my country.

@yo' Oh. I thought it was the same day everywhere.

@FaheemMitha hahahahahahah

@yo' forgive my ignorance.

@FaheemMitha forgive me laughing so badly :D

Though somehow it seems reasonable to expect Good Friday to fall on a Friday.

@FaheemMitha no, the Czech tradition celebrates Monday. We flog girls on that day to keep them fresh :)

@yo' Interesting traditions you have in the Czech Republic. Are these public floggings?

@FaheemMitha yes Good Friday falls on Friday, and Easter Monday falls on Monday (both are public holiday in UK)

@DavidCarlisle that sounds sensible.

"TeX Live has been developed since 1996 by collaboration between the TeX user groups. TeX Live was originally perpetrated by Sebastian Rahtz. Present miscreants are Akira Kakuto, Karl Berry, Manuel Pégourié-Gonnard, Norbert Preining, Peter Breitenlohner, Reinhard Kotucha, Siep Kroonenberg, Staszek Wawrykiewicz, Tomasz Trzeciak, and a cast of thousands."

Are any of these miscreants here?

I think I saw Norbert pass through once. He seemed stressed.

@FaheemMitha depends. usually young gentlemen come from house to house, flog each girl in the house and get a painted egg as a bounty (or a chocolate egg, or money, or brandy, or something)

@yo' sounds like fun.

it is

no excuse me, my teaching starts in 5 seconds :D

@yo' take care.

@FaheemMitha Akira Kakuto answered a question a few minutes ago:-)

@DavidCarlisle oh

For a field in a table, is there a polite way of telling TeX "don't hyphenate"?

@FaheemMitha \raggedright ?

@DavidCarlisle Oh. stick that in the same cell? And why does it work?

@FaheemMitha because tex hyphenates words so it doesn't need to stretch space to justify text to the line width, but if it isn't justifying it won't hyphenate unless the word is actually longer than the line

@DavidCarlisle and what is \raggedright telling it to do? not justify?

@FaheemMitha usually you want it for the whole column so >{\raggedright\arraybackslash}p{3cm}

@DavidCarlisle oh, ok

@FaheemMitha ragged right edge to the paragraph

@DavidCarlisle the para in this case being the column?

@FaheemMitha well each paragraph in the column, if that is what you mean

@DavidCarlisle yes, that's what I meant.

@FaheemMitha mpg, Akira Kakuto, Nobert

@JosephWright thanks.

Looks like Pégourié-Gonnard hasn't updated his SE profile for a while. He is still referring to himself as a student.

@egreg: Buongiorno! Ready for today's readings? :)

@FaheemMitha Hasn't logged in since October

@JosephWright Oh.

@FaheemMitha Personally, I wouldn't like people going to my profile for the sole purpose of checking stuff I say. :(

@PauloCereda I didn't. I just happened to notice it.

Question: if one wants to change fonts temporarily in the middle of a document, it seems you need more precise knowledge than just adding the sty file, which makes a global change / switches the default to that font.

I.e. one needs to specify the font family. But there may be multiple font families, and which one to use may be less than clear.

@PauloCereda Of course! Yesterday I had to read from Exodus. But no bell ringing at Gloria in excelsis Deo: there were too many altar boys, so one of them stripped me off the task of ringing one. :)

@egreg oh no! :)

@FaheemMitha in other words, if you want to change to a different font, you need to know which font you want to change to:-)

@DavidCarlisle Well, when you put it like that...

I think I'll post an example. Then you guys can make fun of me.

@egreg music?

@egreg about that xelatex thing yesterday? You said there was a bug - does that require reporting? If so, should I do it?

@egreg: There was a boy yesterday who was having so much fun playing with one of these. ^^ :)

@PauloCereda hoverboard from back to the future?

@DavidCarlisle Redneck edition. :)

@PauloCereda I guess I should count myself lucky that is not animated.

