TeX, LaTeX and Friends - 2015-04-29 (page 1 of 2)

« first day (1641 days earlier) ← previous day next day → last day (3284 days later) »

2:41 AM

@cfr @Werner Now I've seen everything: Americani{sz}ing the spelling of a working link that doesn't even show up in the rendered answer, but just in the markdown. Amazing. Thanks for rejecting the edits. I wonder if this was a honeypot.

@AlanMunn because the heading text on Wikibooks was Americani{sz}ed at some point, the anchor is no longer valid; the edit was simply correcting the anchor so the page jumps to the intended spot.

@PaulGessler Ah, you're right. I didn't notice the anchor difference. I guess I'll fix it then. Thanks.

3:31 AM

@AlanMunn @PaulGessler The other edit was certainly egregious, though. I looked at that one first and was probably less careful about the second....

3:44 AM

Actually, maybe that one was just to lift the character count...

4 hours later…

8:01 AM

Using xtring what is a good way of testing for @ in a string, \IfSubStr works fine as long as @ is not a letter, it which case it does not catch it. What am I missing?

\documentclass[a4paper]{memoir}
\usepackage{xstring,url}
\begin{document}

\def\test{daleif@math.au.dk}
\makeatletter
%\fullexpandarg
\IfSubStr{\test}{@}{
\expandafter\url\expandafter{\test ddd}
}{
\test rrr
}

\end{document}

8:32 AM

\documentclass[a4paper]{memoir}
\usepackage{url}
\begin{document}

\def\testa{daleif@math.au.dk}
\makeatletter
\def\testb{daleif@math.au.dk}
\def\testc{daleif://math.au.dk}

\def\z#1{%
\edef\tmp{\noexpand\in@\string @\expandafter{\expandafter\detokenize\expandafter{#1}}}%
\tmp
\ifin@\expandafter\@firstoftwo\else\expandafter\@secondoftwo\fi}

a \z\testa{yes}{no}

b \z\testb{yes}{no}

c \z\testc{yes}{no}

\end{document}

@daleif ^^^^

8:47 AM

@DavidCarlisle interesting, thanks. Any idea why an xstring test would not work? I'm parsing some address stuff where the last line in some input might be an email address.

@daleif My only experience with xstring is in questions on this site, and I seem to find that it's used rather less in the answers than the questions:-)

@DavidCarlisle too bad. It ought to be able to do some interesting things. Your macro works fine in my application

@DavidCarlisle Every time you do that, God kills a xetten.

9:18 AM

20 rep points after logging in, hm ... Looking for details, BOOM! Several badges, upvoing and upvoting reversed. Da ist man einmal nicht da ...

9:30 AM

0

Q: Figures generate improper text spacing/breaking with IEEEtran and latexrelease package after MiKTeX update

After updating MiKTeX on 23-April-2015, compilation of my TeX files that use the fixltx2e.sty package was broken. According to the LaTeX News Issue 22 article, the fixltx2e package has become obsolete and the functionality has been integrated into the LaTeX kernel. To fix my compilation problems ...

floats miktex ieeetran

@DavidCarlisle @JosephWright The new stuff is already life in MikTeX?

@Johannes_B Yes: Christian has his own update schedule that we can't do much about

@JosephWright Wow, i had no idea this would go into the wild that fast.

9:55 AM

@JosephWright so it is a "bug in MikTeX"?

@yo' No

@yo' I suspect it's one of the things we've been worrying about for some time: invalid assumptions about fixltx2e/format/...

@JosephWright yeah, possibly

10:47 AM

@JosephWright been at a meeting, I guess I need to read that question?

@Johannes_B forced by texlive policy really, things can not go to texlive, even pretest unless they are on ctan, and if they are on ctan, anyone can use them. So we had test release of latex via dropbox, but really only one or two people test that, to get wider testing once a texlive pretest starts we need to make it all live

@DavidCarlisle I can understand the policy, but it is somehow strange.

@Johannes_B for GA there should be another version of the packages on CTAN. Maybe beta/RC flags should exist for package updates.

@yo' There is the TLcontrib system but it's never really taken off, and you have the same 'does anyone test it' issue

@yo' Nobody is testing properly, it would never go live.

@Johannes_B but you could at least make it the other way around: Do proper TL tests before the thing goes truly live on CTAN

11:03 AM

@yo' Who is testing the kernel stuff? @DavidCarlisle and @JosephWright and who?

@yo' perhaps for latex you could do that, but it isn't clear about the 1000s of contrib packages, ctan is the only place with the infrastructure to accept author submissions

@Johannes_B our code is all brilliant and doesn't need testing

@DavidCarlisle they need not use the feature and sumbit only releases ;)

anyways, you seem to know better than me what you say :-)

@Johannes_B depends what you mean by test, we have a fairly extensive test suite (recently ported to l3build by Joseph) but that tests internal consistency of the core release, you don't find it breaks memoir or xeCJK or whatever else it breaks until you get a wider pool of people testing it on real documents

@DavidCarlisle Exactely, you need the weirdness people put in documents. You cannot think of the stuff people have if you were on LSD, and this is the code breaking stuff.

