@yo' J sounds are problematic as there is no way to write the sound in non-borrowed letters. So when people start talking about 'garages' you have to do something radical... Hence, 'garejis' ;)
@cfr But back to the circularity, I don't think it's really circular. We first define 'symbol' as anything that can be drawn without lifting the pen. Then we define 'diacritic' which is anything that can be added to an existing symbol in a position above or below the baseline. (This might be too strong). Then we define n-graph which is a set of n adjacent symbols that behave as a linguistic unit. Languages then get to decide if 'letter' includes only n-graphs of n =1 (English) or n≥1.
(can't edit because of length) My definition of n-graph meant to include symbol+diacritic(s).
@cfr And if the language is consistent, then 'letter' is used for e.g. sorting, and naming. So for Welsh, digraphs are sorted and named separately, and for English they aren't.
@AlanMunn No, the upper case one. It was introduced in ancient Latin for distinguish the two sounds (voiced or unvoiced) of C, when lower case didn't exist; compare the abbreviation “C.” for the praenomen that was read as Gaius.
@AlanMunn So neither 'i' nor 'j' are 'symbols'. They are 'symbols' plus 'diacritics'?
@AlanMunn Yes. At least, those we classify as 'letters' are sorted and named etc. separately. This is not true for all digraphs. But chwaer comes after cyn in the dictionary. Indeed, chwaer is in a later section of the dictionary since it does not begin with C.
@yo' It means 'I'll see you!' Hopefully that is true... ;)
@cfr Yes, they would have to be under this scheme. Consistency has its downsides. But given that the upper case versions don't have them you might be able to make a principled exception for the dots.
@cfr Ah, so Welsh is more messy, since some digraphs are letters and some are not? So you have digraphs that behave as single sounds but are still considered two letters?
@egreg So then yes. :) I had no idea about the history of G.
@Canageek In the ACS case, it's the fact that they allow people to upload sources: for that to work you need to be sure that only a standard set of packages are used (and in their case no e-TeX)
Good morning. Quick question on tikz-cd: Isn't it possible to draw some tikz-stuff inside of a tikz-cd diagramm? See here: tex.stackexchange.com/a/231419 I tried to put nodes with coordinates or in respect to "current bounding box" but I failed. Can't find anything about that.
park my bicycle is more likely to survive logistical delays revive. Be revived only by bicycle pumps acted as a long-pressing on. And once tried to written this occurs. It is like a request for a substantial answer to the young and humane.
thank you..!
@LaRiFaRi See section 3.1 of the manual for a hint. You could do something like adding execute at end picture={\node [below left] at (current bounding box.north east) {stuff};} to the options for the tikzcd environment.
@DavidCarlisle the first one (\textsubscript was almost already in the dev version comming from a discussion with Ulrike some time ago, just haven't released it). I'll wrap it in the suggested test. I'll have a look at the float one.
@daleif thanks: the float one looks trickier as I'm not convinced the patch was ever doing the right thing, but the textsubscript one is more pressing in a way as it gives an error from \newcommand as the class is loading so means you cant use memoir with the test release at all.
@LaRiFaRi Whether they have internal names I don't know, but you could add a name with every matrix/.append style={name=m}. Assuming only one matrix is created. Then you could do execute at end picture={\node[above] at (m-1-1.north) {stuff};}
@LaRiFaRi Also, take a look at section 2.3 in the manual.
@DavidCarlisle, no that float patch is from before I took over. No real idea why it is needed. I'll have to take a closer look. It will not be until late next week.
@DavidCarlisle do you want a copy of the dev memoir to test the stuff on?
@TorbjørnT. Thank you so much. This is cool. The first approach did not work, but you pointed me to the right stuff. I just used |[alias=X]| inside the cells and they work as would be expected
Friends, have you ever used tgtermes? I tried it instead of tgpagella and the result was good. The reason I tried it was because my book was reduced from 96 to 91 pages because of the font. :)
@daleif Is it public? that is can I say anything on latex-l where there is a public test of the format that there is a test memoir release here... or even "put \makeatletter\let\@textsubscript\@undefined\makeatother before documentclass while testing with memoir, it will be fixed in the next update..."