@FaheemMitha Loading fonts in PDFLaTeX is very different from e.g, XeLaTeX because they understand different things (the latter accepts system fonts usage at almost no cost). So fonts to be used have to be correctly mapped regarding the family, the size, the shape, the series and a couple of other things. Using a package, in the majority of cases, eases this task because it does the dirty job for us (but it assumes you want to rely on the font throughout the entire document). >>>

>>> When you need to use another font mid-document, you need to provide as much info as possible in order to make the font appear correctly. That's basically what David said, you need to know which font you want to change to and know a couple of additional settings if you need more fine tuning.

@FaheemMitha Animation wouldn't be a problem, I'd be more worried if this comes with a looping sound. :)

@PauloCereda :) We don't have it. Around here a different device is used, called racola (in Italian, raganella)

@egreg Oh my! It looks so cool! :)

@PauloCereda I just want the same default it would use if I was to do \usepackage{...}. Surely that is possible?

@FaheemMitha I think Tom showed something yesterday, IIRC. It's a way of doing it.

@PauloCereda I don't recall.

@FaheemMitha usually it's just a matter of \fontfamily{zzz}\selectfont the package documentation (or source) should tell you the family name

@DavidCarlisle well, in at least one case I had problems.

@FaheemMitha as always: no MWE, no solution :-)

@yo' I'm planning to post an example.

I hope you had a good teach.

@FaheemMitha 'twas fun.

@yo' Always?

@FaheemMitha no, it's not always fun

Anyone using the Debian AUCTeX package here?

@yo' what are you teaching?

@JosephWright: ^^ you would love to be my reviewer. :)

@FaheemMitha calculus

@yo' elementary, medium, advanced?

@FaheemMitha that's a very good question

@yo' it is?

@FaheemMitha what is MWE ?

@FaheemMitha it's advanced calculus at a very elementary level.

@barznjy Minimal Working Example.

@yo' oh. do you teach them about limits? continuity? differentiability?

@FaheemMitha multidimensional integral. Now. For 2 exercise lessons

@yo' yawn.

@FaheemMitha you wouldn't yawn having 32 students in the classroom.

@barznjy My White Elephant. :)

@PauloCereda is it still alive?

@barznjy white elephants are always porcelain, unless you live on the north pole where polar elephants live.

@barznjy Sadly no. :) youtube.com/watch?v=CLnADKgurvc

@PauloCereda nice .. thanks

@PauloCereda by watching this, I miss 25 years ago ..

@barznjy I know the feeling. :)

@yo' I've taught too. And sometimes I feel sleepy. But it doesn't do to yawn in public, that is true.

@FaheemMitha say "sleepy" once more :D Before coming to Prague, I haven't slept for more than 6 hours a night the last almost two weeks. Here, I didn't sleep well due to feeling sick until last night, where I finally got 9 hours of good sleep.

@yo' sleepy.

Always glad to oblige.

@yo' I hope you are feeling better now. Get more exercise, more sleep, and try to keep the stress levels down. I recommend yoga. Dancing is also good.

Gyms are also Ok, but can be boring. For yoga you want a trainer/class.

@FrankMittelbach Thanks for the explanation – makes me feel better about my recommendation to the AUCTeX folks. And no, not in any official sense, but I'd like to. This is just a humble contribution :) (albeit a long-overdue one…)

@FaheemMitha With printing from Debian, are you having the same problem I did? The only solution I found was to use lp. tex.stackexchange.com/questions/213201/…

@AndrewCashner yes, exactly.

Though not just egaramond.

@AndrewCashner thanks for the pointer. One more reason to consider upgrading to jessie. You're a Debian user too, then? I think you said so earlier. Still using wheezy?

Apparently they never got around to fixing this 2 1/2 year old bug in wheezy. Useless people.

Hmm, the version mentioned at the end of bugs.debian.org/cgi-bin/bugreport.cgi?bug=692828 is 1.12.2-3, and that is the version of cairo I have.

@AndrewCashner I guess I can just print direct from the command line. I usually print from evince, but that is not going to cut it in this case, it looks like.