@Johannes_B right, you need to have the stuff people got when they were on LSD

11:13 AM

@yo' I somewhere read the lab-book entry of the guy experimenting with the LSD stuff in the beginning. Describing how he left the lab and riding the bike home in slow-motion. Was quite interesting.

@Johannes_B lol

@yo' There is even a wikipedia entry :-)

History of lysergic acid diethylamide

The psychedelic drug (or entheogen) lysergic acid diethylamide (LSD) was first synthesized on November 16, 1938 by the Swiss chemist Albert Hofmann in the Sandoz (now Novartis) laboratories in Basel, Switzerland. It was not until five years later on April 19, 1943, that the psychedelic properties were found. == Discovery == Albert Hofmann, born in Switzerland, joined the pharmaceutical-chemical department of Sandoz Laboratories, located in Basel as a co-worker with professor Arthur Stoll, founder and director of the pharmaceutical department. He began studying the medicinal plant squill and the...

@Johannes_B wow!

How do I go about informing the developers of PGF/TikZ of a bug and attempted fix? They don't seem to look at sourceforge (at least, not in the past year)...

@Josh There is a pgf mailing list you could try.

11:28 AM

Ok, I'll give that a go.

@Josh I would still also file an official bug report. It's a large project; even if you have a feeling that currently the maintenance is not optimal, it is important to have everything in one place.

I filed the bug report a year ago :P

@Josh doesn't it mean, before v3.0.0?

no, after v3.0.0

I'm pretty sure at least

ok then

11:31 AM

yeh, v3.0.0 came out in 2013

@Josh There is also the patches section, where you can suggest your patch sourceforge.net/p/pgf/patches/search/…

@Johannes_B I'll have a look at that too, though judging by the activity, it seems pretty dead too :(

@Josh But at least, the patch is out in the world. Others can see it, even if not officially within the project.

@Johannes_B

@Josh Looking at that, there is quite the reaction delay ;-) sourceforge.net/p/pgf/patches/12

11:36 AM

@Johannes_B Yeh, it seems like it. I've made a mention of the fix over at tex.stackexchange.com/questions/86588/… because that's where it was originally found

@Josh Is that your rep on github?

What do you mean?

Oh, yes it is

Once the fix is in TikZ, I intend to remove it

@Josh Please have a look at github.com/Qrrbrbirlbel/pgf. There is you savest bit i see atm.

What is that repository? There is no description or anything (other than "PGF/TikZ")

It seems like a (random?) TikZ library that someone has written

@Josh Right, that is just a library. Nobbz on the other hand seems to have done some work on the whole thing. Both users are active here and on TeXwelt.de github.com/NobbZ/pgf

11:50 AM

@Johannes_B The Nobbz repository is just a mirror of the CVS repository. I guess I could fork that, apply the patch, and create a pull request too, but if they don't look at SourceForge, I doubt they would look at a github clone of sourceforge... I'll try and email the mailing list and submit the patch on SourceForge.

@Josh Seems to be a good idea.

@daleif You're missing expl3. ;-) VVVV

\documentclass[a4paper]{memoir}
\usepackage{xparse,url}

\def\test{daleif@math.au.dk}
\makeatletter
\def\testl{daleif@math.au.dk}
\makeatother

\ExplSyntaxOn
\NewDocumentCommand{\MailOrUrl}{sm}
 {
  \IfBooleanTF{#1}
   {
    \daleif_mailorurl:V #2
   }
   {
    \daleif_mailorurl:n { #2 }
   }
 }
\cs_new_protected:Npn \daleif_mailorurl:n #1
 {
  \tl_if_in:fnTF { \tl_to_str:n { #1 } } { @ }
   {
    mailto:\nobreakspace #1
   }
   {
    \url { #1 }
   }
 }
\cs_generate_variant:Nn \daleif_mailorurl:n { V }

12:09 PM

@egreg I know, expl3 still gives me an headache ;-) And I'm not even sure I understand what is going on here. I hope to learn expl3 a bit better over the summer

@daleif I define an internal macro that checks for @ in the “stringified” argument, so a possible @ given in a \makeatletter context will become again catcode 12.

@daleif he just copied my code and changed @ to _, \in@ to \tl_if_in: etc.

@egreg I don't quite understand the variant stuff

@DavidCarlisle :P

@daleif The kernel provides \tl_if_in:nnTF, but we want that the first argument is “stringified”, so I create \tl_if_in:fnTF that does f-expansion (in this case x-expansion would be OK too) on the first argument, before the standard tl_if_in:nnTF acts.

@egreg And the V stuff?

12:18 PM

@egreg @daleif so basically rather than having to explictly do \edef to a temporary macro and then \expandafter tricks to expand that into the argument as in my version, you just change an n specifier to x (or in this case f) and the pre-processing of the argument happens behind the scenes.

@daleif value so \foo:V\zz is \expandafter\foo:n\expandafter{\zz}

Hey

I want to set up a math font for use with TeX... so it takes a lot of time, that doens't bother me. Where do I start? Is there any documentation for this? I've only heard people trying to deter me from such an undertaking.

