TeX, LaTeX and Friends - 2015-06-27

« first day (1700 days earlier) ← previous day next day → last day (3530 days later) »

cfr

00:01

@AlanMunn Is it a problem? I'm really looking for cases to test prooftrees or justtrees on, and I'm adjusting them as I go when I find things which seem useful. (E.g. if I can add something small which will let me answer a question much more simply than the current answers.) I'm particularly interested in cases which require text or numbering to the left and/or right.... Only I don't want to flood the front page with them. I can just keep the answers, if necessary, without posting them.

cfr

00:12

@ArthurReutenauer The earlier thing you pinged me about. I hadn't read what you'd just said above. I just responded to the ping.

Arthur Reutenauer

@cfr Well my answer is above :-) About accents and diacritic signs generally. There was a very nice talk by Karel Horák about accents in Czech once in BachoTeX (around 2008 I think), I though it would be nice to complete it.

Ha, it was 2007. Here is the abstract: gust.org.pl/bachotex/EuroBachoTeX2007/prog.html#Horak_1 and proceedings were published in the TUGboat the next year, his article is here: tug.org/TUGboat/tb29-1/tb91horak.pdf

By the way @yo' should like that too :-)

2 hours later…

Nasser

02:31

Latex is not consistent. To change the color of the text, one writes \textbf{text} but to change the font of the text, it is {\small text}. They should both work the same way.

Paul Gessler

03:25

@Nasser {\bfseries text}, {\small text}. There, now it is consistent. :-)

Canageek

04:08

@Nasser That drives me crazy as well. \textbf{foo} vs {\bf foo}

Nasser

04:20

@PaulGessler thanks, will then use {\bfseries text}. But then again, this means there is more than one way to make the text bold? This is not good. A language should have one and only one obvious way to do something. I do not like TIMTOWTDI languages :) en.wikipedia.org/wiki/There%27s_more_than_one_way_to_do_it

4 hours later…

percusse

07:57

@Johannes_B I didn't get the reference. Did I do something about this?

Enthusiastic Student

08:55

Hi

May I ask a question about TeX editors?

Torbjørn T.

@EnthusiasticStudent As David would say: Don't ask to ask, just ask. What were you wondering about?

Enthusiastic Student

:))

Sure.

I have recently installed TeX Live 2015

it has TeXworks editor installed

Joseph Wright

@EnthusiasticStudent Good plan

@EnthusiasticStudent Indeed: you are on Windows then

Enthusiastic Student

@JosephWright Yes I am on windows 7

Today I saw WinEdt

Can I have both winedt and texworks editors?

Joseph Wright

@EnthusiasticStudent Another editor with a long history

@EnthusiasticStudent Yes

Torbjørn T.

08:58

@EnthusiasticStudent Yes.

Joseph Wright

@EnthusiasticStudent TeX files are just text files: you can edit them with whatever you want

Enthusiastic Student

So I should just download and install the windedt editor?

Torbjørn T.

@EnthusiasticStudent If you want to try it, yes.

Enthusiastic Student

No need to configure the Path of tex files or anything else?

Torbjørn T.

@EnthusiasticStudent Never used WinEdt, but usually the binaries are placed in the path, so you don't have to configure the editor.

@EnthusiasticStudent If WinEdt cannot find pdflatex etc., then you may have to point it to the correct folder. But again, I'm not familiar with it.

Joseph Wright

09:02

@EnthusiasticStudent WinEdt has a config wizard that will find your TeX system

@TorbjørnT. I used to use WinEdt ;-)

Torbjørn T.

@JosephWright Then I'll leave you to answer any further questions.

Joseph Wright

Morning @DavidCarlisle

David Carlisle

@JosephWright hi

@JosephWright wrote some chained callbacks last night to check your allocation code, seemed to work:-)

Enthusiastic Student

Thanks all.... @JosephWright @TorbjørnT.

Joseph Wright

@DavidCarlisle Good: I introduced no new errors

Enthusiastic Student

09:17

AH! I did not know that winedt is not free! TeXworks is free... Better I read their website before I download winedt.

Joseph Wright

@EnthusiasticStudent You can try WinEdt without paying

Enthusiastic Student

@JosephWright Yes, one month evaluation is free of charge; but I want to have something for long time use; I think that texworks is more suitable for me. I have used it about 3 years, from the first day I got familiar with TeX and LaTeX

Good question here: tex.stackexchange.com/questions/339/latex-editors-ides

463

Q: LaTeX Editors/IDEs

What editors/IDEs are available for easing the process of writing TeX/LaTeX documents? Please state some useful features like code completion, spell checking, building final DVI or PDF files, etc. This question is undergoing a systematic refurbishment, see Let’s polish the Editors/IDEs questi...

editors big-list ide

Joseph Wright

@EnthusiasticStudent Well yes

David Carlisle

@JosephWright I wouldn't commit to saying that:-), but as I mentioned earlier examples of multiple functions used on the same callback were rather thin on the ground for the existing code, I'll probably tidy these up later and make an example or test file that might come in useful (it's a pointless callback function just abusing the input buffer with gsub but easy to test multiple replacements done in various orders)

Joseph Wright

@DavidCarlisle All true

Enthusiastic Student

09:30

Good day/night friends... thanks for answering my question. You are always so kind to me.

Paulo Cereda

09:55

Top of the morning, gentlemen!

So in a couple of weeks I am about to get me a new bicycle. I had no idea they now have disc brakes!

