@AlanMunn: But now the OP has changed his mind and accepted another question using ntheorem (while the OP was using amsmath), so it's not fear of packages.
I meant "accepted another answer"....
Ah... no. I read amsthm in the OP's question, but it is amsmath... Forget all I said!
@GonzaloMedina Just write something in the "description" area in your profile and you are done. It seems StackExchange also parses links and display them nicely in the pop-up window. ;-)
@AlanMunn: from Schubert I like The Trout, but more for the interpretation of Jacqueline du Pré than for the piece itself. I can perfectly live without Brahms.
@AlanMunn: I love du Pré... and yes, I like her Elgar Concerto, and yes, you're right it's a little bit too romantic for me. But my love for du Pré is bigger.
@AlanMunn, @PauloCereda: Now that we talk about scores, I've never used LaTeX to write music (I know that there are some packages, though). Has anyone of you ever used LaTeX to write a score?
@PauloCereda: If you're bored with palindromes ... you seem to know a bit about the data explorer, can you dream up a search that gives a list of people who have never asked a question? I could get a list of the number of questions someone asked, but wasn't sure how to get the complementary list.
@AndreyVihrov That's good! I was just trying to modify Paulo's to filter out those who hadn't asked many questions, but putting a cap by reputation is equally good. Thanks both of you!
Actually, I ended up doing 100 * n/m to get a percentage. Still worth knowing how to do the cast, though. Okay, so for users over 1000 reputation, we have 13% who haven't asked a question. Compare that with 22% for maths, 21% for English, 14% on cooking, and 10% on SO itself. So I don't think we're doing too bad, there.
@AndrewStacey In TikZ? Interesting things (z-buffering, pseudo-3D-shading) going on that would warrant one of us asking a question so you can show the result?
@PauloCereda Can you say how many pentagons and hexagons are on the fullerene surface? Without counting from the nice image Andrew has posted, of course. And without getting an old soccer ball, that also would be cheating. :)
This is an exercise I always give when dealing with graphs in my courses.
While we're talking of data explorer, it's clear that even tag badges are connected to popularity, although in this case it's tag popularity. I just earned the tikz-pgf tag badge, and while I know a little tikz, I wouldn't call myself an expert (especially compared to e.g. @Jake and @Andrew) but tikz questions are both frequent and popular, so their answers gain lots of votes. Things I actually know a lot about, like trees I'm unlikely ever to get the badge.
@JosephWright I hope that's just a feature of using x-y-z coordinates. The z-coordinate points in the direction (-.3,-.3) (or something like that) which skews the image.
On a sidenote: Now I'm addicted to these BuckyBalls. And I blame @AndrewStacey =P I never heard of a buckyball, so I googled and came with these magnets!
This is ridiculous! I appear to be about to hit the rep cap again, and this time it's for my answer on \end{document} (last time was for pointing out that one should use \verb for verbatim).
@NN no, bridge building games are still in my heart. =P I tried Minecraft too, but I always die in the Survival mode. It's like Darwin saying I'm too stupid for those games.
@AndrewStacey answers to simple questions seems generate substantial rep. people might not understand witty answers so they don't vote (me included...)
@PauloCereda Yes: 12 pentagons and 20 hexagons. Each pentagon is surrounded by 5 hexagons, and each hexagon touches 3 other hexagons and 3 pentagons. Let p be the number of pentagons and h the number of hexagons. We need to count V, total number of vertices; S, total number of sides; F, total number of faces.
Now V = (5p + 6h)/3 (each vertex is counted three times), S = (5p + 6h)/2 (each side is counted twice), F = p + h. From Euler's formula, (5/3-5/2+1)p + (2-3+1)h = 2, so p=12. It's easy to get the number h, now, isn't it?
@egreg Oh. My team wants to host Italy, so there's a big crowd to support la Squadra Azurra in here (there's a nice video about it here). I'll try to get you a uniform. :-)
I have a one-inch top margin, which works fine. BUT, if I have a page that's particularly empty (because LaTeX decides not to start a new section toward the bottom of the page), all the text on the page gets moved down one line. How do I fix this? I am using a custom .sty file which basically loa...
