"@JG There are many factors that enter into a hiring decision, that's true. But you have to hire on potential. My advisor (who has a PhD in philosophy from Harvard) once said that Harvard should only hire dead guys, since they had a proven track record, and wouldn't impact on the amount of teaching currently done by the tenured faculty there. :-) "
@JosephWright Is there any chance I can talk you into putting in the plurals into siunitx? I keep typing \{2E-4}{\seconds} and then have to go delete the s.
@egreg I found another use of *: Exited nuclear state.
@BrunoLeFloch @egreg: Your solution is more comfortable. I think for an introduction it's a bad method to combine two "nearly equal" modules. It's something for advanced users ;-)
@JosephWright My blood pressure was getting a little low, so thanks for pointing out that question on meta. I've yet to find a way of putting my response that won't get me barred from the entire network so I find I can't post an answer. However, I've answered your much more reasonable and community-minded question instead and voted to close the original one as "Not a real question" - I found I was a bit spoilt for choice in reasons!
@AndrewStacey As @AlanMunn pointed out yesterday, there are times when it's best to compromise, and this is probably one. That was why I answered saying 'okay, we will simply not abbreviate at all in the FAQ'. As you also say, there is a logic to that for new users. (Checking the log, before we had 'Welcome to TeX.sx' we had just 'tex.stackexchange.com'.)
@JosephWright In which case it isn't a compromise! It's a resolution that is (hopefully) acceptable to all - even random strays from other SE sites that have never used our site before. My intention in my answer to your question was to say that in that place, any abbreviation is wrong. Also as Alan pointed out, they can't stop me saying "TeX-SX" in my comments.
@AndrewStacey I took a quick look around the other FAQs for the network. Most use '<site> - StackExchange', a few use just 'This site ...' without a name (e.g. electronics) and the stats site simply says 'CrossValidation' without the SE part at all.
@AndrewStacey I see your point about 'resolution' versus 'compromise'. Perhaps 'solution' would have been better.
@JosephWright So in fact there's no "common usage" and the fact that we call the site "TeX-SX" (since just "TeX" would be wrong) is therefore not out of line with other sites. Harumph.
@N3buchadnezzar who's organising the LaTeX course at NTNU? I keep seeing posters for such and meaning to send them an email about this place, but keep forgetting.
@AlanMunn Oh my! Those are really "estações bem definidas"! :P I once went to a city in the other extreme of São Paulo to give a course; I got 42ºC there, it was impossible for me to do anything! :)
I'm fine. :) I'm in a hurry to finish some stuff before Carnaval starts (this Friday). Brazil stops during this event and you can't hear from anyone. :P
@JosephWright Finally I've found a use for English-SX: they don't use the dashed form on their FAQ. I think we can point to them and say "They ought to know".
@AlanMunn Oh yes, sorry. You'd set parse-numbers = false, which means all input is treated in math mode. The alignment is then done using two boxes, but that means that additional $ cannot be safely tolerated. In a 'normal' table (without datatool) you'd need \multicolumn for the header.
@JosephWright So what's the best workaround? I created a \comma command that puts the comma into a savebox, but that seems like overkill. (The problem isn't the $ itself, it seems, but the , inside a $ ... $ element.)
@JosephWright Great, that works perfectly. I had an inkling that it would be something like that but I couldn't find it in the docs without knowing where exactly to look. Thanks. Now off to earn a Necromancer badge. :-)
@JosephWright I've sent the PT-BR translation to the TL team, but somehow the .po file was damaged when attached to the mailing list (data was scrubbed). The file was all messy and almost broke tlmgr. :) Now it's fixed. :) tug.org/pipermail/tldoc/2012q1/000423.html
The About page and the title for the main site page and for meta all say 'TeX - LaTeX - Stack Exchange'. We've had some discussions, for example when we had a skills contest and about the user greeting, stressing the need to cover all TeX-related matters and not just LaTeX. The 'TeX - LaTeX' thi...
