@egreg -- certainly the ams classes load amsmath by default, but they have the ability to pass options to amsmath, so that's a legitimate possibility. i'll query the locals.

@barbarabeeton Help! They will fail me in my examinations!

@PauloCereda -- they're just trying to get you to pay more attention to your thesis. they're tough customers, but they're not vindictive. ... so, how is your thesis?

@barbarabeeton it's evolving. :)

@DavidCarlisle I was on jury duty for a thesis last year, and when I got the printed copy I opened at random and the very first thing I say was a misaligned multlined :-)

@PauloCereda well in some ways that's what \usepackage{mathtools} is (so we could just get @daleif to fix it there:-), but the experience of fixltx2e wasn't great, we had to maintain two code paths for years, and 99% of users used the code path without any of the fixes.

@DavidCarlisle well, there should be something like compatibility=2015 or compatibility={aligned=2015} or similar probably

@yo' but as egreg said, not always easy to expose that to users using somerandomthesis-or-journal.cls that loads amsmath internally.

@cfr we could blame @egreg instead.

@DavidCarlisle Well, the interface exists (our good ol' PassOptionsToPackage), the question is are people going to get it...

(I mean, as your fix is bullet proof, they shan't need to, but still)

@yo' yes but if the point of the option is to enable old code to work as before, it shouldn't take too many code changes to enable it. I don't know I'll not rush to add anything. (although I'd like to update amsmath soon so that generalised fractions work in xetex (@barbarabeeton:-))

@DavidCarlisle well, currently your code change is basically \appto\aligned\zz\appto\multlined\zz\appto\gathered\zz

@yo' yes (although a real fix might be to do the test but then save the result and not add the \, later rather than add a correcting \! at that point, ends up looking the same but a bit more efficient in terms of output.

@DavidCarlisle well, we don't "fix" that much of amsmath it is probably more of an addendum. I think empheq does more fixing (Morten had his hands quite dirty in that one). At least empheq provides a correction to ntheorem such that a tombstone can be placed at the bottom of the display instead at the baseline.

Does someone have an idea how to force the line marker on the legend line?

@UlrikeFischer :-)

@JosephWright I invented a new chemical compound:-)

@DavidCarlisle ooh

@DavidCarlisle ?

@JosephWright An undocumented feature of some reaction.

Happy 50th anniversary, Star Trek!

@PauloCereda :) Qapla'

@JosephWright .. tex.stackexchange.com/questions/328735/…

@PauloCereda ^

@PauloCereda Long live and prosper!

@DavidCarlisle “Spuck”?

@egreg according to the website it goes by the more catchy name: CelebriDucks-MR-SPOCK-Rubber-Duck-STAR-TREK-Bath-Toy-Ducky-Figure

@DavidCarlisle Definitely better, I agree

Friends: suggestions for someone to invite to UK-TUG, 15th October, Oxford, UK?

I was wondering about @topskip

Is @Kurt sometimes around in chat? He asked how far my fillPages.sty got. Just wanted to know if he needs it.

@egreg ooooh

@JosephWright ooh a good name!

@JosephWright Hans?

@DavidCarlisle Also one I thought of

@JosephWright I suppose it's chicken and egg if you can get someone that you can advertise early enough you might actually get an audience, otherwise it's a bit embarrassing to fly someone in to speak to half a dozen people.

@DavidCarlisle plus a duck.

@DavidCarlisle That's true, but I think it will happen anyway :-)

@PauloCereda You're on Skype, that's different!

@PauloCereda ducks can fly themselves in

@JosephWright awww <3

@DavidCarlisle :)

@DavidCarlisle oh no

@DavidCarlisle We'd need a tag for that though. (I removed the latex3 tag from the question and replaced it with whatever I replaced it with. I don't recall @egreg being an option.)

@cfr It's never my fault

@cfr tex.stackexchange.com/questions/tagged/gregorio

many of you may know about this already, but in case you don't, and in case you might be able to attend, there will be a celebration of sebastian rahtz and his work at wolfson college, oxford university, on september 27. space is limited, so registration is required. click here for information on how to register.

@barbarabeeton Announced to UK-TUG

@DavidCarlisle I see you are in a good mood!

@JosephWright just trying to be clear:-)

@DavidCarlisle Of course all upvotes in that tag go to me as well.

:)

@JosephWright ooh

@JosephWright clearly not our Bruno, team members never that polite to each other

@DavidCarlisle :)

@DavidCarlisle I thought that willingness to be on a team with someone itself is a sufficient sign of politeness :-)

@yo' You don't see the internal mails ;-)

@JosephWright I can poke German people on Threema!

@JosephWright I don't need no seeing everything :-)

@PauloCereda :)

@yo' :)

@JosephWright -- just a couple of hours ago, i was notified that the url published for textures in my column in tugboat 36:1 (last year) has been infected with ransomware. there's at least one question on the main tex.sx site about "what happened to textures", and at least one, tex.stackexchange.com/a/109051, was answered by me with a link specified. would it be appropriate to add to that answer an update saying do not go there and why?

