Lol, then what you propose is reasonable

@manooooh Start with \newcommand{\diff}{\mathop{}\!\mathrm{d}} (if you really want an upright “d” (bleah). This way you can simply type \diff x and not worry about \,

\iiint_M \div \vec f \diff x \diff y \diff z seems much more readable, doesn't it? No braces around the integrand.

@egreg you are right in everything, as usual. With regard to the braces, I had written the code a few months ago, so the code was outdated, but with the differentials my hands are already used to write \,\mathrm d, so implementing that feature will cause a mind blown, but I really want an upright "d"!

@egreg Would you recommend \[\iiint_M \div \vec f \diff x \wedge \diff y \wedge \diff z\] with your definition of \diff?

@marmot LOL, it fits very well \div in this context!

@manooooh I get a freaky sign that I know from the calculates of my childhood.

@marmot the \div sign indicates "division", right?

@manooooh I guess @egreg had redefined it to become div (divergence).

@marmot oh okay. I mean that if it really means "division", I believe that no one in life has ever used it, both for the blackboard and for a rigorous mathematical document. Who came up with that command? :P

@manooooh Probably the Egyptians. It looks very much like the stuff they wrote on their obelisks.

Oh, good point

@manooooh I was wrong, it was the Mayans.

@marmot Nobody needs the standard \div symbol. ;-)

@egreg I agree. The Mayans are no longer around.

Hahahah

@manooooh \let\div\relax \DeclareMathOperator\div{div}

@manooooh Or \DeclareMathOperator{\rdiv}{div} (“real” div)

@marmot The \wedge should be needed, but it's customary to omit it

@marmot If you're worried about the spacing, don't be.

@egreg where does r in \rdiv come from?

@egreg Well, you need to give the students the chance to see it once before you tell them that you are going to omit it. ;-) (Yes, it is the spaces, but it is almost invisible.)

@manooooh Just popped to the top of my head: “real div operator”

@egreg ... as opposed to \idiv (imaginary div operator) and \gdiv (Grassmannian div operator) ;-)

@marmot TeX inserts just a medium space between a binary operation symbol and an operator. It's similar to a+\sin x.

@marmot The second misses the thin space between \vec{f} and dx, which is needed

@egreg Yes, good point!

Good night all!

Good night and THANK YOU!

Buona notte!

In TikZ, when dealing with polar coordinates (e.g. 30:3cm), what is the reference point used by the system

@subhamsoni The origin at (0,0) and the positive x-axis has polar angle 0.

@marmot @marmot thanks. In PDF, where can be locate (0,0). Is it at the exact centre of the page

@subhamsoni No, it is not the center of the page. It might even not inside the page. TikZ determines the bounding box from what you draw in the picture. And then it will put the picture where you add it to the text unless you use overlay, in which case you can use absolute page coordinates.

@marmot @CarLaTeX wanna play Pinturillo?

Does anyone know of any repository about documents made with (La)TeX in which you know that I can extract ideas, styles, formats, etc. that are approved by you?

Probably arXiv? But what concrete examples are the most illustrators? What documents are those that you approve? There are many

I am looking for math documents, if possible

I am breaking a \tableofcontents in beamer with the option allowframebreaks; however, the break has a huge gap between the subsection bullet point that starts on the new frame and the next section. Is there anything that can be done about this or any suggestions on a better way to approach this problem?

@CarLaTeX -- Speaking of theses, ...

@JosephWright and @UlrikeFischer -- But that's current tar pits. Visiting the La Brea Tar Pits is a very interesting experience and I recommend it highly. Of course the creatures that wandered into those tar pits while they were active probably weren't too overjoyed about the experience.

@barbarabeeton I worked on it this morning :)

@CarLaTeX -- Excellent! (My husband and I spent most of yesterday helping out a friend whose computer decided to misbehave while he is working on his dissertation in art history. He has hundreds of files, both text and graphics with a 400-page annotated bibliography (all in QuarkXpress) and his computer refused to let him back it up. Potential tragedy! Still problems, but at least there's hope now.)