Christian Hupfer

Good maen

@ChristianHupfer Hy

Christian Hupfer

@1010011010: On the hunt for new tex fields no one has gone before? ;-)

@ChristianHupfer Imagine you have everything to typograhical perfection, EXCEPT the math which is a completely different font. I just find it very hard to accept. And I don't think I will. Since I'm working on something "real" this time, I can't justify some random math font.

12:35 PM

@1010011010 do you mean set up an existing math font for tex, or make a new math font (that is are you wanting to use tex macros or a font editor)?

Christian Hupfer

@1010011010: Typographical perfection? That sounds very promising. You are right, the math font should be correct then. Ahead a lot of work I see for you. Unclear the path is ;-)

@DavidCarlisle I want to create math support for a specific font I'm currently using in my body text. I'd guess I'd need a font editor at this point?!

@1010011010 yes but are you prepared to draw all the mathematical symbols in a style matching your body font. that is where the work is (for the stix font that took over 10 actual years and how knows how many man years)

@1010011010 Font designers are paid to do stuff like that and they know how to do that. There are many symbols you would have to design. Are you willing to engage in a project where the result won't be satisfying?

@Johannes_B I'm a perfectionist... that's all it takes I think... That, and time. I'll have existing math fonts as guidance to achieve a similar optical result. I don't need every single math character ever, for my current project. I can continue after the project is done, as something on the side.

12:43 PM

@1010011010 I hope you will open-source the math font? As we are in need of good math fonts as you have noticed.

@Johannes_B I don't know if it's allowed. The foundry that made the text font is pretty strict, from what I heard.....

It's a commercial font.

@1010011010 since by far the largest part of the work is drawing all the glyphs that determines how long it will take, If you just need + - then an afternoon, if a reasonable subset of the uniocde math range as in stix you need a couple of thousand or so glyphs and it should take years. But it is exactly like asking "how do I draw a picture" There are several ways it depends on what tools money and talent is involved.

@1010011010 Pity

@1010011010 but 99% of the time you don't need to do that, you should be able to find a reasonably matching math symbol font and then set things up to use the letters from your text font. Do you really need to re-draw < ?

@DavidCarlisle I don't necessarily need to, but I find it weird to get all the kerning, protrusion settings, inter line spacing/inter word spacing up to 0.1% accurate by zooming in 1200% all day long to fix things, then use a set of random math symbols to go with the font... It just feels so wrong

I don't care about time investment if the result is that much better. Okay, I won't need fancy symbols e.g. for the vmatrix environment or what have you, but some math radicals can definitely be improved. It'll take a while, but once you got something that works for you, a couple of glyphs per day isn't an unreasonable amount of progress to expect, no?

12:51 PM

@1010011010 some people have commented adversely on my artistic ability and so I am not sure I'd have confidence to do two a day. Can you see anything wrong with the image here for example?

2

A: Scale image based on the size of another image

This shows two figures each scaled by the same amount, such that the larger of the two is \linewidth wide: \documentclass{article} \setlength\textwidth{.5\textwidth} \setlength\textheight{1.2\textheight} \usepackage{graphicx} \begin{document} \sbox0{\includegraphics{house.png}} \makeatlett...

@DavidCarlisle I must say I'm baffled...

@1010011010 "baffled" ?

Yes, quotes would've been appropriate. Thanks.

@1010011010 I am sure you meant "impressed" but I'll let it pass. So If I were asking how long it would take me to draw 2000 characters to make a math font, what answer would you give me ?

@DavidCarlisle Hey, you're the math guy! ;-)

Regardless, it depends on how productive I'll be, sure. Some characters can be taken directly from other fonts, think of absolute value signs and all that trivial stuff. The math support is pretty bad as it stands, but it's not diabolical. Basic signs like the greek capital Sigma for the summation sign can probably be scaled, there's a letter pi... just to name a few examples. Just need to fill in the blanks and respect the font style as much as posible.

And so... after having the math glyphs all compiled in a single font, I can just tell TeX to use OML encoding for that font? After otftotfm'ing it?

1:09 PM

Best table in world golatex.de/viewtopic,p,71084.html#71084

@1010011010 Why not using XeLaTeX/LuaLaTeX? I bet your commercial font is open type as well.

@Johannes_B I'll do that as well. Currently I have an ifxetex conditional in the classfile to use fontspec, and just use a map file/fontenc and so on if pdflatex is used.

@1010011010 I don't know what conversion steps would be needed to use the font with pdfLaTeX.

@1010011010 If you take glyphs from different fonts and mix them up in one, won't the result look like a ransom letter?

@Johannes_B Some math symbols (arguably) don't inherit any style since they are so minimal, absolute value signs are the prime example that come to mind... At worst you have to increase thickness.

@Johannes_B I guess this comes pretty close... tex.stackexchange.com/questions/119714/…

1:28 PM

@1010011010 I'd start with looking around gust.org.pl/projects/e-foundry/tex-gyre that shows you what needs to be produced and here and there are also links to papers showing how they set about doing it.

user image

@1010011010 As you can see, even the plus minus equal signs differ significantly.