@ChristianHupfer Dude, why did you delete your answer? It was a legit answer.

@JosephWright I'll see if I can work up a MWE

I found a downside of using emacs on windows: getting C-W mixed up as it closes wthings in windows, and cntrl-V does annoying things in emacs

@DavidCarlisle Your expl3 skills need improvement

\usepackage{expl3}
\ExplSyntaxOn
\cs_set:cpn { Gscale@div } #1 #2 #3
 {
  \cs_set:Npx #1
   {
    \fp_to_decimal:n
     {
      \dim_to_decimal_in_sp:n { #2 } / \dim_to_decimal_in_sp:n { #3 }
     }
   }
 }
\ExplSyntaxOff

@egreg like my Italian, my expl3 is faultless

@egreg I just compared and your code doesn't look any different from my answer.

@DavidCarlisle But Frank will be happier

@egreg or you will be:-)

@egreg (@JosephWright) see the xetex comment on that answer?

@DavidCarlisle Guess I'll have to take a look

@DavidCarlisle Interesting: the lines are also repeated at normal size

@egreg yes, seems that way (arithmetic overflow somewhere?) need to make an example that's doing less stuff...

@egreg @JosephWright if you separate xdv and pdf steps then xdvipdfmx complains a lot:

xdvipdfmx:warning: --- M = [5.30937e-06 0 0 5.30937e-06 181.336 -177.793]
xdvipdfmx:warning: Transformation matrix not invertible.

@DavidCarlisle uh-oh

I suspect the numerical analysts at work might have a view on the problem that's being posed here:-)

@DavidCarlisle oh no, math people

@PauloCereda what would you rather have, chemists?

@PauloCereda Actually, numerologists numerical mathematicians

@DavidCarlisle What have the chemists ever done for us? :)

@yo' oooh

@PauloCereda I have a list around here somewhere

and the notion you're (not?) looking for is ill-conditioned problem

@PauloCereda I might flag that for moderator attention

@DavidCarlisle Try with dvipdfmx?

@PauloCereda Oh, right. Made the plastics from which rubber ducks are made

@DavidCarlisle Don't you dare. :)

@Canageek ooh I take that back. <3

@JosephWright same

@DavidCarlisle OK, so could be the TeX code or the driver itself

@JosephWright What has the driver ever done for us?

@JosephWright or just that it's being asked to scale to unreasonable levels...

@JosephWright well, if you check, the matrix is not singular, it's simply a shrinking by 10^6 and translating --- far from being singular

ooh a plural matrix

I think that the driver is simply not capable of shrinking that much.

What have plural matrices ever done for us?

I had a similar problem recently when an image had dpi set to 1 (one). Caused pdftex to go insane.

@PauloCereda Matrices for plurality! "We want to have opinions, too!" they shout!

@yo' LOL

@yo' 10^-6 not 10^6 :-)

@DavidCarlisle s/shrinking by/shrinking by a factor of/

same with latex/dvipdfm

@DavidCarlisle Hmm, still could be either end

How should I cite tex.se answers that provide a code snippet or technique I use in an assignment?

@hexafraction Under the answer there's a “share” button; right-click and select “copy link address”, so you get a URL pointing to the answer.

@egreg Thanks!

@hexafraction You're welcome! Provided you cite at least one of my answers. ;-)

Funny enough, I'm actually referring to one of your answers :)

(the one about \systeme for systems of equations including modification of the package code)

@barbarabeeton thanks:-)

@DavidCarlisle -- hope it gets the desired result. no promises, but i tried.

@JosephWright -- any ideas about my earlier ping, about adding to an existing (old) answer?

@barbarabeeton incidentally did you try the updated version of genfrac that gives consistent sized brackets with your example see the code above the line at tex.stackexchange.com/a/328222/1090

Dammit, I found another bug in @Cle­mensNieder­berger 's packages

@clemens Found a bug in your packages: \dento{4} should give a superscript 4, and it isn't. Do you need a MWE?

@clemens And if so, where should I send it?

@barbarabeeton Not sure what can be said bare edit the answer

@Canageek Look at mychemistry.eu/about (or in clemens's package documentations) for an email address

@JosephWright -- thanks. have added a warning to my previous answer.

@egreg Thanks

@JosephWright So, I'm saying several prior papers were wrong in my first primary author paper...including someone who might be a reviwer on the paper. (eep). Minor error for most (forgot to account for obvious disorder) but this one person wrote a whole paper based on this bond angle....

@Canageek Oooh

@DavidCarlisle -- have now tried it with both lm (as you had it) and the stix fonts. with stix fonts, the first \binom in each of the triples after the sum has smaller parens than the other two. the parens in the third line, with k at the top, are also skimpier than with lm. i'll do more tests over the weekend and report by mail.

@barbarabeeton With TWELVE uppercase letters!

@egreg -- with twelve BOLD uppercase letters

@barbarabeeton Oh my!

@PauloCereda -- yes, the situation is quite dire. never mind distressing.