@egreg -- Only if you're teaching arithmetic in primary school. I'm sure DEK had a reason for providing it; it may make sense in some of his exposition of algorithms.

@barbarabeeton Argh!

@CarLaTeX -- Understatement.

@manooooh -- I urge you to not put the \[ ... \] in the system environment. It's probably infrequent, but if you ever need to include something else in the same display as the system, you're shooting yourself in the foot. (Actually, your example problem using two systems side by side is a demonstration that this is a bad idea.) it's not so hard to input the \[ ... \] when needed, and it will be clearer to a reader of the source.

@barbarabeeton you got mail. :)

Is anyone here using TeXShop on an OS earlier than Mojave and suddenly having problems? Mine suddenly started hanging completely.

@AlanMunn Let me check, hold on.

@AlanMunn it seems to be working here.

@PauloCereda It now seems to be working for me again too. Very strange behaviour though. I guess I'll keep using it and see if anything breaks.

@AlanMunn I will give it another go during the day and see if something odd comes up.

@AlanMunn I thought macOS never hangs. Seems it does hang just like Windows.

Which makes me not want to get the new Macbook Air 2018 even more.

@Jasper he's talking about an app, not a system...

@Jasper Well the OS doesn't really hang. It's easy to force a hung application to quit. In my experience the Mac OS is very robust.

@PauloCereda And he's gone... :)

@AlanMunn oopsie. :)

@PauloCereda Do you have Mojave running on anything yet?

@AlanMunn No. My Mac isn't supported. :(

@PauloCereda An antique! And I guess no incentive to get a new one any time soon?

There are some patches available, but Mojave on an unsupported Mac hardware can be very glitchy!

@AlanMunn Not with our prices. :)

@AlanMunn ^^ the cheapest one!

@PauloCereda That's crazy. About 2.5 times the US price.

@AlanMunn Yes. I love Apple hardware, but I cannot afford these prices.

@AlanMunn My iPad does not upgrade to version 10.0 or higher anymore...

@PauloCereda Same with mine and my iPhone too.

@PauloCereda I'm stuck at 10.x for them, I think.

@AlanMunn Oh. :) The number of apps I can install are now greatly reduced. :)

@AlanMunn: Ross and Will brought their powerful Macbooks to Rio. Thankfully, @JosephWright didn't leave me alone and brought his Dell laptop (I had one too). :)

Frank was a Mac user as well, IIRC.

@PauloCereda I've been hanging on to my 2015 MBPro because I hate the new one. C. has had lots of problems with hers, and is now back to using her old MBPro too.

@AlanMunn Interesting! I have a 2007 MB model running Lion. :)

@AlanMunn and @PauloCereda thanks for the feedback, though I would think that if an app hangs, it might have something to do with the OS. After all, what is the OS if not for the apps that we use on top of it?

@PauloCereda Hehe, Apple products are so expensive. Maybe the cheapest one there is the old Macbook Air 2017? The only downside is that it doesn't have a Retina display.

@AlanMunn Yes, I wonder why they have such different prices in different countres.

@PauloCereda I have been thinking of getting a Dell but I keep reading its laptops have fan problems. Sometimes the fan keeps spinning even though things are not hot.

@Jasper I've had no issues with my XPS, nor has a friend with the latest model

@JosephWright That's good to hear. Of course, the problems I was referring to happen to both the cheap Inspiron and the expensive XPS. The Dell XPS 13 and Dell XPS 15 are their most expensive machines for non-gamers.

@Jasper Yes

github.com/tabu-fixed/tabu/pull/14 in case someone wants to have a go at reviewing

@mirabilos I'm very short of time now, but hopefully someone can look it's pretty broken now so you are unlikely to make it worse....

@DavidCarlisle tex.stackexchange.com/questions/476472/… - fair?