@JG Not really. Although there isn't a good introductory, philosophically informed introduction to decision theory, so I guess parts of the thesis could make a book.
@Canageek Oh, I only do really basic stuff. But yeah, some people do logic basically full time.
@YiannisLazarides in your reply to tex.stackexchange.com/questions/44241/… you say one can move the blocks around. Certainly true, but then you are back measuring and given dimensions which isn't how the designer looks at it. Or am I missing something?
Oh yeah, here's an example of different attitudes to logic: I was reading about lambda calculus, and when it said "types" I was thinking of types à la Russell (i.e. in a kind of hierarchy of mess) but the paper I was reading was thinking about them as types of data. I'm not sure there's any technical formal difference, but the difference in outlook made some of the comments seem weird.
@JG Quite the reverse: I am to produce a thesis from which I can carve off chunks to publish as articles.
@JG Well, it's supposed to be a (long) cohesive argument. There are, in fact, chunks that are more or less self contained (including a couple that I've already extracted for publication), but their primary purpose is to be part of the main argument.
@FrankMittelbach I was thinking of a grid type of moves. Tchichold's cover is a nine horizontal grid or so I think. So was thinking of additional commands. That is a normal start for designers. So sort of turtle graphics but to grid positioning and steps. Plus relationships to adjacent objects.
@Yiannis as a general designer toolset that would be most certainly a very interesting approach and most certainly very valuable. So looking forward to your ideas (usually finding the right approach is 3/4 of the whole thing ... implementation is details :-)
@YiannisLazarides but on that oarticular title page I'm fairly sure that it isn't a regular grid or rather it may be a regular grid for the two main rules, but the inner title and author names have a direct relation that jumps any grid .. that relationship is most certainly not just by chance
@Seamus Ah cool! :) I never delved into this subject, but it's fascinating. Curiously, I have the Topoi book (reference 46 of the article), it's a very interesting approach. :)
@FrankMittelbach Sure. I have been battling with this sort of idea, since my entry in the TeX competition. Liked your coffin presentation but as you say it may need renaming:)
@YiannisLazarides that's just my weird kind of humor ... and it fits and so far nobody came up with a better name. I'm not so much concerned about the name but the syntax approach, even though it is already fairly usable.
@PauloCereda I've had an interest in category theory for a while, but never had enough of an excuse to actually look at it in detail. The connections to logic and computation might just make it reasonable for me to spend some time looking at it. (But I have to finish writing me thesis first, I guess)
@YiannisLazarides yes an underlying grid would be really good to have. But in fact that perfectly fits into the approach already, come to think about it. At the moment we usually start with this stupid \JoinCoffins \result \coffinA just to set up a \result box we can reuse. Instead we could go \InitGrid \result <parameters> and then have all the gid lines and intersections as poles available
@FrankMittelbach I am actually quite amused as LaTeX is the in-thing at the moment with all the youngish hackers -- the guys that get excited over haskell etc..
@FrankMittelbach Give me a few days and I can come up with some ideas:)
@PauloCereda Strange enough when I wrote that question, was thinking to compare the coffins with my technique as ... this is as pauper coffins are to Tutankamon's sarcophagus. Looked it up in wikipedia and could not see any handles and so left it and of course there are no handles in a pauper's cardboard coffin:)
@PauloCereda the latter being a poor choice of name also.
@JosephWright Ahhhh. I just saw a bunch of 'x's in names and figured they all meant the same thing. I was wondering why some were on the start and some were on the end.
This is possibly a very obscure problem. I tried searching for solutions but I had a hard time getting search engines to accept a literal ", &" string.
The problem is basically this. I'm using natbib with round and authoryear options. I'm using the citation style of Monthly Notices of the Ro...
@DavidCarlisle It's well known that the less technical and elaborate your answer is, the more reputation you get. Perhaps people upvote more if they can easily understand the answer.
