When you absentmindedly forget that [ and ] don't need to be escaped. :)

@AlanMunn ... and to put a backslash after "e.g." ? ;-)

@JosephWright hyperref tries to disable some keys by redefining them, but this fails if done in a group (and so in AtBeginShipout). The suggested fix is to define a global \define@key variant to reset the key, but as far as I can see no keyval system has a "gdefine@key", so I'm not sure if this is a good idea only for one special case. What do you think?

@UlrikeFischer Don't really like the idea of a special case, and it gets tricky to implement. I'd imagine we can arrange to \aftergroup a function that does the work

@UlrikeFischer do we need the key itself to be global (rather than the handler for the key to have some global effect?)

@DavidCarlisle well \Hy@DisableOption{pdftitle} simply calls \define@key to redefine the key to issue a warning from now on. Imho redefining it so that the code first escapes a possible group as suggested by @JosephWright would be the right thing, the main problem is that the grouplevel can be 0 or 2 (and perhaps also 1 depending on what users do).

@UlrikeFischer We can check that and keep inserting the code until we hit level 0

@JosephWright The original definition is this

\def\Hy@DisableOption#1{%
  \ltx@ifundefined{KV@Hyp@#1@default}{%
    \define@key{Hyp}{#1}%
  }{%
    \define@key{Hyp}{#1}[]%
  }%
  {\Hy@WarnOptionDisabled{#1}}%
}

@UlrikeFischer As @DavidCarlisle says, we can do it at one redirect by having a global handler for the key

3 stars already on "put a backslash after "e.g."" I'm being punctuation shamed. :)

@AlanMunn I thought I had learned somewhere that you always put a comma behind “e.g.” (and “i.e.”).

@AlanMunn Style guides do not agree on whether or not a comma should follow both these abbreviations.. So never mind that, then.

@HaraldHanche-Olsen I use a comma after 'i.e.' but usually not after e.g. But here I'm being shamed for spacing after the . :)

@AlanMunn I changed it to 4 since you objected to 3

@AlanMunn I got that. But if there had been a comma, the spacing issue would not arise.

@HaraldHanche-Olsen Well this particular section of the paper is a direct quotation. I don't think I'm allowed to add a comma, except for with an obnoxious '[,]'. :)

@marmot Yes, @marmot s pretend to be all nice and friendly when really they're scheming in their burrows about bad spacing.

@AlanMunn Spacing is the single-most important issue if you live in a burrow.

@marmot lol

@AlanMunn I always use e.g., :)

@PauloCereda That is because you are fond of smileys, no?

@mickep ooh

@mickep I would have expected “e.g., (ooh)”.

@HaraldHanche-Olsen Mhm, you have a point.

@mickep @PauloCereda I am mean. :)

@HaraldHanche-Olsen ooh

@HaraldHanche-Olsen oh

Typographic question: Should parentheses ever be italic (say, if they are written in a theorem environment where the text is italic)?

@mickep No strong opinions. I prefer upright parentheses in equation references (as in equation (7)), though.

@mickep But it would lead to a visual clash if the previous comment was in all italics except the reference (7) – an upright parenthesis immediately followed by an italic one is certainly going to look bad.

@HaraldHanche-Olsen I agree completely when it comes to those. In another example here, we have something like "Proof (of Theorem 4)", where all text is set italic. I just changed to upright parenthesis, but I cannot say what looks best.

@mickep In that particular case, I would probably set only the word “Proof” in italic.

@mickep I have no opinion on the use of italics in theorems (which I find an odd practice to begin with) but there's a package to do this automatically that arose out of a question here.

@AlanMunn Thanks! (This document is written in ConTeXt, and I have no problems in getting the parenthesis I have in mind upright or italic.)

@AlanMunn Btw, I don't know the origin of italic in theorems, but looking at books, that must be very old.

Anyone know if this answer about crossref's in biblatex is still accurate? tex.stackexchange.com/a/148978/10038

@mickep -- AMS discourages italic parentheses in theorems because they are often intermixed with math that contains always-upright parentheses, and that's disconcerting. In non-mathematical text, though, where upright parens won't appear, italic parens are usually okay. (And by some, preferred.)

@StrongBad That answer is specifically not about biblatex, so it's unclear. biblatex and biber have extremely sophisticated cross-reference mechanisms. They make no mention of how your .bib file needs to be organized, so I would guess that it doesn't apply. But I don't use them, so I don't know for sure.

@AlanMunn I totally missed that it was for bibtex. I feel like the biblatex documentation is clearer than the bibtex documentation, but that is not saying much.

@StrongBad Sadly the biblatex documentation is more of a reference manual rather than a user manual.

@StrongBad Not that I'm complaining. I love biblatex. :)

@PauloCereda You use DuckDuckGo right? Is there an equivalent of the site: prefix that works with Google?

@PauloCereda Never mind. It seems that it works. For some reason I thought it didn't.

Would “rem lyx” be a good answer to tex.stackexchange.com/q/495356/4427 ?

@egreg Now, now, don't be hating on the poor LyX users. ;)

@egreg Shouldn't it be "RIP lyx"?