« first day (5192 days earlier)      last day (26 days later) » 

00:22
Does texdoc.net/pkg/<pkg> still exist? I've always used it as an online alternative to texdoc <pkg>.
00:33
@Werner it's .org these days texdoc.org/index.html
Hello,
I am ashamed to ask for assistance, but I am stuck with the issue and it most likely would not qualify as a fully-developed question to create an inquiry about on StackExchange by going through the usual procedure (for it seems rather trivial).

I can't figure out why alignment is off; the two expressions have to be exactly aligned by the "sgn" operator:
\documentclass[margin=0.5in,preview]{standalone}
\usepackage{amsmath,fontspec}
\setmainfont{Cascadia Code}
\begin{document}
\noindent{}test\\%
00:51
@bp2017 wasn't this asked on site within the last day or so? \phantom{|a|=} is as wide as {|a|=} which isn't as wide as |a|= as the {} form a mathord so you lose the relation spacing to the right of the =. incidentally |a| is wrong it should be \lvert a \rvert` to get correct spacing in general
@bp2017 oh it was your question:-)
@bp2017 but if you want alignment on the red line why have you put the alignment marker, & to the left of the |a| ??????, also never have \\ before a math display.
01:15
@DavidCarlisle, I can'tt thank you enough. I should've read into @egreg's answer more carefully, maybe that way I could pinpoint the reason. As for \\ and math display, I have recently ran across a question of interest on TeXExchange that had you commenting about this very issue. It would take me some time to memorize this. Thank you.
cfr
cfr
01:26
@DavidCarlisle yes, but I'm not in a shell.
02:11
I just don't get it...
\begin{document}
\noindent
\begin{alignat*}{1} % 1+0 = 1
&|a|=\sgn\left(a\right)\phantom{{}\cdot{}a}=
\end{alignat*}
\begin{alignat*}{1} % 1+0 = 1
&|a|=\sgn\left(a\right)\cdot{}a=
\end{alignat*}
\end{document}
02:31
After joining the two "alignat*" environments into one, the leading space is now identical, but \cdot operator has introduced new spacing issues:
@DavidCarlisle Thanks. There's a whole bunch of answers that reference texdoc.net. Darnit.
 
5 hours later…
07:40
@bp2017 don't use \left and \right there, don't put {}={} after \right), there is no reason for that, and if you want to align the equals signs use multiple alignment points instead of \phantom-trickery.
07:53
\documentclass{article}

\usepackage{mathtools}
\DeclarePairedDelimiter\abs{\lvert}{\rvert}
\DeclareMathOperator\sgn{sgn}

\begin{document}
\begin{alignat*}{2}
  \abs{a} &= \sgn(a)        &&={}\\
  \abs{a} &= \sgn(a)\cdot a &&={}
\end{alignat*}
\end{document}
@bp2017 ^^^
 
2 hours later…
10:07
@bp2017 The “spacing issues” are due to \left and \right, which aren't just useless, in this case, but harmful (they add unwanted space).
@cfr sadly, no. I am frustrated now and have given up :-(
@cfr it seems that hvmath and hvmaths are not installed in default in a full texlive distribution. i wonder why is that the case. these are free fonts, right?
@egreg so slow :P
@DavidCarlisle dvips complains that .mf files are not being found. it does not use the existing .pk files
@egreg sorry for teasing you. Good morning :) How are you today? Is the leg (I think it was) completely healed already?
does anyone know where \Umathsubshiftdown and other commands are documented? i want to pull up the subscripts and push up the superscripts in newcm. i think the original cm fonts had better vertical positioning than LM/NewCM.
10:34
@Skillmon Going well, but it'll take a couple of months for full recovery (hopefully)
10:54
@ApoorvPotnis well the fd, tfm, vf, style files are not a problem. But there are no sources for the free pk-files and the premade ones which are there seems to be only in 300dpi which is not really useful nowadays (the default mode ljfour uses 600 dpi). Also the naming of the pk-files is useless. In texlive the resolution is in the extension, so e.g. wrn10.600pk. So all files would have to be renamed. In miktex it would be easier, as it uses a folder dpi600.