@DavidCarlisle Reading glasses? It has begun. :) I used to call them reading glasses because I was in the denial phase. Nowadays, it's still reading glasses, but I happen to read all the time. :P
@DavidCarlisle There was a bishop in here who used to carry his iPad everywhere. I suspect he even had a Missale Romanum in it, just in case of an emergency mass. :)
@DavidCarlisle not much else than to make some test docs and see what happens with out it. I can add a little extra in there, such that the fix can be disabled from before \documentclass, would that be useful for testing?
@daleif It might be, as you know the note sort of says it is needed without saying when or what goes wrong if you don't.... So I didn't have a feel for it, although unlike the \textsubsctipt one the class loads so anyone who wants to test different version scan patch it up with a redefinition in the preamble, so probably I wouldn't complicate the class with temporary code. Anyone prepared to build a test latex is probably OK with some weird \makeatletter block
@DavidCarlisle I added a test for \MEMOIROVERRIDE, if undefined it will add the \AtBegin... block. I'll remove it again once we have figured out what this fix is actually for.
I'm trying to understand, in some detail, how a toc is generated. But now I'm thoroughly confused and perplexed after examining the .toc file before and after making a ``slight'' modification. I hope someone can get me past this bottleneck. [The final goal is to design a ToC where where I can pla...
@FaheemMitha Story is way more complicated. Google pdftk fedora, there might be some messages on the mailing lists (I won't link them here because they branch in several aspects, so it's preferable for you to pick one of your choice and go along).
It's already confusing for me to understand (as a Fedora user/dev), so it might be a living hell for the uninitiated. :)
Update: 2013-04-18
After upgrading to microtype 2.5 today this specific test cases indeed runs much faster (from approximately 3:00 minutes down to 8 seconds). However, I do have another cases where the compile time has not quite improved as much: only went down from 2:05 hr to 1:30 hr.
Questi...
@yo' ignoring a thousand line log discussing what letter means can't you do (or make listings do) \mbox{a}\mbox{b} to break any connection between the letters?
@PauloCereda I got an interesting workflow now: I run PDFLaTeX, but I have all the arara directives there, and when I need the full process, I just switch to arara for one compilation
@DavidCarlisle that was the surprising thing when I started debugging it. But it behaves even more strangly than I thought. A Q is coming in couple minutes
The image says it all:
I'm not sure what's going on, but it seems that microtype "breaks" the lrbox hack for coloring \lstinline. The code is here:
\documentclass[a4paper]{article}
\pagestyle{empty}
\usepackage[T1]{fontenc}
\usepackage{listings,xparse,xcolor,microtype}
\lstset{
basicst...
Just as a note: memoir v3.7c 2015/03/05 has just been uploaded to CTAN. As mentioned earlier the \textsubscript stuff have been addressed with the fix suggested by @DavidCarlisle. I also added a test to disable the rather strange memoir vs fixltx2e\end@dblfloatpatch. If the macro \MEMOIROVERRIDE exists, then that redefinition is not applied.
I will not be able to look further at the kernel updates vs. memoir or mathtools until late next week (have a talk I have to prepare, rather late as usual)
I know that I could do the reverse, i.e. put Mathematica graphics in TeX-based figures instead of putting TeX-labels in Mathematica plots, as @yo' pointed out. But I believe both of those approaches have their uses.
Thanks for all the help with this @DavidCarlisle and @yo'
Well, it seems somebody has already done something similar before ... And I just found it now. But I still believe my approach is better. He even got a publication out of it, even if it's TUGBoat, it's still something. I wonder if I could do that ...
@PauloCereda If you do try that Mathematica package, please let me know if it worked for you. I couldn't test on Linux yet. It requires gs 9.15 but you already have that.
