@JosephWright :20372050 :20372050 Need? None, I COULD write it by hand or on a typwriter. Would like? Something that will do citations well (biblatex-chem is really nice), chemmacros makes things a lot easier, that chem numbering one (chemcompounds or something like that?)
@Canageek I was thinking in terms of number of journals, but also Nature Chem. annoys me somewhat as there was no need from a 'community' POV to set it up
@JosephWright Given how profitable those journals are, even the 'non-profit' ones, I kind of wish I could invest in one. I hear they consider 30% profit to be a bad year.
@JosephWright Where are they going? I know that the ACS sends a bunch of people on caribbiaan 'meetings' each year and pays the head of it something like half a million a year.
@Canageek The real money is I think in databases: there's been some questions about the ACS and how they view the Chemical Abstracts part of their work
@JosephWright well, that depends on their workflow. I introduce tonnes of errors in Word articles, because I have to convert them to LaTeX, which is a lot of manual work
@JosephWright yep. And some make a PDF of whatever you give them and send it to India for re-typesetting. The errorness of some of the people there is in single errors per 100 pages.
@JosephWright @cfr @yo' On the 'letter' debate. I'm completely with Joseph here. Letters are orthographic elements which may or may not be linguistically decomposable, but are certainly orthographically decomposable. The technical term for letter pairs that correspond to one linguistic unit is 'digraph'. Examples from English are 'th', 'sh' 'ch' etc. These letter combinations correspond to single phonemes (or sometimes more) but are certainly composed orthographically of two distinct letters.
@Canageek CCDC was set up by Cambridge Uni but spun out some years ago as a not-for-profit company (they are based round the back of where I did my PhD)
@Canageek CCDC make money from selling data to industry, mainly
@Canageek Without saying too much, from the figures I've seen CCDC do academics a good deal on searching as we are not their profit centre but do put in >99% of the raw data
@Alan Because frankly, we do have CH, that one is not debatable. But if you wanted to go deeper, we probably have more "phonemes" than that, but these do not get their category.
@JosephWright That makes sense. It has never been one of the expenses we've winced at paying. (We use it enough that we have a deal with the library for a copy on one of our office computers)
@Canageek If you use it a lot that makes perfect sense, but for most people doing synthesis (like me) it's something you use a few times a year. In a small department like mine we'd probably struggle to justify a site license, but we don't have to!
@JosephWright (If I wind up applying to a faraday conference I'll get an RSC membership to make it cheaper, but I'm thinking of one in Krakow instead. )
@Canageek Don't you basically either pay before you go or pay the same as part of the conference fee? Last time I went to an RSC conference my then boss wasn't a member so had to stump up the extra £100-ish
@JosephWright That database profit thing suddenly makes everyone's reluctance to make DOI a standard part of the citation format make more sense. Making papers harder to find makes things more profitable.
@JosephWright Yeah, you pay before you go, but it is cheaper if you are a member, and at student membership rates it means I save money.
@JosephWright It wasn't MUCH money, but still money.
@Canageek I've never really struggled finding papers since things went electronic, although I do wish Wiley would learn from the ACS (best) or RSC/ScienceDirect about how to make a quick lookup work
@Canageek Sounds about right: I think for the students we did things in two parts, but for non-students it was the same either way
@Canageek Ah, you mean journals that are not necessarily published originally electronically. There, my library tends not to have access online anyway so ...
Does anyone know how I can stop multicolumn from spacing text out? I've written an equation followed by a final sentence, but it's spaced out to fill the remainder of the column down to the page.
@JamalS that looks almost as bad as \lipsum, but really how can we guess what's happening. Why \leavevmode there isn't that exactly where you don't want to start a paragraph?
@DavidCarlisle About errors in the log file: the lines starting with ! can be extracted easily, but it's not clear how many of the following lines also belong to the error. Is there an easy solution to this? I would like to extract all of the error description instead of just the !-line, but it's not clear how many lines to preserve afterwards.
@DavidCarlisle Maybe because of -interaction=nonstopmode ... I need to run latex so it will never wait for any user input from the terminal. Is there a different way than this?
@JamalS you have a different icon? I posted an answer, I also removed the font commands from \title and \author although I expect you'll put them back:(
Chat message I received from an intern today at work: That was my assumption. I assume everything from before 2016 does not work. I was laughing so hard. :-)
@DavidCarlisle Your solution creates a second page with part of a column, and on the last page the equation is separated, and text at the end is shifted to the right.