@lockstep I tried to help, but without an example it's difficult to go deeper. I suspect there's something wrong in the macros used or in the way \section is managed. I vote for closing.
I sometimes feel that I want a "watch list" as well as a "favourite" list. There are questions that I want to keep an eye on, but wouldn't put on a "favourite" list. For example: tex.stackexchange.com/q/28414/86. It's clear that this is potentially a duplicated, but I feel it's polite to wait until the questioner confirms it, or enough time has gone by to assume it. So I need a "take a look at this in N days time" reminder.
@NN In my opinion it is fine for a line to begin with a quotation mark. Note that tying longer words has negative effect on TeX's line breaking algorithm.
From reading Meta, it looks as though the "favourite" stuff is actually what I want here. From reading meta.stackoverflow.com/questions/53585/… the only thing I object to is the word "favorite" [sic]. Apart from a potential badge, there's no benefit to the questioner from "favouriting" a question.
*shouts*
Though I guess the reminder bit is a bit different, but maybe it's close enough that they wouldn't add a whole new feature for just this.
I think it would be very useful if SO added a Watch List feature. More than once I have arrived at a question and wanted to reply but I had no time at that moment or needed time to prepare some code to post... then I needed to bookmark so I could go back later.
Is there a watch list feature alre...
I am new to latex.
I have to prepare a one-page document (so float is not important).
I want to have 3 pdf graphs at the top of the page with 2 graphs centered on the first line and 1 graph centered below it. I want the two lines of graphs to be as close together as possible, so that I can have ...
Sometimes the OP forgets to add a MWE, and in some cases it's really easy to provide one. Would it be OK if someone else adds the missing MWE by editing the OP's original question?
@GonzaloMedina You could state humbly in a comment that you added an MWE for the sake of the quality of the question and that they are free to roll back or add an MWE of their own if they want to.
@GonzaloMedina Just to add: I do that from time to time. I try to do so in a non-invasive way (as in I just add to the question, with minimal editing of what's already there) and make it clear what has been added by me (I often prefix it with something like: Added (by Andrew Stacey)).
I'd like to be able to use a tool like grep to extract an entire bibtex reference rather than just the matching line.
For example, grep Smith mydatabase.bib would return lines that contain the word Smith. I'd like to have a command-line program that returned the entire bibtex reference that enclo...
@lockstep: Talking about the knuth badge, I'd like to suggest Don Knuth to the "Epic rap battles of history" series, say, Donald Knuth vs. Bill Gates, Donald Knuth vs. Donald Duck...
I'm trying to refer to an appendix in my thesis with \ref, but it generates an odd number (3.2). Is there a way to get something which makes more sense? (for example, that it refers to the appendix letter)?
@JosephWright: I know that the \prg_quicksort is marked as "Experimental", but I can't get the given example to work. Does it work, or should I avoid it like the plague?
Would a question about customized syntax highlighting in texworks be considered off-topic? (I'm consideriding to write one, plus an answer with my settings. I'd be interested to see other users' settings, too.)
I'm trying to create a Layout for user manuals for a couple of software applications we made.
One important aspect of the layout is that it must use TrueType fonts.
In my .lyx documents, I found a "\use_non_text_fonts true" line, and I added to my layout, but it doesn't work.
Where and how I m...
@lockstep Seems okay to me, as we have other editor-related questions. In particular, as TeXworks is Tex-focussed this seems in many ways more 'on topic' than questions about some other editors
@JosephWright Brilliant! That compiles, at least. I must confess that I did wonder about the extra parameters but thought it might be one of those funny macros that eats more arguments than you think it should.
@PauloCereda Yes, it's based on one in TUGboat years ago
@Raphink I didn't get a notification: thanks. That's a complex layout, and is currently well beyond what we have sorted. For example, we've not taken on flowing between frames (probably needs the output routine, for a start).
@AndreyVihrov We don't currently have an array data type for ordered data. (At some stage we'll have to port datatool, but that's not quite right, I think). I guess @BrunoLeFloch will handle this: he's good on efficiency.