@ApoorvPotnis then it's looking for a different resolution than the ones you have
Anyone seen/used the new 'subissue' feature in GitHub?
@ApoorvPotnis well the tex command is the luatex manual but that is a hook to the opentype math parameter you can perhaps start here learn.microsoft.com/en-us/typography/opentype/spec/math
@JosephWright I asssume someone has
@ApoorvPotnis you could use dvips -D 300 to tell it to target 300dpi (I think) but teh pk files would need tohave 300 in the name or be in a dpi300 directory see the standard cm pk files are in /usr/local/texlive/2024/texmf-dist/fonts/pk/ljfour/public/cm/dpi600/cmti10.pk at 600dpi
@ApoorvPotnis why do you want bitmap fonts anyway?
11:10
@JosephWright no. Where is it?
@bp2017 why are you not putting & at the alignment point? Why use an align environment then align by hand by inserting \pahantom everywhere?????
@DavidCarlisle really? interesting, in window they use extension. Do you know if both work?
@UlrikeFischer the mk--whatever scripts if making new ones use an extension but I noticed (just now) that the standard pk fonts are in a directory although of course the map file ensures they are never used anyway.
@Werner it wouldn't be the first time if we did a global edit on the site to update those
@DavidCarlisle you are right, I have on dpi600 folder in texmf-dist too, so probably both work.
@bp2017 you shouldn't have to remember not to use \\ there as you should almost never use \\ outside of an alignment so it should not occur to you to use it at all.
@ApoorvPotnis they are free as in no cost but not free as in open source (which texlive needs) in particular, as you have found, there are no metafont sources so the fonts can not be generated at the resolutions used by default. But apart from the technical issues as the ctan page shows ctan.org/pkg/hvmath-fonts they have a "No Commercial use" which is explicitly not a free licence and means the fonts can't be distributed with texlive (or linux distributions)
11:24
@DavidCarlisle :)
@UlrikeFischer For me, latest siunitx issue
@DavidCarlisle Maybe the type1 fonts costs a lot of drach^H^H^H^H^H€
@JosephWright github.com/josephwright/siunitx/issues/769 ? what is subissue there?
@DavidCarlisle Oh, so no £...
@JosephWright seen yes (coincidentally yesterday evening in the .multichoice: bugticket) but haven't used it, yet.
@DavidCarlisle ^^^ What I see
11:35
@JosephWright I don't see that.
@JosephWright ^^^ ah
@JosephWright as in this chat, you have extraordinary powers
speaking of new gh features, on a recent PR I got offfered a "try the new merge experience" alert, which I didn't dare try.
 
1 hour later…
13:00
@UlrikeFischer Yeah, I was wondering generally - I'd not seen it before
 
1 hour later…
14:25
@JosephWright I think I finally got a fix for the .multichoice: error (almost) ready.
The one case shown by muzimuzhi that uses a nested \keys_define:nn still breaks for me.
@Skillmon Cool - looking forward to it
@Skillmon I have an siunitx bug to fix, plus a tough one to think about, and xbemear is on my mind
@JosephWright it's just an incredibly long list of \@@_reset_var:N calls (turns out there were quite many status variables missing)
I wonder whether it's smart to keep all that on the input stack or whether at some point we should think of a stack-variable.
@Skillmon The input stack approach is taken from pgf: probably we could change and suffer no major consquences
@JosephWright well, we get problems if many input stack using macros are nested inside each other. I don't expect too many too deeply nested \keys_set:nn calls, so I guess we're fine with it. Let's see and when the truck hits us we can still wobble around it.
@Skillmon Ah, true: extremely unlikely to be an issue I think
14:55
The new siunitx issue turns out to be a great test for subissues:
@JosephWright a feature like that already existed on GitLab, iirc.