@JosephWright yes probably, I can't remember it coming up before, I suppose most of the packages got written then never changed, so deleting stuff never came up

@DavidCarlisle I was thinking of e.g. the oldcomments and such in the core sources

@JosephWright In some XPS models the webcam is placed at the bottom or at the side of the screen, which is stupid. In the latest models the webcam is correctly placed at the top of the screen.

@Jasper Mine is at the bottom, so is my friend's with a new one. To be honest, I don't find any webcams that great.

@JosephWright They are not great, but they are good enough to serve their purpose. For you to chat with someone online face to face, or for you to make youtube videos. 720p is just good enough.

@JosephWright are that, one of my finer typographic efforts:-)

@barbarabeeton you quoted an example where \[ and \] were in the preamble of the document, not inside the document environment. So I do not understand what you meant

@barbarabeeton you quote the example where in the same environment where you need more than one system. I need this kind of examples, so put the \[ and \] inside the definition of the system will not allow it to achieve this requirement. So the solution is to use \[ and \] inside document

@manooooh -- What is important is not that the \[ ... \] are in the preamble of a document, but that they are in the definition of an environment. That means they will be applied every time you use that environment. Each one will start a new display; they cannot be set side by side. Your next comment above shows that you do understand the problem when it's explained. Good thinking.

@barbarabeeton sorry, I still do not understand. I understand that it is advisable to write the delimiters in the definition of the environment (as it happens with align), but if you need to write something around it then the delimiters should go in the document, not in the definition, right? If it's wrong, what solution do you propose?

I need to write stuff around system (same if I would use align)

@DavidCarlisle oh okay, thanks… that’s very sad, as tabu is the only sensible table package for so many uses…

@mirabilos tabu contains some interesting ideas but ... it's a car-crash interface wise

@manooooh -- The "delimiters" that specify a display environment should be used with care in definitions. (I've put "delimiters" in quotes here, since the usual meaning is for visible "fences" in a math expression, not the start and end of an expression or environment.) Consider the difference between "outer" environments (equation, align, gather, etc.) and "subsidiary" environments (split, aligned, gathered, etc.). The outer environments establish math mode; (cont'd)

(cont'd) the subsidiary environments do not -- they must be used only inside an already existing math mode scope. Thus your system, since more than one may be needed within a single display (math mode), should be defined as a subsidiary environment, and the math mode needs to be established within the document level, not as part of the new environment.

@Jasper No programming language will keep you from writing faulty software. It's a bit unfair to blame the OS for such loopholes...

@Jasper I have 4 Dell machines and no complaints at all. :)

@mirabilos the author has been saying it's unsupported since around 2011, it's been running on autopilot but finally broke, I set up the github so people have a place to report issues and suggest fixes (thanks) but I simply don't have bandwidth to undertake to try to fix all the issues reported.

@DavidCarlisle ok

@DavidCarlisle here we go again? :)

@mirabilos tabu was never really my favourite table package, it breaks many latex syntax conventions, especially teh >\raggedright syntax which it redefines in a way incompatible with every other latex package.

I’m willing to try and fix things that pop up, but my TeX skills aren’t very high

yeah, the syntax is awful

but it works well

and consistently (in the output and accepted inputs, not the syntax)

@DavidCarlisle, @JosephWright ^^

true that

@PauloCereda What do you mean by "any breaks change something" ? ;-)

@marmot tex.stackexchange.com/questions/476472/… “That goes with the fact that in (La)TeX someone will depend on every single internal macro you have”

@marmot :)

@PauloCereda -- Why limit it to LaTeX?

@barbarabeeton Oh I will update accordingly. :)

@mirabilos Sure, without \marmot my avatar would be a bit silly.

hello

I have a question for any latex gurus out there

I have an array in an equation environment, and I'm trying to make an l shaped box around three of the elements

does anyone have a good suggestion as to how to do that?

@Anthony There are many options and the answer depends on how fancy the shapes will be. If you want to have highly customizable, fancy shapes, tikz with the tikzmark library may be an option. Other options include a plain \fbox or tcolorbox.