@JamalS they are wrong:-) Logically they are wrong as \title is supposed to take text that may be set in different places eg front matter and main title and ... same as \section argument may appear in the toc and the style should be determined in each case by the document style not the argument. Practicallyspeaking \maketitle isn't expecting font commands there so whether \author{\large me} affects just teh author or the rest of the title or the rest of the document is a matter of luck
@JosephWright collection of true books you mean? I've got only the TeXbook and then @wipet's yellow book somewhere I think, quite teared in pieces since the glue had dried.
Generally, we expect all questions here to be written from the point-of-view of genuine curiosity, and we expect all answers to be genuine attempts to provide real information and analysis that directly addresses the question.
However, the community has indicated that, if regulated carefully, Pu...
@DavidCarlisle every good Christian knows at least some Jewish traditions, and therefore knows what Purim is about, there's a lot of deep thought in Purim Torah, actually :)
@Canageek Sounds like a case where we'd say 'Southampton' (national service, rotating anode, confocal mirrors, most powerful non-sync system in Europe)
@JosephWright Nice. If we ever really need, my boss has said he'll look into sending me to lightsource.ca but so far I've been able to tease structures out with Cu/Long frames/150 K
@DavidCarlisle Chag Purim is a joyful holiday celebrating the victory of Jews. The joy is really the central motive of the holiday and good humour is an important part of joy!
@JamalS Using \be and \ee in your source makes the typescript difficult to scan. Use a good editor with autocompletion, if you prefer. But think to when you have to read your source.
@JosephWright It was the Bruker head crystallographers machine before we got it, then we upgraded it. So it is old, and a bit cranky, but was TOP of the line when made.
@Canageek Have used Bruker-Nonius but not a 'pure' Bruker. The national service here ended up going for Rigaku for their most recent set up. Oxford (now Agilent) is a good system for routine work.
@Canageek Never used a copper system (we didn't have one when I was at Southampton and don't have one here: if we had the money I'd push for a dual source)
@yo' As for Czech, there are two possibilities: we can define 'letter' to be 'orthographic phoneme' in which case, of course then Ch is a 'letter'. But this is certainly not how English defines 'letter'. Again, as @JosephWright said, Old English ð and Þ are single letters which in Modern English are written as digraphs of two letters, even though they correspond to single phonemes.
@yo' But the point is that in Czech (as in English) 'C' exists as a letter, and so does 'h' so to say that the digraph 'Ch' isn't orthographically composed of these two letters is really odd to me.
@AlanMunn well, the other option was to make it ȟ (which I would like actually). Because otherwise the Latin alphabet is just too short for our language :)
(well, X could be used, but it was already used in many borrowed words, moreover, we've got a strong German influence suggesting the spelling "ch" for the sound [x])
@egreg I have no problem with that. It's uncontroversial that digraphs behave linguistically as single things, and so being in the sort order is part of that.
@Canageek Ah, well then I suspect she won't have much appreciation for what LaTeX can do. Unless there's some package for writing discourses. (Which are actually really hard since they involve overlapping speech and a whole bunch of other weird things that are hard to put down in written form.)
I once spent hours learning enough TeX to format my ex-gf's resume for printing on the computing center laser printer (back when laser printing was magical) and used up most of my monthly laser printing quota printing copies of it -- all under the mistaken belief that she'd see that she was craz...
@AlanMunn I think you miss my point: searching in general is covered by tools that do that, there's not a pressing need to add all of these things to your 'own' database unless you actually want to cite them
@JosephWright Searching locally through a small number of papers I know are the ones I want is a lot faster then going through the full list every time
@JosephWright I was more joking, than not, but I do find its bib info very annoying, mainly because it has no full first names, which in Linguistics are almost always required.
@JosephWright well, you should cite the more the better. For instance, the full names if you know them. And I always use AMS's MathSciNet and its nice BibTeX export, I only add the full names if they're missing.
I'm trying to use the "precompiled preamble" approach with the exact steps from this question. I see absolutely no speedup in compiling afterwards. My preamble is very short and only has \documentclass{standalone} \usepackage{amsmath}. Is it normal that there's no speedup?
@Canageek No, because the citation is meant to be referring to the original publication and should therefore be derived from it alone. This is pretty well established in some areas, more humanities I guess, where life can get a lot more 'interesting'. You can of course add it as 'clearly editorial': D[onald] Knuth ...
@Szabolcs you need big packages like tikz to make it worth it. the document itself won't run faster you only save the load time which these days for amsmath is not a lot of time
Or is there maybe some trick to keep the pdflatex process running and get it to output several documents as requested? I'm trying to compile a line or two of LaTeX (just math) and get a PDF from it. I need to do this many times so I would like to make it as fast as possible, so I'd like to reduce the overhead to the minimum
@JosephWright What if it is a review that says J. Wright at the top, then has a biography with a picture and full name at the bottom of the first page? (I've been reading a bunch of those as of late)
@Canageek Now that's a tough one: I have to say I'm not sure how I'd handle it. I guess as the data is in the original paper as a whole one can use it.