\documentclass[border=10pt]{standalone}
\usepackage{fontspec}
\usepackage{unicode-math}
\usepackage{libertine}
\usepackage[libertine]{newtxmath}
\begin{document}
\begin{tabular}{rc}
	\small tasty flower dog &
\setmathfont{latinmodern-math.otf}
$\displaystyle a+b-c=x $ \\
\small tasty flower dog &
\setmathfont{texgyreschola-math.otf}
$\displaystyle a+b-c=x $ \\
\small tasty flower dog &
\setmathfont{texgyrepagella-math.otf}
$\displaystyle a+b-c=x $ \\
\small tasty flower dog &
\setmathfont{STIXMath-Regular.otf}

@Johannes_B Plus and minus signs are in the font. As I said, it's just some very specific symbols that are missing. Mostly math radicals.

@DavidCarlisle Which papers are you referring to? The funding papers don't give much insight...

@DavidCarlisle I somehow know how you feel. Playing around with fontspec and xelatex, i had to kill xelatex because it wasn't responding. ^C didn't work.

@1010011010 don't know I've read papers on the gust project somewhere, I assumed they were there, they may have been in tugboat. google is your friend:-)

@DavidCarlisle Thanks I'll look into it.

2:00 PM

@ChristianHupfer If you look at the traffic over at goLateX today, you will realize, today seems to be one off those days.

Christian Hupfer

@Johannes_B: I've not been on goLaTeX for weeks now...

2:12 PM

@ChristianHupfer Have you ever watched Braindead? ;-)

Oh, one more thing, if anybody knows a sort of efficient (but manual !!) way to map each character to the right control sequence, and so on, I'd really appreciate some kind of example/guidance on that... I saw the math abx fonts but I couldn't find any specific unicode characters in the definition files there

Well, it doesn't have to be unicode, but it would be nice.

Battery's almost up. Gotta run.

@1010011010 Good start might be fntguide. See you :-)

Christian Hupfer

2:28 PM

@Johannes_B: Nope

2:51 PM

@1010011010 just copy the lines from unicode-math.sty which is nearest we have to a standard set of tex names for unicode math characters (or better simply use unicode-math package to do that, and just specify your new font)

Wow, a clearpage before each section, and subsection, and subsubsection. I hope the author is planting new trees every day.

@Johannes_B ... trees .. paper... oh yes I remember that, it's what people used to use to read documents isn't it?

@DavidCarlisle I can hardly remember it. :-)

Now the template introduces beamerarticle

1 hour later…

4:27 PM

@yo' Thank you! Things seem to have settled down now :)

This is quite an interesting read: github.com/novoid/Student-Supervision-Process

4:59 PM

A class defers loading a package to AtEndPreamble, but the user already loaded the same package and an option clash occurs. How can i prevent to load it the second time?

@Johannes_B add the option on the first load

@DavidCarlisle That's so simple, didn't thought of that. Thanks :-)

Hi, I'm having a strange biblatex/biber problem. I compile, run biber get no errors, or warnings but still no citations or bibliography appear.

5:14 PM

@AlanMunn What does the blg say?

[1] Config.pm:320> INFO - This is Biber 1.9
[1] Config.pm:323> INFO - Logfile is 'handout.blg'
[158] biber-darwin:277> INFO - === Wed Apr 29, 2015, 13:10:38
[159] Biber.pm:333> INFO - Reading 'handout.bcf'
[376] Biber.pm:635> INFO - Found 1 citekeys in bib section 0
[467] Biber.pm:3051> INFO - Processing section 0
[529] Biber.pm:3189> INFO - Looking for bibtex format file 'johnson.bib' for section 0
[3469] bibtex.pm:990> INFO - Decoding LaTeX character macros into UTF-8
[4886] bibtex.pm:864> INFO - Found BibTeX data source '/Users/alan/Library/texmf/bibtex/bib/johnson.bib'

@AlanMunn There should be more of it. Output written on ...

@Johannes_B So then maybe biber is not finishing? The bib file is very large. Perhaps I'm just impatient.

@AlanMunn Never hurts to trash biber and start over http://tex.stackexchange.com/questions/140814/biblatex-biber-fails-with-a-strange-error-about-missing-recode-data-xml-file
Look who asked the question ;-)

@AlanMunn You might have more insight from a terminal window.

Gotta run for now, be back in an hour.

@Johannes_B No, that error I know. :)

@Johannes_B I think it's just super slow.

5:33 PM

@SeanAllred good to hear!

5:46 PM

Popularity lemma: Question with 1000 views gets only 1 score :)

@yo' Link?

@JosephWright It's not on this site: electronics.stackexchange.com/questions/103678/…

@JosephWright Same idea with \exp_not:c :)

@egreg I was just writing up the detail

btw, got my book of abstracts working, with a little help of low-level \endinput :)

6:05 PM

@JosephWright I was sipping my wine on the train. :)

@egreg I wonder if you sip your wine when you ride the bike on German highways :)

@yo' I'll buy a flask with a small tube. And offer it to the police, in case they stop me.