David Carlisle

@PauloCereda hopefully more success than last time:-)

Paulo Cereda

@DavidCarlisle Exactly. :)

David Carlisle

Paulo Cereda

@DavidCarlisle awwww

Johannes_B

10:38

@percusse That was a reference to the former discussion in may where SE rules where criticied.

@ArthurReutenauer The only way to find that out is by asking. In germany we have a saying: Fragen kostet nichts :-)

Paulo Cereda

@Johannes_B Ein Bier bitte. :)

Johannes_B

11:12

@PauloCereda The yearly Bergstadtfest is happening right now, lot's of beer is served. ;-)

Paulo Cereda

11:26

@Johannes_B :)

David Carlisle

11:51

@ArthurReutenauer one for you:

I would guess that in text mode the entire word is being passed to the otf handler and ligaturing is happening there, but in math mode as tex is positioning each character, each letter is being sent separately. hmmmm — David Carlisle 26 secs ago

Johannes_B

12:19

@StefanKottwitz @JosephWright @clemens latex-community.org/forum/viewtopic.php?f=42&t=26440#p90454

Alan Munn

13:09

@cfr No, I don't think it's a problem at all. I just thought it was funny to see all these old tree questions that I'd given an answer to revived with a new one from you. And I agree with you: your take on these is really nice. I also liked the idea of clear cutting with forest. :)

David Carlisle

@Nasser no, they shouldn't.

@Nasser {\bfseries xxx} isn't the same as \textbf{xxx} usually the latter is better but in some contexts you need \bfseries (same kind of contexts where you need \small typically). It is very rare to need a size change in the kind of context where you would use\textbf

Joseph Wright

@PauloCereda Provided they are not really cheap ones you should be OK

2

Q: How do LaTeX macros interact with the TeX engine?

From wiki on LaTeX the basic requirement is to have a TeX compiler (which is used to generate output files from source), fonts, and the LaTeX macro set. Assuming I have a TeX engine and I would like to add to it LaTeX support. I would expect it to be similar to any language where I shoul...

tex-core compiling latex-project

Thoughts on the dupe or otherwise

David Carlisle

13:26

@JosephWright marginal but as he accepted your answer could un-dup it. but the followup comment shows he's still confused anyway, so probably intended to ask the "difference between latex and tex question"

@JosephWright :(

0

A: Incorrect Cyrillic (Macedonian) glyphs in Computer Modern

This isn't directly related to cm but looks like a bug in the core latex utf8 mapping, this fixes it for now \documentclass[varwidth]{standalone} \usepackage[utf8]{inputenc} \usepackage[T2A]{fontenc} \usepackage{xcolor} \makeatletter \AtBeginDocument{% \DeclareUnicodeCharacter{0403}{\@tabaccklu...

@WillRobertson hi

Will Robertson

@DavidCarlisle hi :)

Joseph Wright

@WillRobertson Hi

Will Robertson

amazing, all my friends are here!

David Carlisle

@JosephWright did you see @WillRobertson's question on site?

Joseph Wright

@DavidCarlisle One for path level one?

@DavidCarlisle Yes

@WillRobertson :-)

Will Robertson

13:37

kind of a worry, hey David? We might have to tell people to use \textit after all :)

David Carlisle

@JosephWright I would guess so, it looks like a straight typo (I always got accents the wrong way round doing French O'level:-)

@WillRobertson depends a bit if it's fixable in xetex, if not then yes

Will Robertson

fingers crossed. Thanks for the reminder about base mode…

David Carlisle

@ArthurReutenauer ^^^^^^^

Paulo Cereda

@JosephWright Oh they are expensive. :)

@WillRobertson Hi from a duck! :)

David Carlisle

@JosephWright running build check now (although I doubt any of our test files pick up Macedonian accents...)

Arthur Reutenauer

13:53

@DavidCarlisle Ha, well done, pity I never noticed it! :-)

David Carlisle

@ArthurReutenauer but I meant you to look at Will's problem (which I can't fix in xetex) not the cyrillic accents:-)

Arthur Reutenauer

@DavidCarlisle Oh sure; since you didn’t specify a chat line I just jumped to what was the most obvious ;-)

David Carlisle

@ArthurReutenauer this one tex.stackexchange.com/questions/252493/…

Arthur Reutenauer

@DavidCarlisle Thanks, found it. No time right now, I’ll make a note of it.

David Carlisle

@ArthurReutenauer thanks I did have a quick look at the xetex source in that general area but decided i didn't have a clue:-)

Will Robertson

13:58

@DavidCarlisle — btw things are going well futzing around inside of unicode-math

Arthur Reutenauer

@DavidCarlisle I may reach the same decision ;-) But I’ll look into it.

Jun-Goo Kwak

Hi, I have a quick question. Is it possible in the array environment to draw lines from one element to another element in a different row?

Will Robertson

I had my head inside NFSS for probably too long today and I think it overtaxed my brain

David Carlisle

@WillRobertson no point if the engines are all broke anyway, may as well give up and blame @ArthurReutenauer :-)

Will Robertson

nah, even if the ligatures aren't there it's still very worthwhile

David Carlisle

13:59

@Jun-GooKwak pstrikz or tikz can do this

Jun-Goo Kwak

@DavidCarlisle Thanks. I'll look into the documentation.

David Carlisle

@WillRobertson if you published a math paper with a missing ligature you'd kill @barbarabeeton. Do you want that on your conscience?