@Szabolcs there have been people who have built tex servers that keep running but it requires changes to the source, and as far as I know they were never generally available
@yo' I am using LaTeX to generate labels and annotations for a figure. I call pdflatex from some other system. Each label will cost a pdflatex call, i.e. about 0.7 seconds. It would be nice to speed this up. But if I can't, it's not a big deal. It is not possible to run the calls asynchronously.
@DavidCarlisle OK, I'll take that as not possible :)
@JosephWright Also, dammit, Aaron left my middle initial out of one publication, which means now it will always lack it. Which annoys me as my name is really common.
@JosephWright I think all chemistry journals just initials anyway, so the only irregularity would be if middle initials are appearing and disappearing between citations, which annoys the heck out of me.
@Canageek The flip side of that is my dissertation, which gets cited a lot, contains my full name, yet I never use my middle name or initial in anything else I write.
@yo' Indeed. @AlanMunn had better rewrite all the Welsh ones, too. These are not treated like the examples given from English or like similar examples from Welsh. mh is a phoneme but it is not a letter. Similarly si etc. But Ch, Ll, Ff etc. are completely different.
@AlanMunn You have even more funny examples in Croatian (that's a good language for these things). There are three letters "d", "ž" and "dž", with the pronounication [d], [ʒ] and [dʒ]. The reasons why the letter "dž" is kept as a letter are moreorless historical.
on an entirely different topic ... has anyone here had an idea to submit an article for the next issue of tugboat? the deadline is coming up soon. if you have something, please send a note to tugboat@tug.org as a heads up.
@AlanMunn The alphabet I learnt has 28 letters or 29 if you include J. Of those, several are written using a pair of glyphs which individually represent other letters in the alphabet. They behave as one letter. For example, they are separate letters when putting things into alphabetical order or when assigning things to sections of a dictionary.
Also, they matter grammatically. If 'ch' was 2 letters, then you'd have 'fy nghhwaer' just as you have 'fy nghath'. But that's not how it works. You have 'fy chwaer' because ch is not subject to mutation (ever), whereas c is.
I borrowed an image from Herbert's answer without his permission.
User defined data:
\def\M{3}% columns
\def\N{3}% rows
\def\scale{1}% scale
\def\filename{herbert}% filename
% Specify the cropping area.
\def\L{-2}
\def\B{-1.5}
\def\R{2}
\def\T{3}
Steps:
Specify the number of columns and ro...
@cfr But that's again confusing linguistics with orthography. There's no question that linguistically, digraphs are monolithic, and so can have different phonological behaviour compared to their component parts. Similarly, the could well be treated differently for sorting (e.g. there's no good reason why we shouldn't sort all the 'th' words as a group before or after the 't' words in English.) What's a bit odd to me and Joseph is using 'letter' in this way.
@cfr @yo' So your operational definition of letter is 'group of 1 or more symbols that behave as a unit with respect to orthographic and linguistic conventions.' Our version of letter is just the 'symbol' part.
@AlanMunn What I mean by letter is what functions as a letter. What is taught to children as a letter. What gets its own section in the dictionary. What you need to treat as a separate letter in order to get the mutations right. If you are using some technical term 'letter' in a specialist sense, fine. But Ll is a single letter in the ordinary sense of the word.
@AlanMunn maybe. We have it the other way around: Letter is what functions as a letter, no matter how it was written. If you started writing Ǩ instead of CH, nothing would change at all. So it is a letter. For me, TH in English would make a good letter :)
@AlanMunn But that is not what ordinary speakers mean by 'letter' unless they happen to speak only a language in which there is a single glyph for each letter of the alphabet. Anybody who learnt a language where that's not the case (at least as a child) will say that 'letter' is what functions as a letter. Regardless of how it is written. For that matter, what makes 'i' a single letter but 'ng' two? Just that the latter has a horizontal space, whereas the former has a vertical one?
But the problem with English is: the language is far from being phonetic, so it's completely not clear what should be a letter and what should not. From the historical perspective TH is clear, but what about SH (Š in Czech) or CH (Č)?
I borrowed an image from Herbert's answer without his permission.
User defined data:
\def\M{3}% columns
\def\N{3}% rows
\def\scale{1}% scale
\def\filename{herbert}% filename
% Specify the cropping area.
\def\L{-2}
\def\B{-1.5}
\def\R{2}
\def\T{3}
Steps:
Specify the number of columns and ro...