@egreg you mean a camelbag?

@yo' Like what those Formula1 drivers have under their helmet.

@egreg possibly. Or like what some cyclists use

6:14 PM

@yo' Cyclists tire up, I don't. ;)

@egreg famous last words.

@yo' Well, after 650 km tours I'm a bit tired, I admit.

@egreg the rule I use is: stop after each 2 hours for at least 10 minutes. This means we'll make two stops on the night trip to Darmstadt

@yo' I have no fixed times for stopping. It depends on the road I'm doing.

@yo' And no, no night trip.

@egreg well, I stop at least that often. On very complicated journeys, I usually stop more. For instance with 3 bikes on the roof and a car almost overweight, I stop more often to check the bikes and to let the care rest a bit. It's also a bit more dangerous when the car is really heavy.

6:23 PM

@yo' I'm always wary of cars carrying bikes; once I found one of them at the side of the motorway, the driver was running to recover a couple of bikes that fell down.

@egreg I have seen this happening, but I (1) inspect all the mechanical parts of the mounts beforehand, and (2) fasten the bikes really well (much better than what the mounts' manual says) using the cargo belts

also, having 3 bikes, if one of the mounts failed, the other two would most likely prevent a catastrophe (seen this happening; one of the rods got lose; should the bikes not have been fastened together, the bike would have felt off the roof)

I get the warning pdfTeX warning (dest): name{ejercicio.1.27} has been referenced but does not exist, replaced by a fixed one What does that mean? The \label is present, and the \ref gives the correct output, but the main problem is that clicking on it takes me to the first page of the document (with hyperref), rather to the point to where \label is.

@Manuel have you applied \phantomsection if it's something non-standard? And have you tried removing the .aux and .out files?

@yo' No \phantomsection needed (apparently), since it's one environment more, I reference tons of them but that one (and other three) just don't want to work. And yes, the .aux and .out have been redone.

@Manuel well, then the only advice is: cut it down to a MWE and post a question on the site :-)

6:33 PM

@Johannes_B It turns out the mathematical support for the updated version of the font is far better. This may not actually take years after all.......

@yo' The question was more about what exactly was going on (in case it was a common weird problem), but if it's not common, I will have to try to create a mwe.

@Manuel it happens sometimes to me. I haven't found the exact issue though, I think that exploration in this way may be useful.

@yo' The thing is that I have ten \begin{ejercicio} \label{ex:whatever} .. \end{ejercicio} but just one in a 200 page document gives that weird warning (and, even while the \ref gives correct output, the hyperlink points to the first page of the document, the “fixed one”).

The same happens with one in a hundred with a different environment prop and the same with another exercise.

@Manuel yeah, I understand. I can only guess that it gets lost in some weird place like the output routine or whoknowswhat

@Manuel A 200 page document consisting of one file per chapter, neatly included in the main document with an \input per chapter. ;-)?

6:42 PM

@1010011010 Neatly included with \include per chapter :)

@Manuel Snap! Anyway, shouldn't it easy to figure out the issue then? I don't know how many chapters you have...

@AlanMunn Sometimes biber gets confused, so starting over applies to more than this message ;-) Are there any news? Did you have success?

@1010011010 Good to hear. May i ask which font you are using?

@Johannes_B typofonderie.com/fonts/le-monde-livre-family/#glyphs

Some glyphs are not in that glyph table, e.g. some swashes.

@1010011010 Not necessarily. The moment I remove something it starts to work. I'm working on it.

@Johannes_B I'm also taking suggestions on pairing it with a titling font...:-) right now I have myfonts.com/fonts/latinotype/arquitecta and myfonts.com/fonts/emigre/mr-eaves-sans-opentype... (only these 2 unfortunately... I was really looking for something experimental but I can't find anything matching)

7:08 PM

@1010011010 I somehow like Architekta. But i really don't have enough experience in choosing fonts. You could ask in some typography chatroom for suggestions, though.

@Johannes_B I had a preference for that too... Typophile.com it is, I suppose. Though they seem less active than they used to be:-(

@1010011010 I am not sure what i feel about le monde livre. The font seems to be very very stern.

@1010011010 graphic sans-serif fonts ... I'm not personally fond of these :)

@yo' Haha... I've gone through a trillion titling fonts. There's always something wrong with them. I guess it's because they have such large point sizes that every single detail that doesn't add up is immediately revealed...

Okey, I think this is a minimal working example.

\documentclass{book}

\usepackage{amsthm,thmtools}
\declaretheorem[numberwithin=chapter]{ejercicio}
\usepackage{hyperref}

\begin{filecontents*}{chap1.tex}
\chapter{chap}
\begin{ejercicio} \label{b}
Foo
\end{ejercicio}
\begin{ejercicio} \label{a}
	\begin{enumerate}
	  \item Bar
	\end{enumerate}
\end{ejercicio}
\end{filecontents*}

\begin{filecontents*}{chap2.tex}
\chapter{chap}
Baz \ref{a} y \ref{b}
\end{filecontents*}

\begin{document}
Filling the first page to show where the \verb|\ref{a}| points.