@marmot thanks for the reply - I don't really want anything highly customizable, I'm just looking for a way to have three fboxes in an array, in an lshape, that are joined rather than separate

@Anthony A tabular, then?

@Anthony Something like this?

\documentclass{article}
\usepackage{amsmath,array}

\begin{document}

\[
\left[ \,\begin{array}{@{}ccc@{}}
  \cline{1-2}
  \multicolumn{1}{|c}{1} & \multicolumn{1}{c|}{2} & 3 \\
  \cline{2-2}
  \multicolumn{1}{|c|}{4} & 5 & 6 \\
  \cline{1-1}
  7 & 8 & 9
\end{array}\, \right]
\]

\end{document}

@egreg you're the best

that's exactly what I wanted

thank you

@Anthony Change the last row into \multicolumn{1}{@{\hspace{\arrayrulewidth}}c@{\hspace{\arrayrulewidth}}}{7} & 8 & 9

you're a wizard

what do the @{}'s mean?

@Anthony Remove the padding before the first/last column.

I see - but more generally it seems like the @ symbol has some semantics - does it always mean remove space?

You used it in the array definition, but also in the last line you just posted

@Anthony @{...} inserts ... between columns, instead of the standard padding.

I see, thank you!

@vlg please could you start to show complete code? Why should we guess what you are doing?

@vlg It's a code command: you'll probably want a document wrapper:

@UlrikeFischer There is no need for guessing if you can read and understand English.
    \ExplSyntaxOn
\sys_if_engine_xetex:T {
    \ExplSyntaxOff
    \RequirePackage{polyglossia}
        \setdefaultlanguage{english}
    \RequirePackage{fontspec}
        \setmainfont{XCharter}[Ligatures=TeX]
        % \setmonofont{Century Gothic}[SmallCapsFont={TeX~Gyre~Adventor:+smcp}, Ligatures=TeX]% fixme: escape the `:`?
        % \setmonofont{Century Gothic}[SmallCapsFont={Arial}, Ligatures=TeX]%
}%
\ExplSyntaxOff

\usepackage{expl3}
\ExplSyntaxOn
\cs_new_eq:NN \IfXeTeXTF \sys_if_engine_xetex:TF
\ExplSyntaxOff

\IfXeTeXTF
 {<true code>}
 {<false code>}

@vlg ^^^

@vlg TeX tokenizes when text is read: you can't use \ExplSyntaxOff in an argument where the rest of the argument is to be handled with different catcodes

I understand. I assume the same for any catcode transformations is true, eg, makeatletter?

@vlg Yes: it's a feature of how TeX works

@vlg You (I) could use a different construct at the code level, but I doubt that is useful for you

@vlg but you hadn't shown the cause of the problem until Ulrike asked for the example.

@JosephWright It's at least interesting, I'd like to know how you'd go about using a different construct

\usepackage{expl3}
\ExplSyntaxOn
\sys_if_engine_xetex:TF
 {
   \ExplSyntaxOff
   \use_i:nn
  }
 {
   \ExplSyntaxOff
   \use_ii:nn
  }
 {<true code>}
 {<false code>}

@vlg ^^^

@vlg The trick is to switch catcodes first, then read the 'payload' later. This is needed e.g. for verbatim, but to be honest I'd not do it in a document

@vlg [SmallCapsFont={TeX~Gyre~Adventor},SmallCapsFeatures={Letters=SmallCaps},.

@vlg you could use ifxetex package then \ifxetex ... \fi which is more designed fro document use

@vlg More effficient:

\usepackage{expl3}
\ExplSyntaxOn
\use:n
  {
    \ExplSyntaxOff
    \sys_if_engine_xetex:TF
 }
 {<true code>}
 {<false code>}

@DavidCarlisle Yeah, but leaves us loading code inside a conditional

@DavidCarlisle Are the if..tex packages somehow different different to the sys_if_engine _... cmds?

@JosephWright fontspec is Wills code and ifxtex is his conditional so if it doesn't work, we know who to blame..