15:11
@Skillmon Sure, it's in various tracker - partly depends on whether you just want issues-for-users or workplanning-for-teams
@Skillmon I thought I'd try it - honestly makes no odds to me, but I can see for a wider project it might be useful even if it's the same person doing all the work
Perhaps for xbeamer (I want to look at that properly - next week)
@JosephWright agreed, as long as you don't have another software for task planning etc. (looking at Jira and friends)
@JosephWright prod me when you've reached that point, I'm still interested in that project.
@Skillmon Indeed - each 'vendor' wants people to stick with just their tool
(sorry, wrong link)
15:51
@JosephWright what about the test file naming suggested by muzimuzhi? Fine or should I fix it?
16:11
@Skillmon Ah, yes - we use m3<module><number> for expl3 tests generally, so he's right
@Skillmon, I can name one reason for using {}={}: in my case, it aligns the expressions perfectly, although by inflicting the cost of inserting extra space before the \cdot (which is better than having the expressions to be disaligned).
@Skillmon, Why is \phantom a trickery? Isn't \phantom intended to render characters invisible while retaining the space they occupy?
@egreg, the \left and \right are not useless to me in this case because I've always found operators such as cos, tan, and sgn to be more readable with parentheses around their arguments. Besides, what if a fraction were to replace "a" in "sgn(a)," would we still consider \left and \right useless?
@egreg, some advise to use \left and \right because they span the appropriate vertical space to accomodate the expression inside the grouping symbols, others warn that they're harmful (add unwanted space)... I don't know who is right. (Could it be the person who has implemented \left and \right in a way that breaks spacing, or is it an intrinsic part of how things work in LaTeX? In any case, I wonder if there's a workaround without laying aside \left and \right.)
@bp2017 but using phantom for aligning when you are in an algn environment is weird (and very inefficient)
@bp2017 the advice was not that you should not use parentheses it was that you should not use \left and \right.. (x) is not the same as \left(x\right) and the second is almost always wrong.
16:29
@UlrikeFischer i see
cfr
cfr
@ApoorvPotnis hvmath-fonts are not free. they are freeware for non-commercial use. (according to ctan.)
@DavidCarlisle just wanted to see how helvetica math looks. i thought that bitmap font might look good in print
in any case, i've given up :-(
cfr
cfr
@UlrikeFischer @ApoorvPotnis this is why my answer said to rename the subdirectories. then you don't have to rename every pk file.
i renamed the subdirectory to dpi300
the dpi300 directory contains all the .pk files. i also transferred the hvmaths.sty, and all the .vf, .tfm, and the .fd files the same directory, which contains test.tex and the dpi300 folder
also, is there a way to view the .pk files like we view .otf files in a font editor, like fontforge?
the .pk files, in principle, contain all the glyph info, right?
my working directory looks the same as in your answer @cfr
@ApoorvPotnis tex related programs are about the only ones remaining that can read that format, so once it's working you can make a tex file to show them eg type pdflatex fontsmpl answer the questions
16:39
i see. so basically the .pk format never got popular out of the tex world
@bp2017 You have \left(a\right), not a fraction in between the parentheses. Here \left and \right are not only useless (they choose the same size), but also harmful, because they add unwanted space. I never said that \left and \right are always useless, did I?
does anyone know if there was a particular reason the subscript and superscript shift values were changed for latin modern, as compared to computer modern? i suppose the bluesky type1 versions have the same metrics as the original bitmap ones?
@ApoorvPotnis that is over-estimating the non-tex use
cfr
cfr
@ApoorvPotnis and they are no longer popular in the tex world either because they do not scale. you can see the fuzziness even in my answer.
indeed. tbh, i think the prelualatex way of using fonts is a real headache. how is a normal user supposed to make sense of all the .mf, .vf, .tfm, .pk files?