Will Robertson

Hah! I'd just blame it on the publisher.

(Apparently I have to lead how to do threads properly in this thing. How do you manage David? Do they have an emacs interface to chat.sx??)

s/lead/learn/

David Carlisle

@WillRobertson not quite yet, only the main site stackapps.com/questions/3950/sx-stack-exchange-for-emacs

Will Robertson

@DavidCarlisle AMAZING

Arthur Reutenauer

14:03

@Johannes_B Not sure what you mean here. I was referring to people who ask a question with a precise solution in mind, and refuse all the answers that don’t recommend that exact solution. But the solution they originally had in mind doesn’t work, of course - otherwise they wouldn’t have asked the question - and what they are seeking is a slight modification of their “solution” in subtle, unspecified, ways. This does happen a lot.

David Carlisle

@WillRobertson to reply select reply from the menu at the left of each comment or use the arrow on the right

Arthur Reutenauer

@Johannes_B In a way we’re all in that situation to some extent, but sometimes it’s more obvious than other times :-)

Johannes_B

@ArthurReutenauer Ah, now i understand.

Will Robertson

@DavidCarlisle Ta. It's less so the how, more so the remembering to :)

Paulo Cereda

Quack. :(

Gonzalo Medina

14:15

@ArthurReutenauer No. Both tilda and tilde are valid words but mean different things. tilde refers to the accent, tilda is the second person simple present of the verb tildar (which has as one of its meanings "to put the accent").

Johannes_B

Has anybody ever seen \end{textit} before?

Gonzalo Medina

@Johannes_B No. I had seen \begin{itshape}...\end{itshape} since some people tend to use font switches as environments.

Will Robertson

@GonzaloMedina Yah, \end{textit} would be pretty bad :)

Johannes_B

@GonzaloMedina @WillRobertson I wish the user had taken care of error right away. Now, the OP needs to fix that log file golatex.de/viewtopic,p,73217.html#73217

Gonzalo Medina

@ArthurReutenauer I wrote "second person" and it should be "third person".

Paulo Cereda

14:24

@Will: quack?

Will Robertson

@DavidCarlisle It turns out that unicode-math already (supposedly) activates mode=base. Which is good (subject to further tests) except that I'd forgotten about it. Sigh :)

David Carlisle

@WillRobertson :-)

clemens

Now I end up explaining a WordPress plugin on the main site o.O

Will Robertson

@PauloCereda quack :)

Paulo Cereda

@WillRobertson yaaay :)

barbara beeton

14:35

@Johannes_B are you referring to this?

1

Q: How to change the shape of reference numbers in both bibliography (as boldface) and citations (as italic)

Using the answer provided in How to change the shape of reference numbers in the bibliography, I have a document where the reference numbers in the bibliography are shown as boldface. Now at the same time that this number is shown in boldface in the bibliography I would like that when citing tha...

bibliographies formatting citing

i've just taken a look at this, and it seems that an explicit italic correction is ignored because of the machinations in \cite processing. i'm about to put together a complaint for latex-bugs.

David Carlisle

@PauloCereda getting a bike has made the BBC news:

bbc.co.uk/news/world-latin-america-33271772

Paulo Cereda

@DavidCarlisle ooh

@David: it's a very strong controversy in SP. :)

Will Robertson

user image

Great success

David Carlisle

@WillRobertson luatex?

Will Robertson

@DavidCarlisle Yep

@DavidCarlisle All the mode switching happens behind the scenes, so everything kinda just works.

David Carlisle

14:42

@WillRobertson not sure I like the monospaced ff ligature in \mathtt but I suppose that's a font choice?

Will Robertson

@DavidCarlisle Yeah that's super ugly. I haven't investigated why it's there exactly, but yes — a font thing.

David Carlisle

@WillRobertson so what to do about xetex

Johannes_B

@barbarabeeton hi, no, actually it is this golatex.de/viewtopic,p,73217.html#73217

Will Robertson

The ligature can be disabled with \setmonofont[Color=550055,Ligatures=NoCommon]{texgyrecursor-regular.otf} — feel free to ask Jacko why you'd have monospaced ligatures in the first place :)

David Carlisle

@WillRobertson yes I was just wondering why anyone would have sat down and created such a thing

Will Robertson

14:46

user image

The XeTeX version

@DavidCarlisle I have to admit in theory it might have seemed like a good idea at the time :)

Note the difference between \mathbf and \symbf in the XeTeX version — it's still a huge improvement

@DavidCarlisle So I'd say we can live with it until we can find someone to fix it for us! (tex.sx bounty??)

David Carlisle

@WillRobertson yep. Could consider having an option to make \mathit basically be \textit but that has other side effects and probably adds confusion to an already confusing array of options. Hopefully xetex can acquire a hook to fix this.

Will Robertson

@DavidCarlisle The bigger worry for me now is how to update unicode-math with sufficient backwards compatibility to keep people happy and also without ending up with a boggling number of interface choices. Lucky I have a couple weeks before this conference :)

David Carlisle

@WillRobertson why longtable v5 is in github and not on ctan (and most of the intended updates are not even checked in to github) not sure there is ever a good answer. (see starred comment on the right:-)

Will Robertson

@DavidCarlisle :) Reminds me of the "I assure you [my problems] are still greater" quote — yes unicode-math is very small fry in comparison!

David Carlisle

@WillRobertson not really, I don't think I've ever needed a longtable in any document I've ever done, but math fonts are somewhat more core, sorry:-)

Will Robertson

14:58

@DavidCarlisle Hah!