@barbarabeeton well, I'd even suggest removing the link completely for the time being

@yo' -- do html comments work in answers?

@barbarabeeton not sure, but there's revision history. You can always put something like --link temporarily removed as the website is currently compromised, see the revision history for the link-- in place of the link

@yo' -- okay, thanks. (i keep forgetting i can see the revision history.)

@barbarabeeton well, even non-users can, AFAIK

@yo' -- done. thanks.

@barbarabeeton you're welcome. Now I'm sorry, I wanna finish the sheet music I'm working on

btw, I just realized the annotation in the sheets is done in three different languages. This is screwed. @egreg @Paulo

"tempo di blues, medium swing" at one place, and "pravou ruku lze hrát v některých slokách o oktávu výš" at another place

@barbarabeeton will look with stix, although of course since any font just has an essentially arbitrary list of sized brackets to start with there are always discontinuities where it jumps to the next one. In the end the question is not whether the \left\right version is identical to the withdelims one, so much as is it, on its own, a reasonable implementation of \binom and friends.

@JosephWright @barbarabeeton I think the comments to tex.stackexchange.com/a/109051 should be purged; they are completely irrelevant at this moment

@yo' done

@JosephWright thanks

@DavidCarlisle -- i think what i see in your test, with stix fonts, is not a show-stopper, since it's consistent. consistency, and appropriateness, are what i'm looking for. more later. right now, i have an appointment to get shorn.

@JosephWright I've gotten my paper down to 119 citations

@Canageek that's more than what my thesis has!

So, just saw an answer that the OP had awarded a bounty, but neither upvoted nor accepted ...

@TorbjørnT. wasn't it an auto-award? got a link, please?

@yo' Nope, it had zero votes. tex.stackexchange.com/questions/326797/…

@TorbjørnT. oh rly. say, generous people :-)

@yo' That is more then my undergrad thesis had >.>

@Canageek First thesis has 4. Second thesis has 13 + 7*wiki + 1*web.

@yo' Those don't sound like thesis at all.

@Canageek degree project is better

@yo' Most chemistry thesis I've seen run 200-300 pages for an M.Sc and Ph.D. I've heard tell of 700 page ones.

@Canageek no way!

@yo' Of course, I heard that McMaster chem brought in a 300 page limit for a Ph.D. thesis after that. Not sure if that story is true or not though.

@JosephWright How long was your thesis?

my PhD thesis has 86 pages, the text itself is 64 pages

@Canageek Just short of 150 pages

@JosephWright and what percentage of figures and tables? :-)

@Canageek My master thesis was around 70 pages ...

@TorbjørnT. What discipline?

@yo' Obviously a reasonable number of graphics: I am a synthetic chemist after all

@Canageek Physical oceanography.

@TorbjørnT. That is about the length of my girlfriend's Lingustics thesis

@JosephWright yeah :) for me, the figures and table occupy I guess at most 10 pages in total (20 figures and 4 tables, and the tables aren't huge, nor are the figures actually)

@Canageek So it's just chemists that are crazy.

@TorbjørnT. and philosophers

@JosephWright Given what I know about synth. org. chem. you spent 1 page on each step of some horrible natural product with a name like 2,2-dimethylkillmenowol right? ;)

@TorbjørnT. I believe 400+ pages is common in some areas

@Canageek No, not total synthesis :-) I'd say 'organometallic methodology/mechanism'

@JosephWright For M.Sc? Or PhD?

@TorbjørnT. PhD: some parts of earth science, I'm told

@JosephWright McMaster Chemistry has a page limit of 300 of Ph.D. and 200 for M.Sc.

@JosephWright I could ask my dad; he's an earth scientist :)

@Canageek lower or upper?

@Canageek That would be tight for bio: 250+ is very common

@yo' Upper.

@Canageek ah ok

@JosephWright Yeah, it is for chemistry as well.

SFU does not have one, luckily.

I wonder if anybody actually reads those 300 pages lol

@Canageek Going that way, yes: I'm in favour of shorter theses, but of course this depends on the examiners too

I believe that only 2 people actually read my thesis, and that's the two opponents (maybe), and remember it's short with 64 pages

@JosephWright I mean, I'll be trying to fit 5-6+ years of work into mine (Combined M.Sc./Ph.D.)

@Canageek Yes, but that should be possible: it's not everything you did, it's a 'story' of some part of it

@yo' a minimum of 6 will read mine (My prof, 2 comittee members, external, internal)..

@Canageek I didn't include about a third of my research into my thesis, because it didn't quite fit by topic

@Canageek ARE YOU SURE? :D

@JosephWright Ah, we usually do one chapter on each project/result and have a story inside of that

@yo' Yes: Each will hand back a LARGE list of things you need to fix

@Canageek They might :-)

@JosephWright Everyone who has graduated from my group has gotten one.

@Canageek France where I defended has this system as well, but luckily none of the 2 reporters suggested any large changes; I basically only added one reference and corrected some typos

anyway, bedtime, I get up quite early tomorrow. Good night!