16:45
@ApoorvPotnis they don't: millions of users use pdftex for decades without knowing anything about those extensions.
cfr
cfr
@ApoorvPotnis well, postscript type1 fonts were used long before luatex and widely used beyond the tex world. I've never used .mf or pk fonts in a real document.
i see. my main exposure has been to the .otf fonts, which are easy to use in the supported engines
cfr
cfr
@ApoorvPotnis it depends on your perspective. yes, for an arbitrary font, it is easier with fontspec. but setting up fonts with fontspec is often more involved than \usepackage{<something>}, which is all most users ever deal with for pdftex.
@ApoorvPotnis but people have far more font issues with portability, so easy to pick up some random font installed on your system that doesn't work anywhere else.So much easier when everyone used cmr10
hmm. it would have been probably better then if cmr10 was a slightly thicker font, with a complete coverage of glyphs, so that users don't ever need to load different packages for different gylphs
cfr
cfr
16:51
@ApoorvPotnis users regularly post questions with code I can't compile because I don't have, say, Arial or Times. or I might have the ttf rather than the otf of Droid or whatever.
@ApoorvPotnis it was more than thick enough when printed with wet ink dot matrix printer on soft paper The type1 versions ended up being a bit thin when seen on screen....
ah, commercial fonts could be a problematic issue
cfr
cfr
@ApoorvPotnis the font encodings Knutt worked with had only 128 slots. even postscript type1 have only 256.
hmm. memory problems of the past, i suppose
cfr
cfr
16:52
Knutt -> Knuth
@ApoorvPotnis good quality type1 fonts included things like small-caps, but they were in a separate font because there were not enough slots for those as well as lower case. pdftex cannot use fonts with more than 256 slots directly.
@cfr it's probably then time to switch to modern engines
@ApoorvPotnis Yes
@ApoorvPotnis LuaTeX
@ApoorvPotnis that is not always possible, arxiv and many journals require use of an 8 bit tex for example, that will change but pdflatex is by far the most used format at the present time.
@DavidCarlisle Well yes but ..
Hans has a point
@JosephWright ... but life isn't always as you'd wish?
16:59
@DavidCarlisle :)
in reality though, i for some reason see a lot of inertia when it comes to typesetting. Springer and almost all the math/physics books i've read, even the modern ones, still use pdflatex for some reason. surely, the big publishers have lots of money to update their systems
@ApoorvPotnis For print in English (and indeed most western European languages) pdfTeX is OK - but there are real issues beyond that
i hope arxiv supports lualatex when i start publishing papers :-)
@ApoorvPotnis What area do you work in?
For me, arXiv isn't an issue :0
@ApoorvPotnis they probably don't have lots of money and don't want to spend it, and if printing is your main buisness, changing the core typesetting engine is a high risk activity.
@ApoorvPotnis they are working on it....
17:02
@DavidCarlisle I think it's reasonable to say Springer are not so badly off :)
@JosephWright i am a math student (completed my masters). i wish to become a physicist. not a researcher yet though
@JosephWright They have money but how much they make on journal printing (and how much they want to spend on changing anythng) are not so clear
@DavidCarlisle Indeed they are :)
@DavidCarlisle Yes, that's fair
cfr
cfr
@ApoorvPotnis I've essentially never shared an actual .tex file. either it is .pdf or converted to word.
> In 2013, the London-based private equity firm BC Partners acquired a majority stake in Springer from EQT and GIC for $4.4 billion.[11]

for springer books. not sure about the journals
17:06
@ApoorvPotnis well if they gave half of that to @JosephWright and me, we'd help them switch to lualatex.
i shall try to persuade springer executives then...hehe
@DavidCarlisle, why use an align environment and still resort to \phantom: an align environment is the only means (I am aware of) that allows for the conditional curly brackets, and expressions in my examples would rely on curcly brackets (they were not shown for the simplicity's sake).
@cfr i think authors share a .tex file with both the journal and the book publishers.