@vlg well the tests are more or less the same but by design has a name with just letters so it can be used in a document with \makeatletter or \ExplSyntaxOn

@DavidCarlisle True, true

@vlg if you are using \ExplSyntaxOn or \makeatletter in a document it is always a sign that something is not right with the layering, those commands give access to commands that by design were made inaccessible in a document.

@DavidCarlisle Well, it might be ... but the tests for expl3 are hardened up in some places

Also, aren't like all the l3 cmd's and names, and macos and literally everything with a colon and/or underscore somehwer? I know it's mean for programming, but leaving abs. nothing to the user is harsh?

@vlg well you do not need ExplSyntax on/off then, just as you don't need \makeatletter, if the package is loaded with l3 conventios

@vlg It's a layer for programming: that's the entire point!

@vlg yes exactly: expl3 is the programming layer, none of those commands are intended for use in a document/

@vlg For example, we've provided a very small package to make \fpeval available as a document command as it's really really ueful and users shouldn't have to worry about the code-level \fp_eval:n

I'm still not getting how, and or why, I should go about transitioning my package to l3; if necessary at all? There are certainly a few small things that could be beneficial regarding clarity and optimizing

@vlg On 'why', it comes down to if you want/need a large set of programming tools

@vlg Only you can decide if it's necessary

@vlg you don't have to, expl3 provides some features to make it easier to program some things, but if you don't find use for it then there is no need to do so.

@vlg On 'how', start small, I guess, and build things up. We have conventions that don't really mesh well with mechanical translation of classical TeX code.

Even in my testing folder, i get errors and things I don't understand, rereading the modul docs again and again. I might be too stupid to do a transition, given how I'm not too familiar with the TeX basics.

Every time I see @DavidCarlisle do something, it's depressing :D

@vlg well you're in luck,I haven't programmed a large body of tex code this century:-)

@vlg Like I said, start small: try some small ideas. Make them into examples, and ask on the site, or if they are really small, ask here

@DavidCarlisle Er, what about all of that stuff with UTF-8?

@DavidCarlisle You're just showing off. It's like Schwarzenegger saying how he's not been working out whilst looking like a marble statue of Greek god.

@vlg You should meet Bruno (@DavidCarlisle, @UlrikeFischer, @egreg)

@JosephWright most of that was really from xmltex which was 1997 or thereabouts:-)

@DavidCarlisle Well yes, but it was a big change to the kernel

@DavidCarlisle How is your name second on the l3 team if you haven't done anything major?

@DavidCarlisle Let's see if my big plans for the fragile commands come off ;)

@vlg He said 'this century'

@vlg because c is early in the alphabet and I was working in L3 in 1992

@DavidCarlisle And I think Frank hadn't come up with the 'add people in order' idea then

@DavidCarlisle Should be Frank, Rainer, Chris, you, or Frank, Rainer, you, Chris?

@JosephWright chris before me (and some others no longer active)

@DavidCarlisle Ah right: Michael? Alan? Robin?

@JosephWright yes and denys

@DavidCarlisle Ah, right

@DavidCarlisle We should probably record that somewhere: one day, someone is bound to want to know

@JosephWright I think I may have been just before Alan, we got tricked in to joining at around the same time.

Johannes as well of course.

@DavidCarlisle Ah, thought that was later

@DavidCarlisle Frank's probably still got the emails ;)

@DavidCarlisle A common theme, that

@JosephWright well there is no initiation ceremony so joining is somthing of an undefined concept.

@DavidCarlisle True, true

@DavidCarlisle I meant I thought babel was somewhat separate back then

@DavidCarlisle Nowadays I think 'added to the team list' is a good measure

@DavidCarlisle I see that Rainer mentions in tug.org/interviews/schoepf.html that Frank, Chris and he were 'there from the start'

@JosephWright yes

@DavidCarlisle So Frank probably does have the order right, more or less, on latex-project.org/latex3

Hello @DavidCarlisle @JosephWright and all

@touhami no that is a misunderstanding utf-8 covers all of unicode there are no non-utf characters

@touhami what is true is that you may use a character for which there is no code defined with the currently loaded packages . Most of unicode is in that state as theer are a lot if Unicode characters.