@DavidCarlisle I've used long table in the documentation for unicode-math, so where does that put us? I haven't used unicode-math for my real work yet either, but perhaps this is the year for it.

Joseph Wright

@WillRobertson I do use siunitx ;-)

Will Robertson

@JosephWright I use siunitx ALL THE TIME

Joseph Wright

And I have the same 'what can I change' problem (v3 would be lovely if it were not for users)

Will Robertson

@JosephWright So thank you again :)

Joseph Wright

@WillRobertson No problem

David Carlisle

15:04

@WillRobertson seriously I think the logical conclusion of all the work on the 2015 release to support unicode texs in the kernel is that fontspec/unicode-math somehow migrates from "contrib" to "core" status. (not sure what the expl3 code implications are though, bit backwards to have latex2 core depend on "exp(erimental)l(atex)3")

@WillRobertson I've never used it, if that helps

Will Robertson

@DavidCarlisle That idea scares me very much. I think both packages would need a solid re-think before adopting them. But I agree I would love to start re-working the NFSS to better allow the layering we've done.

(I did convert the entire NFSS to expl3 a while back, and then didn't know what to do with the edge cases.)

David Carlisle

@WillRobertson scary it may be but saying latex2e kernel supports xetex but by default you can only use 8bit fonts doesn't seem like a story that you can really sell.

Joseph Wright

@DavidCarlisle Doesn't stand for experimental anymore ;-)

@DavidCarlisle, @WillRobertson Re. the LuaTeX stuff, I'm planning to re-add l3luatex to expl3 to cover the bits I am certain we need: a name for \directlua and related things. OK?

Will Robertson

@DavidCarlisle Aye. But then, 2e doesn't support graphics out of the box either :) Just joking. I fully support something happening in this space—it will become a little more clear to me in the next 6 months what my involvement will be able to be. (As always I hope for more but…)

@JosephWright Certainly. Definitely makes sense.

Joseph Wright

@WillRobertson One for the team meeting agenda ;-)

@DavidCarlisle The release approach to l3kernel is nowadays very careful: nothing is getting added that will break stuff without careful notification

David Carlisle

15:10

@WillRobertson well that's the point, it does, that's why graphics/color etc were in the Required packages core distribution, it's not defined in the format, but then neither is \section

Joseph Wright

@DavidCarlisle, @WillRobertson Perhaps we can sort \tl_to_lowercase:n soon as it's the one case we know needs a breaking change

Will Robertson

@JosephWright Oops, I thought we had :) Sorry

Joseph Wright

@DavidCarlisle I should probably do graphics support for expl3

@WillRobertson No: still no final agreement on the one use we do need for \lowercase, making 'odd' chars

Will Robertson

@JosephWright By "do" do you mean a (xtemplate-based) user interface?

Joseph Wright

@WillRobertson No, I mean something at the expl3 level: I did box scaling and the like (still formally experimental) but didn't do graphics inclusion

Will Robertson

15:15

@JosephWright Ah, I see what you mean. Hey, we really should organise a formal agenda for this meeting don't you think?

Joseph Wright

@WillRobertson Well first we should agree we are having some kind of meeting (of the team specifically), but as this is perhaps a unique opportunity for a good number of us to meet up physically it seems sensible to me

@WillRobertson One for the team list I guess

@WillRobertson Email coming up

David Carlisle

15:32

@ChristianHupfer no need to delete the whole answer, just delete the \

Joseph Wright

@ChristianHupfer What @DavidCarlisle said ;-)

@DavidCarlisle, @WillRobertson Hope team list mail gets the tone right: I've no idea what the usual is at TUG meetings in the regard!

@DavidCarlisle, @WillRobertson I guess I better finish Greek case changing!

Hello @StefanKottwitz

Stefan Kottwitz

@JosephWright Hi!

Joseph Wright

@StefanKottwitz How are you?

David Carlisle

@JosephWright we've varied from just chatting over dinner to going a day early and getting the local organisers to allocate a room etc...

Joseph Wright

@DavidCarlisle Indeed

@DavidCarlisle I'm thinking perhaps the first approach

Stefan Kottwitz

15:35

@JosephWright Too busy these days :-o but fine

Joseph Wright

@StefanKottwitz Ah

@StefanKottwitz Not I think as busy as Martin

David Carlisle

@JosephWright, @WillRobertson so who wants to explain the logical nature of tex math font encodings?

1

Q: Why does \mathbf work for uppercase Greek but not for lowercase

I was always using \boldsymbol to make Greek letters bold, but today I made a typo and found that \mathbf also worked for \Omega. Now I am curious, why does it work for uppercase and does not for lowercase Greek? May I rely on this behaviour, or may it change across various LaTeX distributions?

Joseph Wright

@DavidCarlisle That's not our fault :-)

@DavidCarlisle Let @egreg do it!

Will Robertson

@DavidCarlisle This is exactly why \mathbf needs to default to working with greek in unicode-math :)

Joseph Wright

@DavidCarlisle Another reminder that one way or another we do have to move on from pdfTeX

@WillRobertson Indeed

Stefan Kottwitz

15:38

@JosephWright I hope you are fine too? I see you are progressing towards the TUG meeting

David Carlisle

@WillRobertson It would if it loaded a proper large text bold font:-)

Joseph Wright

@DavidCarlisle Also a reason for needing the TeX Gyre math fonts (try explaining that to one J Fine)

Arthur Reutenauer

@GonzaloMedina Thanks, I stand corrected. I really wonder where my former maths teacher got “tilda” then - he systematically called this accent that, which is not his name in French, or English, or even Spanish. Strange.

