@UlrikeFischer yes

@DavidCarlisle It was 2.13 with TeX Live 2019 before this. Now it's 2.14.

That's based on the Debian packages, at any rate.

@HaraldHanche-Olsen Well, ok. But are expansion and execution distinct processes or not? And if so, why does even Knuth use different terminology for them?

@DavidCarlisle It's perhaps more to the point that biblatex appears to require a bleeding edge biber release. If it didn't, it wouldn't be so much of an issue.

The "Table 1: biber/biblatex compatibility matrix" table in the biblatex manual doesn't show the current biblatex version, namely 3.14.

Pg 7 of the manual.

@UlrikeFischer I'm working on it: trying out what is the best approach as there are clearly a few things here I've overlooked in the past

@UlrikeFischer First, I'm trying out in v3 some 'clean' code, then will transfer it across

@FaheemMitha seems that way

@DavidCarlisle then couldn't you do it under the assumption that one can always "fake" a bold version with e.g. \setmathfont{Latin Modern Math}[version=bold,FakeBold=2.5]?

@UlrikeFischer possibly. I haven't looked in a while. Also while the basic unicode font -> the unicode font that is used in the same slot in the bold math variant is probably easy enough (bm aready knows how to decode \Umathchar information), interacting with the various range=... mapping that unicode-math is doing on top may be tricky.

@UlrikeFischer @JosephWright speaking of bold did you see this, it came up just as I was setting off to drive here so haven;t looked at it at all yet but warnings about bx sound familiar, don't know if the PL3 patch helps at all tex.stackexchange.com/questions/528909/…

@DavidCarlisle yes, mathspec uses bx. It probably will be resolved when @WillRobertson sets up fonts as b + bx.

@Annie well occasionally. The question is only a day old, so no need to flag it here really.

Thank you @DavidCarlisle

ooh smooth criminal song

There seems to be something very wrong with the title to this question

@DavidCarlisle ooh

@DavidCarlisle the really astonishing thing is that APL was invented before unicode...

@barbarabeeton a cute egg holder: amazon.com/o/ASIN/B07WD8MF7W/janu20-1-20/ref=nosim

@DavidCarlisle a clear proof that a pizza is a pineapple with smaller \baselineskip ;-)

@DavidCarlisle What that question needs is more TikZ.

quack

@PauloCereda dinner

@DavidCarlisle oh no

@DavidCarlisle That's not a bad idea, having dinner.

Nom nom nom

@JosephWright you remind me I have a branch somewhere with hyperref with a lot fewer dependencies, I really should finish that off :-)

@DavidCarlisle :)

@DavidCarlisle do we really want a tikz logo when there is none for picture mode??

@UlrikeFischer picture mode doesn't need a flashy logo, it stands on its own merits

@DavidCarlisle The plane ;)

@JosephWright have you yet retouched the l3keys implementation of \keys_parse:NNn?

@Skillmonlikestopanswers.xyz No

@Skillmonlikestopanswers.xyz On other stuff ...

@JosephWright yeah, I saw the activity on the case changing thread on Github... Just wasn't sure whether I missed something.

@Skillmonlikestopanswers.xyz Also siunitx

@JosephWright that one I don't follow as strictly. I just know you released some v3 beta a couple of months ago.

@Skillmonlikestopanswers.xyz It was alpha-2

@JosephWright oh, alpha, not beta, my memory playing tricks on me again :)

@JosephWright -- We've got this nice question on meta -- Often referenced questions -- that lists good candidates for duplicates. It's even segmented by topic and the people maintaining it try hard to arrange items in a logical order. Is there any way to point at it in the intro text for the site? The more people who know about it, potentially the fewer duplicate questions would be posted.

@barbarabeeton I learned shockingly late about this meta question myself, so I second this (also @JosephWright)

@barbarabeeton The mods can edit some of the intro text ..

@JosephWright -- Thanks. Let us know when you come up with something, or would like suggestions.

@JosephWright You can also feature it to give it a boost, right?

@yo' Yes, but that's a temporary thing

@JosephWright -- Might it be helpful to change that title to "Often referenced questions (identify possible duplicates)"?

@JosephWright is it by design that your sanitisers in \keyval_parse:NNn might strip braces if the argument is only a single group (non-sense example: \keyval_parse:NNn \foo \bar { {a=b,c=d} })

@Skillmonlikestopanswers.xyz Not by design: can you send me a test file?

@JosephWright I don't have a real-world use case. I was just taking a look at the implementation to get an idea of what it does in more detail (as for example \keys_set:nn does strip spaces around / in key names).

@JosephWright the above would be a usage example, for which I'd expect the outcome to be equivalent of \foo { a=b,c=d }.

@JosephWright Again, I doubt that this will ever be a valid usage, it only makes sense if one passes in a single key which must contain a comma or equal sign, which key names in general shouldn't.

Going to bed. See you tomorrow!

@Skillmonlikestopanswers.xyz That's deliberate, as the key name and path can be given in different ways

@JosephWright even that might be useful

Hi folks, am I the only person who browses tex.stackexchange.com with the following User Interface?

I am not browsing on a smartphone but a computer. But why does this happen?

@MoneyOrientedProgrammer You'vepicked the mobile version of the site ...

@MoneyOrientedProgrammer perhaps you clicked on the "mobile" link. Look if you find a "full site" link somewhere.

Oh OK. I am trying to find the link.

Thank you!

@MoneyOrientedProgrammer full site, at the bottom of the page, on a black ribbon

I got it. Thank you!