@touhami the job of inputenc (or the implicit core of input enc preloaded in recent formats) is to decode the incoming byte stream so that latex knows some byte sequence refers to (say) unicode character 8020, whether latex has a definition to typeset that character not its concern.

@touhami what do you mean by a non-utf8 char?

@touhami if you just mean why does the error message say inputenc then it is still the same code as previously it is just loaded in to the format to save you loading it into the document.

@DavidCarlisle yes this is the point

and yes i know there is \input{utf8.def} in the format

@touhami as I say all that happened (not quite true but not completely false) is that we added \RequirePackage[utf8]{inputenc} to the end of latex.ltx.

@touhami I could have re-written the error messages to use latex@error, instead of packageerror{inputenc} but it really is the inputenc code, just preloaded, so it didn't seem necessary,

but that can confuse new user! and what shoud i say when they come ask me?

@UlrikeFischer thank you for asking as @DavidCarlisle say may be i still misunderstanding

@touhami I doubt that new user care if the message says "inputenc" or "latex kernel" or something else - they would only want to know what to do about the error.

@DavidCarlisle so i can say : the team find no need to re-write the error messages because it is the inputenc code, just preloaded.

@UlrikeFischer by new user i mean users not aware with the changes.

@touhami if they are "old" new users they should know that a lot things in latex can only be explained by "history".

@UlrikeFischer here i think it's logic to ask why i have a message from inputenc since i diidn't load it.

@DavidCarlisle thank you for now. please let me know if any

@UlrikeFischer thank you as well

@touhami I often get messages from packages I don't load myself but which are loaded somewhere in the depth of some other package. Do the users care about this too?

@UlrikeFischer i think yes!

@touhami or another way to think of it is that "inputenc" is the latex input encoding layer and you get error messages from that (just as font loading messages ar prefixed by (font) even though the NFSS code (which used to be a separate package) is preloaded in to the latex2e format.

@touhami :

\def\@font@info#1{%
         \GenericInfo{(Font)\@spaces\@spaces\@spaces\space\space}%
                     {LaTeX Font Info: \space\space\space#1}}%
\def\@font@warning#1{%
         \GenericWarning{(Font)\@spaces\@spaces\@spaces\space\space}%
                        {LaTeX Font Warning: #1}}%

@UlrikeFischer do you? I never get error messages:-)

@DavidCarlisle Out of cheese error

@DavidCarlisle because you do not do any thing else just helping peoples

@DavidCarlisle I'm one of the people who gets contacted by panicking users whose documents don't compile. I know all sorts of error messages ...

good night to all

@touhami that sounds like an epitaph:-)

@egreg The reason you couldn't make sense of the request might actually be "and even challenge some of the best developers" . ;-)

@marmot Huh?

I do not know either lol

@JosephWright Huh!

@marmot Foggy crystal ball. An afternoon in a gym with about 200 kids playing 3 on 3 basketball on eight courts simultaneously can have some side effects.

@egreg Yes, I will play in 30 minutes. I haven't really played for weeks because we had the most rainy winter in decades but now it seems back to normal (some days are sunny and warm, others are warm and sunny).

@egreg Last time you were on the pic, with the nice costume. ;-)

@marmot there are a lot of childs! Are you one of them? :)

@marmot best of luck dear marmot!!! I am sure you and your team will win :)

@manooooh No. We would never allow a duck to be the referee. ;-)

Hahahha

@manooooh Gracias!

@egreg If I look at the left part of your pic this looks more like wrestling to me

@marmot An attempt at a block. :-)

@egreg Do you allow weapons for that? (I sometimes feel that I would set much more efficient picks if I only had a sword. ;-)

@Sebastiano How was your weather. In television they said there could be snow in your region? Was it true?