Joseph Wright

@StefanKottwitz Yes, two talks to write (depending on some discussion, possibly on the way there)

@DavidCarlisle Am reading your testing e-mail now

Paulo Cereda

Oh no, people are talking abot TeX in here. We need some cricket info!

Arthur Reutenauer

15:40

@JosephWright Are you talking to him at all these days?

Will Robertson

@DavidCarlisle Not really I think—you still need to do the mathcode remapping to get plane 1 symbols.

Johannes_B

@StefanKottwitz The golatex-time seems to be almost 5 minutes in the future.

David Carlisle

@WillRobertson yes but not for \mathbf

Paulo Cereda

@JosephWright: any dates for this year's UK-TUG meeting? :)

Stefan Kottwitz

@Johannes_B Just enough advance to the spammers ;-)

Johannes_B

15:41

@StefanKottwitz :-) There was no spam for weeks. Just that strange GloTeX Vanillepudding incident.

Joseph Wright

@PauloCereda Need to hassle people about that, but first we need to sort some other issues (@ArthurReutenauer knows the issue I refer to: he had a committee mail about it today)

@ArthurReutenauer Haven't heard from him other than renewal

Arthur Reutenauer

@JosephWright Yes I was about to talk to you about that. Do you want to discuss it in Darmstadt?

Johannes_B

@PauloCereda Vanillepudding

Joseph Wright

@ArthurReutenauer You were not at the UK-TUG meetings were TeX Gyre Math discussion got really heated

Stefan Kottwitz

@Johannes_B I'll check the time. Just have to deliver a coffee to my girl

Paulo Cereda

15:43

@JosephWright I'm looking for the GuIT meeting, but maybe I could go to ours as well. :)

Joseph Wright

@ArthurReutenauer I think so: Kaveh will be there too so it will be a good time to do it

Will Robertson

@DavidCarlisle Oh, I guess it'd work, but I'm still not sure that's how we'd like to recommend it happen :)

Johannes_B

@StefanKottwitz Kuchen as well?

Arthur Reutenauer

@JosephWright Now you mention it I think I heard about it.

@JosephWright Sure, let’s do that.

David Carlisle

@WillRobertson That's what I'd expect \mathbf (or \mathit) to do.

Will Robertson

15:44

@JosephWright I missed out on opentype math fonts gossip?? Perhaps you can fill me in in Germany

Stefan Kottwitz

@Johannes_B If I would call a delivery service ;-)

Joseph Wright

@WillRobertson Yes, probably the best plan

Paulo Cereda

@Johannes_B Das Vanillepudding. :)

Johannes_B

@StefanKottwitz golatex.de/viewtopic,p,73214.html#73214 Last paragraph. Your whole family could come too :-)

Will Robertson

@DavidCarlisle Right—if \gamma is defined as being a plane1 math gamma, then switching to \mathbf for a font with bold symbols in the same slot should do the right thing.

Joseph Wright

15:46

@DavidCarlisle Could you remind me where luaregisternames is?

David Carlisle

@JosephWright github.com/davidcarlisle/dpctex/tree/master/luaregisternames

sorry should have attached a copy:-)

Joseph Wright

@DavidCarlisle Ah, I checked GitHub but looked for a separate repo, not inside the general one

David Carlisle

@WillRobertson No I think \gamma should be a gamma and then \mathbf should not use the plane 1 slots so you get the normal bold gamma out of a normal bold text font (your \symbf should do the mapping version) but there should be an option to make \mathbf=\symbf

Will Robertson

@DavidCarlisle But what is "a gamma"?

David Carlisle

@WillRobertson U+03b3

@WillRobertson these should do the same thing \mathbf{γ} and \mathbf{\gamma}

Christian Hupfer

15:54

@DavidCarlisle @JosephWright: Nope, my first guess was completely wrong, it's better to drop that answer

Will Robertson

@DavidCarlisle But if you don't remap symbols, then these glyphs are intended for text purposes according to Unicode. You might be able to find fonts that have italic (mathematical) greek symbols there, but that's not how the current Unicode maths fonts work.

To be a bit more specific, they use the text slots for upright and the math (plane 1) slots for italic.

David Carlisle

@WillRobertson yes you need to remap if using the math font but \mathbf should set the mathcodes to use the natural slots so that \mathbf and \textbf are using the same glyphs if using the same font

Joseph Wright

@ChristianHupfer Your call

Will Robertson

@DavidCarlisle Indeed, that'll be how the next version of unicode-math works :)

David Carlisle

@WillRobertson so if I specify a bold text font for \mathbf , and that font has greek and cyrillic, then the greek and cyrillic goes bold doesn't it?

Will Robertson

15:59

@DavidCarlisle I guess I'd have to check, but I don't see why not—actually adding official support for setting up Cyrillic is on my todo list somewhere.

David Carlisle

@WillRobertson sounds good to me:-)

Will Robertson

user image

This is from literal input \def\XX{ffiЖγ}

okay i'm definitely up later than i should be. catch you all soon i hope :)

Joseph Wright

@WillRobertson See you

David Carlisle

@WillRobertson sleep well:-)

cfr

16:36

What's with the sudden fascination with pst-barcode?