@DavidCarlisle, why not use & at every alignment point: (1) there are so many alignment points that I've deduced (possibly, incorrectly) using \phantom instead would reduce complexity, (2) initially, I had two separate "alignat*" environments which had to align with one another, and \phantom was the only solution available (to MY knowledge) for correcting the misalignment between the two.
@ApoorvPotnis Nah, usually PDF (in my area at least)
17:10
@JosephWright oh i see. but then how do the editors make changes? by editors, i don't the journal editors, i mean the typesetting editors (both for books and journals)
@bp2017 apart from looking weird (and hard not to add spacing artifacts as you are seeing), it's somehat inefficient, the content of every phantom is set 8 times (4 for being a phantom and twice for being in an alignment) not that it takes very long to typeset {}={} but I feel TeX's pain when i see the markup done that way.
@ApoorvPotnis They use change tracking in Word (or APP(3B2) or InDesign...)
@ApoorvPotnis In my area: convert to Word, import into their typesetting system (XML-based), edit there, send back Word files for author markup
huh? why this convoluted route though? why not edit the source files themselves?
@ApoorvPotnis often even journals that accept latex source do not use tex for publishing, they just use the tex to get a manuscript then re-key it into a commercial publishing system. Some do that even if they do use tex.
By the way, what could be the reason for the misalignment between the two environments? Weren't they both supposed to begin at the same starting point?

\begin{document}
\noindent
\begin{alignat*}{1} % 1+0 = 1
&|a|=\sgn\left(a\right)\phantom{{}\cdot{}a}=
\end{alignat*}
\begin{alignat*}{1} % 1+0 = 1
&|a|=\sgn\left(a\right)\cdot{}a=
\end{alignat*}
\end{document}
@DavidCarlisle, if I should almost never use \\ outside of an alignment, how am I going to add a blank line?
17:17
@ApoorvPotnis In my area: convert to Word, import into their typesetting system (XML-based), edit there, send back Word files for author markup
@DavidCarlisle A shorter version of what I said :)
@bp2017 you haven't shown code to reproduce but the \noindent is wrong (and doesn't affect indentation) and (by default at least) the two environments are centered so not aligned at all
@ApoorvPotnis I am on contract with the ACS so I know something about the internal process there :)
@bp2017 You can never get a blank line from \\ so that doesn't arise (if you try you presumanly get a warning about badness 10000 which is tex's maximum measure of bad quality output.
@DavidCarlisle i see. probably that's why the math and physics books (atleast the ones i've read) have typesetting deficiencies
@JosephWright i see. i only know of one Indian company, River Valley technologies, which apparently uses a tex based workflow (though they use other software as well)
@ApoorvPotnis but even when a jounal uses tex. they ask authors for a clean tex doument but authors fil the manuscript with custom macros and spacing and other horrors, so when the journal tries to use the document class setting up the journal house style it fails. At that point they have a choice: get a tex specialist to debug the author macros and make the manuscript work, or print it on paper and sent to a country with companies that specialise typing from manuscript at a fixed price per page
@ApoorvPotnis . guess which is cheaper and has more reliable time estimates
@bp2017 you can also see the bad spacing from \left in your image.
17:27
@DavidCarlisle hmm. Do they do the same for books as well? I've seen draft versions of books of the websites of authors, and the published book also contains the same typesetting mistakes. For example,subscripts which are words are in math italics, instead of upright style
@ApoorvPotnis it depends on stuff I don't know. but certainly not every author supplied tex file gets published the way the author expects:-)
Probably the authors don't bother with trivial typesetting stuff then
@ApoorvPotnis yes well look on arxiv where it is author supplied tex and I'd say that kind of comment applies to most of them. typesetting is not a primary interest of most of the authors
@ApoorvPotnis I have books set in TeX, but few journal articles - and most of those are at the physics end of the spectrum from smaller publishers
Sad (for me) although they're probably justified in devoting more time for science, rather than typesetting
17:32
@DavidCarlisle, \noindent works for text that precedes an align environment and starts the first paragraph/line. It doesn't seem to have any effect with "standalone" class. Since I've switched to "standalone" class, \noindent remained as an artifact. We can, of course, ignore it. It is my fault for including it in the code, I should have paid more attention.