@Faheem Well they're almost ready to make Jessie the stable version. And stability has its benefits. In this case it means living with bugs, but at least you know what the bugs are. I have Jessie on another machine and will try this. I've actually been very happy using lp once I learned it. man lpoptions is helpful.

@AndrewCashner you could use lpr. That is just a BSD frontend to CUPS. It just doesn't use evince and associated libraries in the pipeline.

The relevant package is:

ii cups-bsd 1.5.3-5+deb7u5 amd64 Common UNIX Printing System(tm) - BSD commands

gmane.org seems to be having problems. Can someone confirm, please? The front page is returning an error.

@FaheemMitha news.gmane.org/group/gmane.comp.tex.latex.latex3 working fine

@JosephWright yes, for me too.

How about: thread.gmane.org/gmane.emacs.devel/184155

And is the gmane.org front page returning an error for you?

@FaheemMitha Yes, some error on the front page

@JosephWright ok. And that thread above, the links don't work. Should I report it?

@FaheemMitha Thread works if a bit slowly

@JosephWright oh. Doesn't seem to be working for me at all. I'll try again.

@JosephWright you guys use dropbox? I'd assumed git/github or something. Possibly BB.

@FaheemMitha The master repos are SVN on the DANTE server comedy (the same one that runs CTAN). We have a read-only mirror using Git on GitHub, which relies on Will Robertson's PC to keep up to date: github.com/latex3/svn-mirror. DropBox is a convenient place to put 'ready to test' code with all of the PDFs made, etc.: the same as we'd upload to CTAN.

SVN is world-readable should you wish to see it: latex-project.org/svnroot/experimental/trunk and latex-project.org/svnroot/latex2e-public

@JosephWright Oh, I'd assumed you would have switched to DVCS by now. If/when you are ready to switch, consider Mercurial. It's good.

@FaheemMitha Some of the team would be happy to, others much less so

@JosephWright once I go to thread.gmane.org/gmane.comp.tex.latex.latex3/3404/focus=3406 I ahve the same problem as I did with the other link. I guess I'll report it, since someone else says he doesn't see any problem.

@FaheemMitha I like Mercurial but it's clear where things are going: Git is the 'winner', certainly in the LaTeX world

@JosephWright Compared to TeX, DVCS is a piece of cake.

@FaheemMitha It would be trivial to convert the current GitHub mirror to the 'master' repo but we'd loose the world-wide definitive versioning that SVN gives us

@FaheemMitha :-)

@FaheemMitha Certainly not so complex

@JosephWright I think there is space enough for everyone. You should use what you like, imo.

@JosephWright "world-wide definitive versioning"? I don't follow.

@FaheemMitha I've shifted my stuff to Git from Mercurial

@FaheemMitha We have one master SVN and each checkin has a defined version

But it is certainly true Git is dominant. I think currently Facebook is keeping Mercurial afloat. Which is sad.

@JosephWright BB?

@FaheemMitha Was on BitBucket, but am moving the GitHub as most other LaTeX code is there

@JosephWright is that like a tag?

@JosephWright Github doesn't support Mercurial.

@FaheemMitha No

But you said:

@FaheemMitha No: SVN has a simple linear list of checkins starting at 1

@FaheemMitha Yes, to Git from Mercurial

@FaheemMitha My stuff <> L3/L2 stuff.

@JosephWright Still unclear what you would lose.

@FaheemMitha Convert Mercurial to Git then upload to GitHub

@JosephWright Sorry, can't read. And bummer.

@FaheemMitha With SVN there is no question what is the definitive version of the code

Maybe Mercurial will eventually be down to Facebook support. Yuck.

@FaheemMitha SVN means we can ensure that each source file links back to the checkin it relates to

@JosephWright Oh, because it is centralized?

@FaheemMitha Yes

@FaheemMitha I know that there are arguments both ways here!

@JosephWright I see. Otherwise you can have different version floating around.