7:12 PM

@1010011010 that's the mistake of using too large titles :D

@yo' The title is always larger than the body text. :/

@Manuel I don't have LaTeX here :(

@yo' Argh

Okey, I will leave it here in case someone shows interest, in case it's not easy, I will post a question.

@1010011010 but it need not be much larger. For instance in amsart, the only thing that's larger is the article title, and only by mag1. Everything else is \normalsize.

7:13 PM

@yo' I'm not saying much larger... I'm saying larger. When you're demanding of the shape, the titling is the first to suffer (to me at least).

@Johannes_B There's no accounting for tastes I suppose.

@1010011010 well, the first to suffer is small print (footnotes, superscripts, etc.)

@yo' Never had issues with small print, though I generally work at unusually high screen resolutions and professional printers, so that may be why.

@1010011010 Funny that you mention that. Just tried the font on the website. Try tasty please.

@1010011010 your readers can if you get a lot of double subscripts :-)

but I agree. In general, I don't much experiement for title fonts. I like some standard ones: LMSS, Palatino, Antykwa Toruńska.

@Johannes_B All right. What am I looking for?

7:17 PM

@1010011010 I think the y and the t don't really match.

@Johannes_B Well the handout had one reference, so I just copied the bib entry into the file with filecontents and it worked fine. So I suspect biber just doesn't like the enormous bib file very much (it has almost 6000 entries).

and also, Antykwa Półtawskiego, if you want to look really old-school :)

@AlanMunn biber is known not to be the fastest tool in the world. :-)

@Johannes_B You think it should look like this? i.imgur.com/sBJUad8.png?1

That's \fontsize{64pt}{0pt}\selectfont Tast\fontsize{62.5pt}{0pt}\selectfont y..

@1010011010 Somehow. Though i am not a type designer.

7:20 PM

@Johannes_B I'll try to test more carefully when I get some time.

@yo' I don't intend to do this tex.stackexchange.com/a/144525/44160

@1010011010 Ah, you scaled down the font. I wouldn't have noticed. You could as well use raisbox to shift it a bit down.

@1010011010 lol

@Johannes_B At smaller point sizes it seems to even out.... i.imgur.com/9o2EZBD.png?1

It's just the tasty thing where it gives something weird.

@1010011010 where have you got the font? The only "Tasty" I see is some awful funny font

@1010011010 well the decision to make the bar of t lower is a bit controversial

7:24 PM

@1010011010 That's right. This is the problem with huge point sizes ;-) and so little text. But i think, even in the last example, that the period at the end of the sentence seems quite big. I noticed the same on the i when trying out the font.

@1010011010 tasty was really the first word i typed in a trilogy: Tasty fried liver :-p

@1010011010 why not, you should incorporate all my answers

@DavidCarlisle I'll make sure to include your picture of the man and the house.;-)

@1010011010 and the plane of course

@DavidCarlisle Almost forgot. Thanks. :-)

@Manuel Yes it is. But i just tried inputting the files, and the weird behaviour persists after multiple runs. Can you confirm that?

@Manuel There is a warning in the terminal output.

7:32 PM

@Johannes_B i.imgur.com/SnYrHBH.png?1 hmmmm

Time for some \catcode'y=\active\catcode'.=\active? (How to do back ticks in chat without activating inline code?)

@1010011010 The first is better

@yo' @1010011010 Mainly because in the adjusted version, the kerning is messed up.

@1010011010 enclose the code containing a backtick in two backticks: `````\catcode`\@=\active `````

@1010011010 ` ` or like this :-) much better ^^^^^

damn I can't get it right now

but this works: \catcode`\@=\active

7:34 PM

@yo' or use \ before the backtick: \catcode`\y=13

@DavidCarlisle escaping is one of the most annoying things, so I try to avoid it as much as possible :)

@Johannes_B @yo' Okay, I changed the kerning now a little as well: i.imgur.com/0oBXPNs.png?1

I guess the kerning between the period and the quotation marks could be better...

@1010011010 the font may profit from ty ligature

@yo' *strongly profit

@1010011010 that's caused by the large kerning of y.

7:38 PM

@1010011010 I like the first one better though, as the heavy period better matches the heavy glyphs. Also, the comma and the colon.

Is heavy the right word here?

@Johannes_B I'd say yes. You can also use "dark"

@yo' dark seems about right. Couldn't translate that to german though. :-)

@Johannes_B typography is like anatomy: you need to know zillions of notions to be able to express yourself.

2

@yo' Yep :-)

@Johannes_B :)

7:40 PM

Here's a home-made "ty" liga... i.imgur.com/nXsu17H.png?1

It's tough I suppose. I mean, you're right that the comma and the period are very thick compared to the body text.

@1010011010 I wouldn't literally connect them though, Just to make them fit very niicely. Also, it's not clear wat should happen with tyy then

@yo' A black hole would be created, sucking up most of the solar system.