@bp2017 in theh bottom one you have \mathinner followed by a mathbin in the first you have a mathinner followed by a mathord(which contains a mathord a mathbin and amathord). Not using \left \right` so getting rigd of teh \mathinner would help, but as I say, using \phantom and getting thespacing right is hard. Much better to use an alignment environment.
@DavidCarlisle, but \\ does start a new line (or pargraph?). (LuaLaTeX, either document class: article or standalone.) I don't get any warning or error.
@bp2017 no I meant it doesn't affect the alignment of \begin{alignat} (in any document class.
@bp2017 No it ends a line. It does not end a paragraph,
@DavidCarlisle, I'll considering using & instead, as suggested.
*consider
@bp2017 take the code you posted above and replace \noindent by aaa\\ The terminal will show
Underfull \hbox (badness 10000) in paragraph at lines 5--6
@bp2017 `badness 10000 is "as bad as it gets"
17:43
@DavidCarlisle, I get "badness 10000" for \\ between the two align environments, but not for \noindent:
\noindent
test\\
here
\begin{alignat*}{1}
&|a|=\sgn\left(a\right)\phantom{{}\cdot{}a}=
\end{alignat*}\\
\begin{alignat*}{1}
&|a|=\sgn\left(a\right)\cdot{}a=
\end{alignat*}
@bp2017 didn't I send you code this morning with correct alignment? Why do you ask the same yet again?
@bp2017 that's what I said. \noindent is just wrong for other reasons (there is no paragraph that could be indented so nothing to not indent, it does not affect the indentation of the following math display.
@bp2017 never use \\ after a display environment and never use two consecutive display environments with no text in between
@Skillmon, thank you. But I wasn't asking about how to solve the issue again, since your code had already done so. I was referring to particular things you've "said" about \left and \right.
@Skillmon surely no one was expecting correct code from such a source?
@bp2017 why would you write \left(a\right) instead of (a) ? it is more to type and messes up the spacing.
@bp2017 my suggestion was to make a bad example by replacing \noindent before the display by aaa\\ that would make the warning. You have test\\ here so \\ mid-paragraph which is completely different and makes no warning
@bp2017 but you don't need them to get alignment here, they just insert extra space.
@bp2017 you're using alignat that provides means to align stuff at specified points, so why do you resort to \phantom? That is just a trick at that spot and nothing more. Did you try my code?
17:53
@DavidCarlisle, I wasn't sure (a) would in all cases scale the parentheses properly even for single-line expressions like "a," so just in case I was using \left(a\right; another reason is that it's easier to flip back and forth between expressions like "a" and "a/b" by pasting them in between \left( and \right).
@bp2017 but it's just wrong, the ( won't stretch and the horizontal spacing is disturbed. Getting bad space is somthing you might tolerate around a large matrix (but see mleftright package to avoid that, but getting it when the brackets don't even stretch makes no sense. do not use \left\right "just in case"
@Skillmon, I have tried your code, and it resolves the issue with flying colors. I just can't understand whether \phantom should never be used inside an align environment (because we can do with & instead) or there are cases where \phantom cannot be replaced with &.
@DavidCarlisle, thanks.
cfr
cfr
@ApoorvPotnis in my area, they almost always demand word or equivalent. almost nobody accepts pdf for an accepted article (as opposed to an initial submission). the exception is maybe one or two journals which are not published by major publishers and they want camera-ready copy. or journals which publish papers in technical areas (but often they only accept latex for technical subjects).
18:11
Is there any drawback in placing a \prop_if_exist:NF with \msg_error:nnn in an expandable command? I think not ... or yes?