David Carlisle

@cfr │┃│┃┃┃┃│┃│┃┃┃│┃│

Joseph Wright

@cfr New user editing a tag in

Stefan Kottwitz

@Johannes_B time corrected

Johannes_B

@StefanKottwitz Nice, thank you. That was driving me insane. :-)

David Carlisle

22 hours ago, by David Carlisle

@Johannes_B maybe that already happened?

Johannes_B

16:47

Chinese und seine Suppe

barbara beeton

17:32

@DavidCarlisle and @WillRobertson -- sorry to butt in here. one of the reasons for adding the math alphanumerics in plane 1 was so that they could be searched in output files. if you say that it's simply an ordinary bold gamma, then it can't be recognized in searching as math at all. please think about this before "freezing" anything.

Joseph Wright

@barbarabeeton I'm intrigued by that

@barbarabeeton At the point of typesetting you loose all of that semantic stuff: you are making marks on a page which need human interpretation

David Carlisle

@barbarabeeton I think making the typeset result readable takes precedence, the plane 1 characters (in all current fonts) are unusable for multi-letter words, so not usable for \mathbf \mathit etc.

barbara beeton

@JosephWright -- on paper, yes. in a file (text or pdf), no -- that's what unicode partially based their decision on. unicode does require markup for some things (and mathml should take care of that).

David Carlisle

@barbarabeeton mathml in some contexts, latex in this one, surely.

Joseph Wright

@barbarabeeton But that's a special pleading for one area: what makes maths so special that it gets this whereas all other areas have to rely on people

barbara beeton

17:40

@DavidCarlisle -- yes, i understand that plane 1 characters aren't meant for multi-letter words, but every single manuscript file i've seen here uses \mathbf{x} for a bold "x" in math. you would do away with that. i don't want to have to explain to ams authors (and production staff) that they need to change that to \symbf{x}!

David Carlisle

@barbarabeeton no neither do I, final call is with Will but I hope the \mathbf/\symbf distinction isn't really exposed to end users, just that both methods are available and when setting up a font whoever is writing the font package has to choose. But for all current fonts the default for \mathbf is best set to be a normal text bold font

@barbarabeeton yes and currently they get a bold x from a text bold font when they do that, all I'm saying is that should be the case also if you use xetex.

barbara beeton

@JosephWright -- no. math is (to unicode) just another "language". why else would there be separate "A" shapes in ascii (latin), greek, and cyrillic? unicode hates duplication!!! this is not just a matter of appearance; it's semantics.

@DavidCarlisle -- computer modern doesn't have a bold math alphabet because knuth didn't have enough room, either memory, or families.

David Carlisle

@barbarabeeton It doesn't have a non bold math roman either, but still as I say above I do not see what you lose if \mathbf is defined to be the same in uniocde-math as classic cm setup.

barbara beeton

@DavidCarlisle -- i will stop my whining if you can assure me that the output (pdf file) of \mathbf(x) is there as a plane 1 character that can be searched as such.

by the time text + math gets to pdf, the only thing that will identify a bold symbol "x" as math is its plane1-ness. (\mathrm is only very rarely used for single letters.)

David Carlisle

17:56

@barbarabeeton no it would be an ascii a

@barbarabeeton yes but how come it's Ok for log to be ascii so no specific math hint in the unicode text, but a disaster if Hom is encoded as ascii Hom ?

@barbarabeeton and surely \mathbf and \mathit should have similar definitions?

barbara beeton

@DavidCarlisle -- you mean \mathrm{a} ? (in math, as in journals and books here, single letters are almost invariably (math) italic; only operator names are usually roman.)

Joseph Wright

@barbarabeeton I guess what we have here is a limitation of the fact that Unicode tried to ignore formatting and it was then added in for maths, whereas in fact formatting is a series of axes (as seen in NFSS) along with others (such as case) which map out a multi-dimensional surface.

barbara beeton

@DavidCarlisle -- Hom is a darn good example. and i think i'll send it off to murray to see what he says about how to recognize "log" or "sin" in a search. it could possibly make them think about adding the latin alphabet to plane 1 for math.

David Carlisle

@barbarabeeton yes in mathematics as a subject I know, all variables are called x and all functions are called f, I was a professional mathematician for a long time:-) but there are many other disciplines that use mathematical layout where identifiers are commonly multi-letter and commonly use font variations.

@barbarabeeton it's no good adding more math alphabets to plane 1 unless the fonts add kerning and ligatures for those characters.

barbara beeton

@JosephWright -- unicode was acting on their usual principles of "no unnecessary duplication", and preferably no duplication at all if they could get away with it. they're loosening up. if you look at recent entries, you will find a number of "duplicate" shapes, with different meanings (usually associated with spacing differences on account of usage), being added.

egreg

18:04

Back from my tour in the mountains.

David Carlisle

@barbarabeeton so to be specific what unicode characters would you like to see from $\mathit{apples} + \mathit{pastry} = \mathbf{pie}$ (question is serious, even if expression is not)

Joseph Wright

@barbarabeeton Yes, I understand that but a multiple-axis scheme doesn't mean duplication, it means having different identifiers for different things (e.g. A = a = \textsc{a} in some contexts so we could store just 'a' and then a case axis)

barbara beeton

@DavidCarlisle -- you're ignoring the fact that \mathit does use the \textit font. the reason for the name is a matter for historical speculation, or asking dek. i can't say what the unicode committee will decide on a particular topic; i can only say what i know they've decided in the past, and the grounds on which the decision was made. that's why i suggested throwing this question out to murray, explicitly asking about the searching aspect.

can we work together on coming up with a really thorny and ambiguous text that we would like to know how to input/encode in order to search?