@SeanAllred: I have Feed The Beast running a couple of mods for Minecraft and I found an epic texture pack named Soartex Fanver. A truly epic gameplay. :)
> it is my pleasure to announce, that the board of DANTE e.V. awards Till Tantau, Joseph Wright, Vedran Miletić and Mark Wibrow with the annual “Ehrenpreis” for their extraordinary and enduring work on beamer and PGF/TikZ. On behalf of all members I thank you for your commitment.
@PauloCereda @yo' I'm astonished that Debian are still packaging this. Is there any free alternative? (I still use it on Arch but Arch does not have the same stance vis-a-vis licensing as either Fedora or Debian.)
Is there any package for automatically inserting small spaces between groups of three digits to improve readability? E.g. to display twelve-thousand as $12\,000$ automatically, even when writing $12000$.
@daleif siunitx or simpler packages that just do grouping in threes eg my old comma package, but I have an answer on site somewhere (but I don't think I'd use it)
@JosephWright I have some mathcode "8000 silliness in an answer somewhere here..
You really really really don't want to do this.
But if you did want to do it, then you could do this, but you just know it's bound to break something.
\documentclass{article}
\usepackage{comma}
\def\commaformtoken{\,}
\edef\mca{\the\mathcode`0\space}
\edef\mcb{\the\mathcode`1\space}
\edef...
@DavidCarlisle My plan is to talk about all of that stuff in Darmstadt: I know Frank said something last year but I've got more to add and have more idea of what is actually implemented!
@PauloCereda Do you by any chance know which underlying bit of Fedora is responsible for generating notifications concerning software updates to the desktop? I can't get this to work on either of two Fedora 21 boxes I've got, and it is a pain. The only documentation I can find is just reassuring blurb from Fedora telling me that this will work by default. It doesn't tell me what to look at if it doesn't. [Why does documentation always assume that nothing will go wrong?!]
@DavidCarlisle -- well some others of us who write documentation try to address what might go wrong, but ... nothing is foolproof, and fools are getting more creative all the time.
@cfr Oh perhaps something along these lines? ask.fedoraproject.org/en/question/41058/… I haven't tried it myself because I always update my system twice a day out of paranoic habits. :)
@PauloCereda I don't want automatic updates. I want notifications of when updates are available so that I know they are there. I don't trust automatic updates - something could get updated just when I'm up against a deadline. I want control. But I want informed control, and I want the information to appear automatically. I just want what was default in Fedora 19 and is supposed to be default in Fedora 21!
@cfr -- oh, i've seen some tex documentation that leaves something to be desired. (even some i've written myself.) really -- it's pretty hard to predict everything that can go wrong, no matter how hard one tries.
(but i wouldn't dare people to try to find out by writing @DavidCarlisle's "use at your own risk".)
park my bicycle is more likely to survive logistical delays revive. Be revived only by bicycle pumps acted as a long-pressing on. And once tried to written this occurs. It is like a request for a substantial answer to the young and humane.
thank you..!
Update: 2013-04-18
After upgrading to microtype 2.5 today this specific test cases indeed runs much faster (from approximately 3:00 minutes down to 8 seconds). However, I do have another cases where the compile time has not quite improved as much: only went down from 2:05 hr to 1:30 hr.
Questi...
The image says it all:
I'm not sure what's going on, but it seems that microtype "breaks" the lrbox hack for coloring \lstinline. The code is here:
\documentclass[a4paper]{article}
\pagestyle{empty}
\usepackage[T1]{fontenc}
\usepackage{listings,xparse,xcolor,microtype}
\lstset{
basicst...
You really really really don't want to do this.
But if you did want to do it, then you could do this, but you just know it's bound to break something.
\documentclass{article}
\usepackage{comma}
\def\commaformtoken{\,}
\edef\mca{\the\mathcode`0\space}
\edef\mcb{\the\mathcode`1\space}
\edef...