@1010011010 :-)

I have used snapshot and bundledoc to archive my dissertation and all needed packages. How much farther should I extend my paranoia? Archive the whole TeXLive2014? Just the binaries and needed fonts?

@1010011010 a black hole? \fontsize{1000}{1000}\selectfont. ?

@AndrewCashner This is what AMS does, yeah. I recommend to give up, it's easier :-)

7:43 PM

@AndrewCashner If you want to be really sure, yes

@yo' @Johannes_B The whole distribution, you mean?

I suppose I should burn a liveCD of Debian Wheezy while I'm at it.

@AndrewCashner yes. But even that need not help since the binaries can stop working

@AndrewCashner this.

@AndrewCashner Packages can change everyday. Not looking at the kernel right now. Ladida whistle

or use plain with some standard fonts.

@yo' I recommend to give up, it's easier :-)

7:46 PM

@AndrewCashner binaries won't do you much good after a while, I have a dat tape somewhere with all the stuff when I left Manchester, but not sure how to read it or use sunos binaries for a motorolla architecture sun 3. Is the text of your thesis so fragile that minor font updates are a bad thing?

well, I'll have to go. I'll play some piano and then go to bed. Good night!

@yo' Enjoy!

@AndrewCashner I will. Though the e'' key is a bit weird recently. I think I'll have to service the piano at one moment.

@DavidCarlisle No, not at all, it's mostly the source code I need. Any future projects based on this will have to be submitted to publishers as Word docs as far as I can tell. And I'll have to rewrite most of it before then anyway.

@DavidCarlisle All I really need is the WEB sources and a Pascal compiler, right? :-)

@yo' That would be a killer easter egg. Somehow make "tyy" map to \fontsize{1000}{1000}\selectfont. :-D

7:49 PM

@yo' Is the key sticking or what?

@AndrewCashner You don't have an old IBM machine?

@yo' PDP10 would be good if not IBM

@AndrewCashner no, but sometimes when I press it, there's a strange feeling, a bit like if there was a crack in something that opens only if you press from the correct angle, or I dunno. But it's strange because e'' is certainly not amongst the most used keys

if I were able to describe the issue, I would consult them on the phone. But I'm not able to describe it :-/

@yo' Oh, it's in the key action, then. Is this an acoustic piano or electronic?

@AndrewCashner electronic.

yes, it's certainly something "mechanical"

it's one of these:

user image

7:52 PM

@DavidCarlisle Thinking truly long term (say 75 years), just to read the source code I would just need a text editor that can read UTF8 encoding, right?

@AndrewCashner print your thesis on a pro printer, notthing else will last 75 years.

@yo' I once repaired an acoustic piano key-action with one of those little metal twist-ties you get holding together a bag of bread at the grocery store. Doesn't look like that trick would work there.

@AndrewCashner no. I even can't reach the mechanics and as you can see from the picture, it's all hidden under the key

@AndrewCashner the text sources are probably readable forever but you'll probably need to move them around (or store in the cloud and let someone else move them) usb sticks and dvd's will presumably be as easy to read as my dat tape by then

anyways, I go. bye!

7:57 PM

@yo' Yes, it's on my shelf. Paper lasts a long time, I can say from having worked in old music archives, unless it gets lost, or burned, or waterlogged... The research should ideally form part of a conversation in the field, and any real value comes from that more than from the ink on paper.

@DavidCarlisle I have a box of cassette tapes (and one DAT, I think) of my performances in college. I kept meaning to digitize them, but never got around to it.

@DavidCarlisle So I should quit worrying about archiving all of TeX? After all I know you guys are never going to change anything, ever, that will break older documents, right? :)

@AndrewCashner we try

@DavidCarlisle I don't suppose you kept the typewriter (Golf?) you wrote your thesis on.

@AndrewCashner seriously though I'd just save your main document sources. For a journal or archive that needs to automatically reproduce a document out of thousands or millions of archived ones, it makes sense to try to freeze the entire stack, but for a single document, even if its a few hundred pages or so, it's likely to be more work to get a 10 year old texlive to run than it is to get a saved 10 year old document to run on a TL2025.

@AndrewCashner No, nor the typist I paid to type it.

@DavidCarlisle Thanks, that seems really sensible.

Paranoia relieved. Now I too will go play piano.

@Johannes_B Yeah, it stays there forever (it's not a thing of the first run). What do you mean? Of course there's a warnign (that's what I saw and started looking further).

8:11 PM

@Manuel I joined in later and haven't read your first posts. I'll take a look at it in a bit. thmtools is a candidate for messing up stuff.

@Johannes_B Nope, it can be just with amsthm. The problem seems to be the enumerate and the problem doesn't go away even if I use enumitem or scrbook.

9:06 PM

@JosephWright Arrrggg, writing chemical formulas without any packages is a huge pain the butt.

[\textsuperscript{n}Bu$_4$N]$_2$[(UO$_2$)$_2$($\mu$-O$_2$)(NO$_3$)$_2$($\mu$-Au(‌CN)$_2$)$_2$]}

@Canageek Indeed

@JosephWright Wait, ACS has mhchem in their template. Will RSC let me get away with that? It would be a lot easier to write with, even if it isn't as good as chemmacros.