Joseph Wright

@barbarabeeton I think we can look forward to some interesting/useful discussion at the TUG meeting!

David Carlisle

@barbarabeeton \mathit uses the same font as \textit and \mathbf uses the same font as \textbf, there is a pleasing consistency there that you seem to object to?

egreg

18:11

@DavidCarlisle Not really: the encoding for \mathit is OT1, for \textit it may be different.

barbara beeton

@DavidCarlisle -- maybe there's a surface consistency, but i think it's only surface. i'm even willing to ask don.

Joseph Wright

@barbarabeeton I guess the interesting thing for me here is that you feel you should be able to do this search at all. With my day-job hat on as a chemist, I can't do the same for a chemical name and expect only to be able to search on an ASCII basis in a PDF. For 'real' chemical data I expect to have to use a structured database.

David Carlisle

@barbarabeeton it doesn't have to be thorny, look at any category theory paper and look for \mathbf{Var}

@barbarabeeton huh?? how surface can it be? in plain tex and by default in latex it's the same font.

barbara beeton

@JosephWright -- and i've heard discussions at the utc regarding chemistry. they won't go near it! but there really are no structured databases for math. i can't even drum up any support for creating a symbols list by area of mathematics.

David Carlisle

@barbarabeeton taking \mathbf{Var} then if you use the plane 1 alphabet in any of the currently available fonts there is no kerning between V and a and it looks horrible, if you do as tex has always done and use a text bold font then you do get kerning and its readable.

barbara beeton

18:14

@DavidCarlisle -- the reason, as i said, is that there weren't enough families, and not enough memory for another distinct math font.

David Carlisle

@barbarabeeton have you seen my latest answer on site? :-)

Joseph Wright

@barbarabeeton I wouldn't suggest going near any arbitrary subject: I'm sure we all have our own conventions :-) (My specific one here is that chemicals in a paper are given bold numbers normally, but if I want to search a paper I'm just looking for numbers, so it's actually easier to look through by hand than ask Reader or whatever to search.)

David Carlisle

5

A: Why does \mathbf work for uppercase Greek but not for lowercase

TeX was designed at a time when using 8bit fonts with 256 characters was considered rather exotic, so while it could do that all the standard fonts only use 128 characters per font. That combined with the limit on 16 math families per expression meant that some "interesting" design decisions had ...

barbara beeton

@DavidCarlisle -- i agree for multi-letter bits. and thanks for linking in your answer that says the same thing about limits. what i care about is the single bold math letter. remember, unicode didn't exist when tex was hatched. personal computers, laser printers, pdf, and the web didn't either. the challenge now is to make "best" use of all these "new" tools. oh -- don't forget accessibility; that's by and large simply ignored these days, but won't/can't be forever.

@JosephWright -- looking "by hand" (by eye?) is practical if you've narrowed something down to one paper. what if you're searching 10 years' worth of papers for a particular compound? i guess that's what your database is for.

Joseph Wright

@barbarabeeton I can't speak for maths but for chemistry I don't see accessibility from typeset works being realistic: need the kind of structures that XML is good at

@barbarabeeton One should never be searching the original literature directly, that's what abstracts, databases, etc. are for (For a start, the legal situation means you can't have access to all relevant data.)

barbara beeton

18:28

@JosephWright -- i don't disagree with the principle that abstracts and databases should be searched first. for math, that should identify the items covering relevant concepts. but i've spent hours poring over pages looking for single symbols -- even though i already knew (because i had the tex source) they were there, and even approximately where. that's how the stix info for unicode was compiled. it's thankless.

Joseph Wright

@barbarabeeton Like I said, I think there is useful discussion to have on this area at the TUG meeting

barbara beeton

@JosephWright -- re accessibility, i can see why chemistry isn't a good subject for a sightless person to be pursuing. but math is different in that regard. there are perhaps not many blind mathematicians, but it's certainly not out of their reach. and the day is going to come when one or more of them is going to cite the american disabilities act and insist that the mathematical literature be made available to them. (it can be done. unicode is part of it. but it needs proper thought.)

@JosephWright -- this area and several others. i'm making my list.

Joseph Wright

@barbarabeeton Can that really work? Straight text is OK but the moment you have a table or figure the PDF version can't describe semantics and you need the XML data version to describe the structure properly

barbara beeton

@JosephWright -- figures would have to be explained in words. an xml structure for tables is probably reasonable, but would need a well-conceived audio (or braille?) output. but the math i'm talking about is more likely to be "equations" (formulas of various kinds). see cs.cornell.edu/home/raman/aster/aster-toplevel.html and allthingsd.com/20130604/…

Joseph Wright

18:48

@barbarabeeton I was thinking that the moment you have say a matrix then anything relying purely on Unicode for semantics is in trouble

barbara beeton

@JosephWright -- for anything remotely two-dimensional, unicode insists that markup is required. that's why (la)tex source is requested by mathematicians who can't read a paper on paper, and the ams will provide that on request, after checking into the circumstances.