@JosephWright Well, one does what one is comfortable with.

@FaheemMitha I'd be happy to move to a DVCS but it would need agreement

@JosephWright Yes, I see.

@FaheemMitha Yes: there's traditionally a worry with TeX that as people can use the sources directly, without compiling, they might have all sorts of 'stray' files around

I personally fled svn with a great sense of relief in early 2006. Haven't felt the slightest desire to go back.

I've been using Mercurial since Feb 2006 or so.

@FaheemMitha As I said, some of the team would be happy to jump, but it's a question of whether that discussion is a sensible use of time!

@JosephWright oh

@JosephWright Ok.

@JosephWright I imagine you might get tired of talking about it.

@FaheemMitha for a centralised project a DCVS mainly just adds complication without so many useful extra features, see RMS's comments in the thread you link to above:-) In practice I can't see switching to mercurial, but we could switch to git if it made sense at some point.

@DavidCarlisle well, it really depends on what you are comfortable with. tastes vary. I can't imagine using svn again - it would just be too constraining. And it's a poorly engieering project in my opinion, too.

@JosephWright @DavidCarlisle RMS has come begrudgingly into 'the future' with the move of emacs from bzr to git. He doesn't really like git, from what I can tell. The think with git (or any DVCS) is that it allows decentralization, but doesn't enforce it. I still use a centralized paradigm with git for my projects – GitHub is that 'definitive source'. As long as everyone remembers to commit/push (which isn't that different from an svn checkin), there are no problems (that I can see).

More importantly, in my opinion, is that it opens up the LaTeX project to the scrutiny and support of the open-source community with integrated tools (courtesy of our webdev friends who love git).

Of course, this doesn't negate the need for consensus, but perhaps it is a good argument.

@SeanAllred Will and I would be happy to move to a DVCS, others I'm less sure

@SeanAllred SVN has been public for many years, and CTAN after all has the sources usable for anyone

@SeanAllred sure but if you always push on commit then basically you haven't got any decentralised features and just have a more convoluted commit syntax, it's not that it's hard but it isn't exactly much of a selling point for anyone to switch:-)

@DavidCarlisle Well the Git argument there is you do something yourself in bits then push a 'summary' for others :-)

@JosephWright Usable, in this case, isn't the same thing. Of course the sources are publicly available :) What I like about the tools available for git (I'll just use GitHub as an example, but perhaps it's not the best tool – I don't know) is that it's built around the idea of a community of contributors. Thus, the tools available reflect that – GitHub's issue and PR system are a prime example. (The usage of the mirror's bug tracker should stand testament to this.)

Also, RMS stopped being a technical person some time ago. Like, in the 90s. I don't think people should be inspired to emulate him...

@DavidCarlisle I use the GitHub mirror for a convenient local log :-)

@DavidCarlisle But the idea is that we don't want an essentially decentralized codebase. We want an authoritative source.

(forgive me if my 'we' is a bit too encompassing :))

@FaheemMitha Exactly.

@DavidCarlisle I get the feeling you're one of the dissenters :)

@SeanAllred Well we have a community: it's called the LaTeX Team and they all have commit access to the SVN ;-)

You know, you can easily make git aliases to have it behave exactly like svn – though I'd of course encourage you not to do this.

I'm one person, and I still find those features useful.

@SeanAllred the main advantage of switching to git would in fact be github which is unquestionably a better issue tracking system than gnats, but having to use git to use github is a bit of a downer:-)

@JosephWright Oh, that's not fair XD You know what I mean.

@DavidCarlisle Oh, posh :)

@DavidCarlisle Have you ever tried out git for a personal project?

@SeanAllred not really: the issue if switching has never really come up, so dissenting or not isn't really the right description

@SeanAllred Yes, but semi-seriously it's not like we get loads of people wanting to send in code

@DavidCarlisle No: Will and I have discussed off-list but ...

@DavidCarlisle I do know that this isn't the first time the conversation has come up (at least in this chatroom)