@bp2017 you can use \phantom there is nothing inherently wrong with it, but sporadically as a last way out to achieve something. In alignat it's probably never necessary.
@JasperHabicht you can do that, but should use \msg_expandable_error:nnn instead, or else your macro isn't expandable
@DavidCarlisle, my takeaway from the conversation: (1) never use \\ after a display environment and never use two consecutive display environments with no text in between; (2) do not use \left and \right to enclose expressions that are (my understanding) one character in height; (3) by default, two (independent) environments (that are next to each other, or otherwise) are not centered; (4) \noindent before math display is redundant.
@Skillmon I saw that ... I just wondered, because in my (probably quite naive) tests, the command at least worked as expected
But ... wait ... I think, the error message just did not show up in this case ... let me check with \msg_expandable_error.
@cfr i see. does your area not involve heavy math equations? i guess it would very hard to type equations in word
@Skillmon, I'll use & instead of \phantom from now on. Although I have one quetion (if I may): what if we had "\DeclarePairedDelimiter\prnthss{\left(}{\right)}" and used \prnthss{a}, would it be any different than using \left(a\right)?
18:27
@bp2017 (3) is wrong. (4) is better said as "don't use \noindent in a latex document
@ApoorvPotnis are you sure about that? You are encouraging people to use lualatex rather than classic tex, the math OpenMath layout in lualatex is modelled on that in Word.
@bp2017 isn't that a syntax error? what happens if you use \left and \right` like that
@Skillmon I think I can live without adding the message in this particular case. I think the error message that will pop up per default will be clear enough. Thanks anyways!
@bp2017 that would be very wrong. Read the documentation of mathtools regarding \DeclarePairedDelimiter and you'll understand why.
@bp2017 assuming you meant \DeclarePairedDelimiter\prnthss{(}{)} then when it does use \left\right eg \prnthss*{x} then it compensates for most of the bad space effects
@DavidCarlisle There is also the mleftright package, but maybe that has bad consequences?
Lots of fun snooker these days x.com/WeAreWST/status/1880027911283175645.
cfr
cfr
18:59
@ApoorvPotnis it does not, no. some journals, as I say, make an exception for papers in technical areas, especially logic. (I'd assume maths, too, though I haven't checked.) I'm not sure what other journals do for papers which involve maths historically. these days, I'd assume word is fine for equations. I'd imagine things like proofs in logic are more problematic, but I haven't tried. (I don't have word, but I haven't tried in libreoffice, either.)
@DavidCarlisle I meant in the sense of actually typing the equation. I had tried to use word many years ago, and I had to click on various menus and symbols to insert stuff, using mouse. I guess I don't have a deeper knowledge of word to use it efficiently. Also, I couldn't figure out correct placement of diagrams on the page, but this is probably just my incompetence.
cfr
cfr
@ApoorvPotnis ethics rarely requires equations, though it does sometimes.
@cfr I tried livreofficea few days earlier. I couldn't see any option to use opentype math. Libreoffice has a dedicated math equation editor, but it wasn't really thatt good
@ApoorvPotnis i can't use Word at all (it isn't emacs) but people do.
@DavidCarlisle I see. Hope your left pinky is okay.
19:03
@mickep I think I mentioned mleftright somewhere as well yes it's similar to declaredpaired delimiter
@ApoorvPotnis see my in depth review of all available editors here: tex.stackexchange.com/questions/85249/…
cfr
cfr
@ApoorvPotnis I only use it when forced. mostly for form filling. even then, I sometimes convert forms to .tex. I hate using it and I have never even opened the equation editor. I'm not sure I knew it had one.
Detailed indeed
@cfr yeah, the equation editor isn't that popular
@ApoorvPotnis You can use it with the keyboard, although it's painful
@JosephWright I'll give it a try once again.
19:36
@DavidCarlisle Ah, I did not follow so closely.