David Carlisle

@barbarabeeton I don't think it's a is documentable interface to tell people to use different markup for a bold V and a bold Var, if if the math font plane 1 alphabet isn't suitable for Var it isn't suitable for V.

barbara beeton

19:03

@DavidCarlisle -- i didn't say it was easy, and i didn't say i had an answer. i'm just asking for you to think about how a letter might be made searchable in the output, which is one of the rationales for why the plane 1 alphanumerics were encoded. (i've seen operator names rendered in fraktur, for goodness sakes. mathematicians will do what they want, but if searching is a criterion, then some rational approach needs to be found.)

Joseph Wright

@barbarabeeton /ActualText?

David Carlisle

@barbarabeeton yes agreed (making, or at least trying to specify, accessible math markup has been a large part of my day job for last 15 years:-) but I think the plane 1 math alphabets are at best only a slight hint in that area, and you need markup structure or tagged pdf for any real distinguishing of the math use, so if bold x comes out as U+0078 in times bold font, I wouldn't lose too much sleep over it.

Joseph Wright

@DavidCarlisle Yes, I suspect you are right

cfr

19:19

@JosephWright Oh, right. Yes, I see. I just looked at the front page and it seemed an odd fashion to have suddenly hit.

Johannes_B

I love soup.

2

Joseph Wright

@DavidCarlisle Anything further to do on the LuaTeX stuff at this stage?

David Carlisle

@JosephWright wait, then post:-)

David Carlisle

19:59

@egreg excellent sporting news: bbc.co.uk/sport/0/olympics/33298267

egreg

20:18

@DavidCarlisle That's shocking!

@Johannes_B Thanks for having pinged me for the openbib question.

Johannes_B

@egreg Someone has to answer :-) You were out all day, i figured you could need a bit help to get capped ;-)

Joseph Wright

@DavidCarlisle :-)

egreg

@Johannes_B That was not a problem. :)

Johannes_B

@egreg :-p

egreg

@Johannes_B And I even had ice cream. :) In one of my favorite places in view of the Dolomites.

Johannes_B

20:23

by the way, who starred that soup post? :-)

@egreg That second part sounds nice, ice cream ... not as much. Do they serve soup?

Joseph Wright

@Johannes_B That would be me: be glad it's not mod-pinned :-)

Johannes_B

@JosephWright :-D Do you love soup as well?

egreg

@Johannes_B The place only sells ice cream, milk and cheese.

Johannes_B

@egreg I am getting a cart and seel different soups on the side of the street, how about it? Would you buy a pot of nice heart warming soup?

hm, i would need chairs, or something for people to sit on.

Hammocks don't seem like a good idea with hot soup.

egreg

@Johannes_B No, thanks. What about a Wurst in front of a beautiful mountain lake, like I had for lunch?

Johannes_B

20:29

@egreg :-) What kind of Wurst?

egreg

@Johannes_B Bratwurst

user image

@Johannes_B There was no choice, actually. Standard Südtirol Wurst, I guess.

Johannes_B

@egreg That is beautiful. Never been to Italy before. Should definitely fill that gap one day.

cfr

Is this on topic?

0

Q: Writing chemical reaction with WP-Quick latex?

I want to show series of some chemical reaction on wordpress using popular chemfig and mhchem package however i always encounter some problem . For example . I want to show reaction in series like this : A + B -> C D + E -> F and much more like that In Wordpress post i post like this isn't it...

chemfig mhchem quicklatex

What is WP-Quick LaTeX?

egreg

@Johannes_B It's a side valley of Pustertal, which is a very beautiful place, from Brixen to Lienz.

cfr

@egreg Prydferth!

egreg

20:38

@Johannes_B From home to that lake it's 231 km.

@cfr Yeah! Dolomites are the most beautiful mountains in the world. Well, very high ranked anyway.

Johannes_B

@egreg Freiberg --> Berlin: 236 km :-)

1

Q: changing font of wordpress quick latex

The font of word press quick latex looks so light and not so good. Is it possible to change the font a bit dark and a bit cool? and i am writing chemical reaction so i must use wp quick latex as i have not found any other plugin that writes chemical reaction on word press on fly. I want to show...

egreg

@Johannes_B Even without high mountains. For the journey back I climbed four mountain passes. Of course, the route was completely different.

Johannes_B

@egreg I am taking care of the dog next week. I want to walk 30 km at least once. I hope it doesn't get too warm.

@PauloCereda soundcloud.com/johnnyjewel/chromatics-into-the-black

cfr

21:00

@JosephWright He's the author of the package... ;).

@Johannes_B Thanks.

Christian Hupfer

21:27

@cfr A WordPress (WP) plugin for LaTeX which is quite comfortable. It's possible to load traditional packages

cfr

21:45

@ChristianHupfer Thanks. Yes. I followed the links from the question posted above and read a little bit about it. Sounds interesting.

Christian Hupfer

@cfr I used it on my WP site, but some pages are hidden so far

1 hour later…

Arthur Reutenauer

22:50

@JosephWright @DavidCarlisle @WillRobertson @egreg I suppose you’ve all booked your hotel room for Darmstadt? If not we need to talk ;-)

egreg

@ArthurReutenauer Yes, I have.

Sean Allred

23:43

@DavidCarlisle I found something very special in a bookstore today

user image

but alas, it is a different image :( I was convinced earlier

370

A: How can I explain the meaning of LaTeX to my grandma?

It does this but it uses a computer and so requires less manual labour. (image from Wikipedia)

« first day (1700 days earlier) ← previous day next day → last day (3530 days later) »

Transcript for

Jun '1527

$TeX - LaTeX$ TeX, LaTeX and Friends

\begin{chat}

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2025 Stack Exchange Inc; legal

mobile