@AndreyVihrov There's a quote from Knuth in that article, "Although the basic idea of binary search is comparatively straightforward, the details can be surprisingly tricky…"
@AndreyVihrov Remember that TeX is primarily a typesetting system, and so for complex data handling there is probably a better way (most obviously doing a LuaTeX-only solution and using Lua). However, I'm sure Bruno will take a look at it at some stage: after all, he's currently writing a regex parser in TeX :-)
@Caramdir I don't know. According to the project page, the development status is set as "planning". We could ask the project team, as they are active in here.
@AndreyVihrov That's tricky business. The main limitation in efficiency in TeX is that you don't have random access: to access one element in a list stored in a macro, you need to expand the macro, and then whatever method you use will take a linear time at least. Binary search would let you find an element in an array in O(n log(n)), which is not too bad, I guess. (cont.)
Other possibilities include: (1) an indexed array data type, of the form \marker 1\marker{...}\marker 2\marker{...}\marker 3\marker{...}, or (2) working digit by digit: 0{0{...}1{...}...}1{0{...}1{...}...}..., and I've had a couple of other ideas over the last few months, but none convinced me that it was the "right one". Do you have a specific application in mind? (cont.)
In particular, what operations do you think should be optimized for (slicing? looping? sorting? extracting one element? finding if an element is inside the array?) What kind of content should be supported?
@BrunoLeFloch Yuo might want to look at datatool (the current internals were suggested by Morten Hoegholm, and are LaTeX3-like, and somewhat similar to the old prop implementation from the toks-based days).
Actually, it is possible to have random access in TeX if you are ready to either use \csname table<index>\endcsname, which wastes the hash table, slowing the rest of TeX down, or to use \toks registers, of which there are only 32767. My current approach to regex matching uses \toks registers for fast access to various states of an automaton: I do everything inside a group so that I can freely use as many registers as I want.
@JosephWright I will, for sure. Not today, though: I need to see my supervisor in ~5minutes and was just passing by.
@BrunoLeFloch I was just going to mention the \csname … \endcsname as the closest to the classic array: You can have random access, but deleting or inserting an element is O(n) in general case.
@AndreyVihrov Of course xD. There are subtelties because TeX does some operations slightly faster than others (e.g., I avoid conditionals as much as possible to test for the end in the seq and prop modules).
@AndreyVihrov In most programming languages, deleting an item from an array is O(n) isn't it? You need to shift all other items, no?
@JosephWright Yes. That's the only sensible way of accessing toks ;-)
@JosephWright Well, I don't mind doing that soon. I have to say that I love this kind of programming much more than working on the galley or the OR, sorry :-).
@BrunoLeFloch Background for others: for LaTeX3 we are not using toks in general as for named variables \edef\macro{\unexpanded{<tokens>}} can hold #.
@BrunoLeFloch Well I can see why. I'm modifying my galley stuff at the moment, and it should be in the SVN at the weekend. I hope we are getting close to something workable: we really need to tackle the OR.
By the way, CTAN updates of l3kernel, etc. have been sent :-)
@BrunoLeFloch Don't forget to finish the l3fp updates :-)
@BrunoLeFloch I can't name anything more specific than implementing a set or a dictionary. My question was more of random nature, seeing that there is quicksort already. After all, LaTeX3 keeps pushing the limits of what can be done within TeX.
@JosephWright In LaTeX2e, the etex package allows you (I think) to allocate a whole bunch of toks at once, so I guess you're meant to access them as \toks\numexpr\base+<index>\relax.
@JosephWright Yes :-). I'll do a last push on l3regex this weekend: unfortunately, I realized I had made my life more complicated than it needs to be (with some elaborate chained list), and that's preventing me from fixing a few annoying bugs.
Then I'll have to be back to l3fp, and fight with it.
@JosephWright Good. I'll commit my prop accelerations tonight.
@AndreyVihrov We're not really convinced about the quicksort. There needs to be something like that, but it's not clear how this should be done and what interface to provide. Any idea is welcome.
I'm trying to create a Layout for user manuals for a couple of software applications we made.
One important aspect of the layout is that it must use TrueType fonts.
In my .lyx documents, I found a "\use_non_text_fonts true" line, and I added to my layout, but it doesn't work.
Where and how I m...
