@MarcelKrüger the problem is that depending on the style the symbols are positioned independently, and only in the raised style they are lined up... And I do not know how to align a node on the math axis, which would be great (the have a base anchor for the baseline, but I am not aware of something similar for the math axis...)
@DavidCarlisle Sorry, maybe I wasn't being clear (as usual). The distance from the margin differs across two pages. That's what I meant by not aligned.
@PauloCereda -- That's great, but you (or someone else?) posted it earlier, so I've already seen it. Mama duck having fun while teaching her little ones.
@DavidCarlisle Hello, this is not very important, but doesn't in the file github.com/latex3/latex2e/blob/main/required/graphics/trig.dtx the line 95 contain \@clxxx instead of \@clxx for the value 180? (if so, the lines 89 ans 220 must be corrected accordingly).
@FaheemMitha saw it later but I don't see any longtable issue. You woul d see the same with a center environment: the text is centred on the text block not the page so if you specify different margins on left and right pagesm you get different margins on left and right pages.
@FaheemMitha I can't understand your concern at all, it lines up exactly where you specified it lines up. As ususal you have shown no code, but presumably you specified that it is left centered or right aligned within the text block
@quark67 oh wow I guess so but you are 27 years late with that bug report, why were you so slow?:-)
@DavidCarlisle I mean, if you are using twoside=semi then the longtable columns will not align vertically. That's all I meant to say. I was not suggesting it was an error.
And if the term "columns will not align vertically" is ambiguous or unclear, I mean that the column separators (the vertical lines) will not form one long straight line across pages. If twoside=semi, that is.
@FaheemMitha they align vertically if you stack the text block vertically, obviously if you offset every other page then every other page is offset. But there is nothing longtable related here you could say the same about any paragraph of text or an image
@FaheemMitha actually you said "It looks like KOMA's twoside=semi setting messes up the longtable alignment," which implies it's longtable related and wrong:-)
@FaheemMitha with twoside=semi they should as the margins are equal. But you are again showing no example and I will not try to guess what you are doing.
@FaheemMitha no, I'm just stating a fact, I came online to see a ping from you that looked like it was stating a longtblee problem. You may say it's just semantics and in fact you were just statung that you were getting the margins you specofied, but OK.
@UlrikeFischer -- From the messages coming through on the conferences list, it sounds like everybody is frazzled. But I expect they'll come through in time. (Maybe just in time.)
@PauloCereda I got a link (it is a bit bewildering as it doesn't say at which time one can use it) but Gert didn't got one. So currently it is a speaker-only conference ;-)
@UlrikeFischer -- Re links for the hoi polloi, here's the feedback from the back room: "It hasn’t. We are working on it. Please be patient." (I shall try to do so.)
@DavidCarlisle I saw something not behaving as desired (or expected). Not being an expert (or even knowledgeable on the topic) I mentioned it here. And mentioned longtable, because that was the thing not behaving as desired (or expected). I did not say it was a bug in longtable. I didn't know then, and I don't know now, what caused it. The reason I mentioned twoside=semi is that in my actual code, if I remove that setting, then things behave as expected.
But there seems to be something else going on, because I cannot reproduce the issue with just the twoside=semi setting in a test example. So if I want to find out what is causing it, I guess I have no option but to strip down my actual code.
@barbarabeeton He claims to be a fully fledged „polites“ of the Republic of TUG - regularly paying his membership fees and wanting his share of the spoils.
@UlrikeFischer -- Gee, I've been part of this gang for so long (since day -1) that I can no longer escape. I haven't yet figured out what the spoils might be ... just more work?
@PauloCereda This actually happened at a recent conference: one of the organisers sent around a personalised link to a webinar -- as a result 90% of the participants showed up under his name. The question and answer session was "interesting" as the chair tried to find a way to say who should ask next :)
@UlrikeFischer -- Lots of stories ... but I've actually missed a few of the more interesting places. I can think of one long-timer who has possibly been to more meetings than I have. He's certainly overworked, but I'm not sure how close he is to retirement. (He may linger past the "usual" age.)
@FaheemMitha -- Well, the Math Society gave me interesting projects to work on, and paid me enough that I never had to get a second job. But any work I've done for TUG has been volunteer, although AMS paid my way to most TUG meetings and didn't require me to take vacation time for those. So if there's money in this, someone else is getting it.
@FaheemMitha if you specify a different margin for odd numbered pages, then it seems quite natural that you get a different margin on odd numbered pages, certainly longtable is not involved here at all.