@JosephWright Actually, they don't say ANYTHING about allowed packages, and say don't make 'extensive' use of \newcommand..... COULD I use chemmacros and chemnum?

@JosephWright Oh frack, I've been doing things wayyyy the hard way. RSC guidelines: "any packages you have used that are not included in a standard LaTeX installation" o.o So I can use anything in TexLive if that is a 'standard latex implementation'

@JosephWright rsc.org/Publishing/Journals/guidelines/AuthorGuidelines/…

@JosephWright (We are aiming at one of two journals, one is ACS and one is RSC)

@Canageek ACS guidelines are slightly out of date: I believe they now have e-TeX on the conversation server

@JosephWright Cool. I assume that siunitx and chemmacros are still verboten since they use LaTeX3? So mhchem and manual units?

@Canageek I'm not sure, really: I don't have access to the conversion server

9:18 PM

@JosephWright It would be nice if they posted those details. :S

@Canageek why do you assume that? it's just a latex2e package in the standard tex distribution, there isn't a latex3 at that level (of course their setup might be too old, but that's rather different)

@Canageek I can ask

@JosephWright That would be good to know. I'm more likely to send this to Dalton though, which is RSC.

@Canageek I can ask them too

Hi and welcome, what we see here is known as the template confusion. A template is supposed to help you, not give you additional trouble. The above, to be honest, is completely useless. — Johannes_B 2 mins ago

9:19 PM

@yo' It was \phantomsection in the end.

@JosephWright That would be amazing if you cuold

@Manuel Oy.

@Johannes_B I thought the point of a template was to make your document look exactly like what the person you are sending it to wants. I.e. I'll move my paper into the ACS or RSC template once I figure out which I'm sending it to, so that the manuscript matches what they want it to be.

@Canageek Well, what is a template? One say x, one believe y others state it's z. It's a mess, really. It is template confusion.

@Canageek The thing newcommers google for is something else than what you just stated. Or think of the templates in the L3 system.

We should call in @barbarabeeton to get yet another opinion :-)

@Johannes_B Aka the word is even more overused then Level in D&D

9:30 PM

@Canageek By now, mhchem also uses some L3. So if the tripwire is there, you are screwed ;-)

@Johannes_B -- barbara beeton thinks that a "template" is basically an "outline" into which contents can be poured. it is not a style (or a document class) or an example file. it should contain a \documentclass{...} line and (perhaps commented out suggestions for) \usepackages, things like (if a book) the commands \frontmatter, \mainmatter, and \backmatter, the structure for the top matter (which is document class specific), and probably a few empty \chapter or \section lines.

i think we need a glossary.

@yo' Sorry for having truncated the conversation about bikes, but my train was arriving and after a fast dinner there was a win to watch.

@barbarabeeton agreed :-)

@Johannes_B -- thank you! now, how do we get everyone else to understand this?

@Canageek See above ^^^^^ :-)

@barbarabeeton An easy to link blog entry could be a start. But who wants to read these days?

9:34 PM

@barbarabeeton Even if there was a glossary that wouldn't help when universities, journals, etc put out 'templates' that are chalk full of formatting things.

@barbarabeeton Markus Kohm as well as @clemens Niederberger have written something about the troubles of overloaded and old templates, both in german.

@barbarabeeton For example: lib.sfu.ca/help/writing/thesis/templates and rsc.org/Publishing/Journals/guidelines/AuthorGuidelines/…

For future reference LaTeX-Vorlagen and Minimale Vorlage vs. maximale Probleme

9:46 PM

Forgot one, oh dear: mychemistry.eu/2013/01/latex-templates

10:04 PM

@Johannes_B I like that one.

@Johannes_B Finding what packages one should be using is hard. For example, everyone uses mhchem, but if you use chemmacros it has support for coordination compounds. Also, I know a lot of people that still use subfigure instead of subfig.

user image

My library grew again today. ^^ It's really a beautifully-done book!

Stefan Kottwitz

@PaulGessler fancy book with strictly justified paragraphs

Huh. Would I cite this as a paper (It has a DOI and is in Science) or a webpage?

news.sciencemag.org/scientific-community/2015/04/…

@StefanKottwitz indeed; I'm still up in the air on my opinion of that. If the text flows well and doesn't seem obviously contrived/massaged to make the rectangular paragraphs, awesome. But if things seem unnatural I have a feeling I'll think it's too much. I'll have to read it to find out for sure. :-)

10:54 PM

@Werner could one adjust the lower margin down a bit? I'm not particular how it is done.

Is having closing associated with a nobreak sensible?

« first day (1641 days earlier) ← previous day next day → last day (3284 days later) »

Transcript for

Apr '1529

$TeX - LaTeX$ TeX, LaTeX and Friends

\begin{chat}

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

$TeX - LaTeX$

site design / logo © 2024 Stack Exchange Inc; legal

mobile