@mickep wise choice
@ApoorvPotnis hit the Alt key once, that should spawn some hints on the lower left corner of each button, you can then enter the characters displayed there to "press" the buttons. The equation editor supports LaTeX syntax at least partly. It's still a pain using Word, but that eases the pain a little bit.
20:21
Never mind, just checked the actions
@JosephWright is there no [fixed-in-dev] tag in the L3 repo? What should happen with github.com/latex3/latex3/issues/1667, closing?
Hi everyone, it's my birthday today! Is it okay for me to ask for a drawing without including an MWE?
20:41
@EmanueleNardi I guess one can always ask, but I don't think you can expect anyone to say they will draw it before knowing what it is about. Congratulations by the way!
 
1 hour later…
22:09
@Skillmon I’ve closed
cfr
cfr
@EmanueleNardi if it is a duck or a rabbit or a duck-rabbit, probably.
@DavidCarlisle Does it mean (a) is also wrong and should be \lparen{}a\rparen? By the way, what was wrong with the symbols themselves for LaTeX and mathtool authors to make the corresponding word versions?
22:25
@bp2017 no lparen isn't a standard command, it's defined if you load mathtools but is the same as ( but you should not follow it by {} as that would affect spacing in general
@bp2017 | doesn't have any spacing assigned but \lvert is mathopen like ( and \rvert is mathclose like ) so you get correct spacing.
@DavidCarlisle, so there's a difference between a simple space (key on the keyboard) and {}? I remember reading somewhere {} being the same as space (character).
@bp2017 compare |-2| and \lvert -2\rvert and you will see why | is wrong
@bp2017 why ask that? you have already shown multiple times that : {}={} is not the same as : =
@DavidCarlisle, I find it odd for mathtools documentation mentioning \lparen and \rparen in the context big parentheses and in the context of \left( and \right). Because when I use \lparen and \rparen, they don't scale for fractions.
@bp2017 they don't stretch but the referenced package uses the name to make modified names
@DavidCarlisle, but how are you going to write \lparen a\rparen without a space before "a"? I must have misunderstood you, I don't know.
22:34
@bp2017 just write what you wrote, that does not make any space
@DavidCarlisle, so {} and " " are different kinds of spaces?
I see, " " and \" " are the same. {} is a delimeter.
@JosephWright in other news: I did a fast regex-based replacement of all the ; in l3fp and got only two failing m3fp... tests (m3fp001 and m3fp-assign001)
@bp2017 well not really.
@bp2017 no
Oh, and m3show003 failed for some reason.
So I only broke 3 tests, now to investigate why
@bp2017 empheq docuemnts why you need names it gives the example \DeclareLeftDelimiter{\llbracket} defining \empeqllbracket you can do that with \lparen to define \empheqlparen you could not do it with ( as you can not have a command called \empheq( it seems a fairly useless feature (and I have never used \lparen or \DeclareLeftDelimiter but it is documented
22:41
@DavidCarlisle, my understanding of TeX was not great to begin with, but I think my memory has become rusty after not using the system for a few years. I think I should go and read a book or two on TeX to refresh it (and learn something new) instead of barraging good folks with questions of trivial sort :)
@DavidCarlisle, I have one last question. It's regarding LuaLaTeX. Is Lua version of LaTeX a complete rewrite that uses Lua, or does it simply use Lua to access LaTeX commands that have already been written separately?
@bp2017 mostly the latter. A few commands have lua versions but mostly only where luatex has dropped a pdftex command so we have to provide a Lua implemented aternative.
@bp2017 although for many people access to Lua isn't the main feature, it is access to Unicode opentype fonts
Ok, all are no-brainers, the verbatim contents of a variable were dumped into the test log which of course changed for a changed delimiter.
@DavidCarlisle, thanks.
23:38
@Skillmon a large edit did you use M-x tags-query-replace ?
23:51
@DavidCarlisle :bufdo %s/.../.../g

« first day (5192 days earlier)      last day (26 days later) »