@JG The OP found the answer by him/herself. Since the question is clearly not relevant (it was his/her fault for using a modified style), the question is closed for our own good. The intention of TeX.sx is to have good Q&A. :)
@DavidCarlisle Welcome to the wacky world of TeX.sx voting practices. Votes are driven by views, and simple questions garner more views than obscure ones, leading to an odd inverse correlation between amount of work spent on an answer and number of upvotes. Of course a pretty TikZ picture will get you votes as well independent of question complexity.
what is the convention if question and answer is given by the same person? am I to accept my own answer if nothing else comes along or just leave it as is?
mathpazo supports oldstyle numerals by swapping the font used, essentially.
I'm intersted in doing the same to the Pagella fonts .. but to do so, I think I need a source description of the fonts. Does such a thing exist? Or is my only option editing them using a font editing program that understands opentype/type 1?
@egreg seems like it (I had a version 1 here but never released, so I got all my files back from ctan (with the help of google to know what they were and where on ctan) and stuck them in googlecode so a public source control, better late than never:-) I expect I'll now get bug reports for xii.tex ...
@DavidCarlisle Actually I'm quite surprised that it took such a long time to be found. Nobody had ever tried a multiparagraph \multicolumn{n}{p{...}}{...} in a colored table before, it seems.
@AlanMunn Thank you Alan! So, I should put the centering immediately after the begin{table} and before the \caption line? How about the \vglue-24pt and \begin{adjustwidth}{}{3em}?
@JG Yes, but you shouldn't be doing any other manual formatting I should think. You should use the booktabs package, and possibly the caption package, which improves various spacing issues with captions.
@egreg realized a few minutes ago that that was part of the question and was just lookin up the definition in eTeX. It's the generation of \relax due to \fi seen too early in an expandable context I guess
@egreg and you have already documented it ... fast you are ;-)
@egreg you mean within the program? Don't really think so. It would slow everything down and there is no way you can really prevent all types of loops. Well this one there could be a count when inserting \relax but ...
@egreg but that would generate additional cases in other places, wouldn't it? TeX is extremely tuned for speed and those kind of htings would alter this
@egreg sn interesting comparison is \immediate\write 0 \if\fi{Hello} This expands the \if as well, but the \relax is not going away to after two attempts you are ok
@egreg well it expands \if being in the <filler> section and \if generates \relax twice. Don't really see this as being less <filler> compared to \if\unexpanded\fi
@FrankMittelbach The <filler> is considered during expansion to find {; the expansion of \if\fi is empty anyway, so \write0 doesn't really see the two \relax as part of a <filler>.
well but with \unexpanded the filler is also considered while looking for {, same thing. The difference is that in the case of \unexpanded the relax is immediately thrown away so it doesn't take part in the \if comparison, while with \write the \relax stays and so TeX hits \fi twice and then determines that the if is \true. With tracing you get {\immediate} {\if} {\fi} {\fi} {true} {\fi}
so the point is more that the execution of \write rmeoves the final filler not the any expansion along the way.
I just said in an answer that I wouldn't necessarily put {itemize} environments into a table. But apart from the list spacing (which can be fixed with enumitem) are there actually any reasons not to do it?
@AlanMunn I saw it; I don't understand what kind of information it should contain. OTOH, itemized lists in table are very commonly made with the-software-I-won't-mention
@PauloCereda It will appear during the next LaTeX course. :)
@egreg perhaps it is later in Germany compared to Italy, but I don't get it. Of course \if\fi does this by itself. But on its own it is clear that the \relax will stay. However, with \immediate\write0 \relax \if\relax\fi \relax {Hello} we are in the middle of looking for { expanding a filler ... but not removing the \relax along the way, right no?
@FrankMittelbach What I'm saying is that the expansion of \if\fi causes the appearance of a \relax, then of another, but then they disappear before being considered as part of a <filler>.
This is possibly a very obscure problem. I tried searching for solutions but I had a hard time getting search engines to accept a literal ", &" string.
The problem is basically this. I'm using natbib with round and authoryear options. I'm using the citation style of Monthly Notices of the Ro...
