« first day (3063 days earlier)      last day (35 days later) » 

12:47 AM
Please have a look to this spam-question, worth to be deleted:
-2
A: What is the best package out there to typeset proof trees?

userFor help to my client, I'm looking for a freelance expert for formal logic, who knows how to formulate proofs, and to write in latex. A complete mastery of English, a doctorate, and an example of an article published with proofs of logic. 600 NIS per hour for the appropriate candidate Please do ...

 
1:21 AM
@DavidCarlisle So what you are saying is, I can't get there from here, until some indeterminate point in the future? That's an OK answer, as it just means that if I want to develop a class that imports pre-specified images, to work across engines, I just have to make sure the pre-specified images have suitable names in their pdf and eps forms.
 
 
6 hours later…
6:58 AM
Is it a matter of taste or does it make the same to write ^T or ^\intercal?
 
7:38 AM
@manooooh Oh, why use \intercal? Of course you are free to chose your own syntax, but the T is standard here among mathematicians, I would say.
 
@StevenB.Segletes partly or just that I didn't want to look at the details at that time:-) You may find Heiko's grffile package does what you need already, but we do hope to get some general support for filenames with spaces into latex, but it will not make the cut for the first texlive 2019 release
 
8:01 AM
@mickep hi mick! I found that symbol because it is what "The Comprehensive LaTeX Symbol List" says, but in 2009 a guy gave and advice of the use of ^T, so I do not know why I have said that LOL! So I guess I will use ^T :). THanks!!
 
@manooooh intercal is just weird, we have an answer about that
9
A: What is the meaning of the name of the \intercal symbol?

David CarlisleThe earliest mention I can find is the SGML standard (1986) which lists intcal in its additional mathematical symbols list under "binary and large operators" the definition being (in full) <!ENTITY intcal SDATA "[intcal]"--/intercal B: intercal--> which means the the entity &intcal; could be u...

 
8:26 AM
@DavidCarlisle oh I did not know. Thanks for the link <3
 
8:37 AM
@DavidCarlisle I like the "which looks like T" part ;-)
 
Could someone look at my question: tex.stackexchange.com/questions/480551/… ? I’m afraid it disappeared in the “stream” of questions
 
@MaxR One way might be to disable the section head in the contents and then to add \section*{Inhaltsverzeichnis} just before the multicols start.
 
9:27 AM
@manooooh Never found anything better. Using \intercal doesn't add to clarity.
 
9:44 AM
Oki doki
 
10:28 AM
@DavidCarlisle No relation to the INTERCAL programming language, of course.
@manooooh I have sometimes used ^{\mathsf{T}} myself. You might want to make a command so that you can easily change it later. (@egreg)
 
10:49 AM
@HaraldHanche-Olsen That's the right way to do it.
 
11:06 AM
has anyone played with the sansmathfonts package? It seems nicer than the default font when used in beamer. But, I get a lot of
LaTeX Font Warning: Font shape `OMS/cmsssy/b/n' in size <15.93346> not available
(Font) size <17.28> substituted on input line 320.
probably since they are specified descretly in the fd files. As far as I can see it iis using PFB files, so how can we get this to just optically scale?
I'm trying to combine it with our standard sans serif font, which does not have math support. Just replacing a-zA-Z0-9 and the operators font with this text font, seems to be working so far.
 
@daleif fix-cm package ought to allow arbitrary sizes I would guess
 
@DavidCarlisle I'm already loading it, does not seem to do anything.
 
@daleif before the class?
@daleif oh hang on cmsssy isn't a standard shape is it,..
 
@DavidCarlisle no idea, I've never understood the way fonts works
 
@daleif found it you need..
 
11:17 AM
This is just for a special case. We are not allowed to redistribute the main font I use here (just changed the family default). Default will probably just be using the default beamer font setup
 
untested but:
\DeclareFontFamily{OMS}{cmsssy}{\skewchar\font48 }
\DeclareFontShape{OMS}{cmsssy}{m}{n}{%
      <5-6>cmsssy5%
      <6-7>cmsssy6%
      <7-8>cmsssy7%
      <8-9>cmsssy8%
      <9-10>cmsssy9%
      <10->cmsssy10%
      }{}
\DeclareFontShape{OMS}{cmsssy}{b}{n}{%
      <5-6>cmsbsy5%
      <6-7>cmsbsy6%
      <7-8>cmsbsy7%
      <8-9>cmsbsy8%
      <9-10>cmsbsy9%
      <10->cmsbsy10%
      }{}
@daleif before loading the samsmathfont package
 
Hmm, that does not work as it gives some mk errors. I'd probably need to take a look in the fd files that comes with the package. It this the idea for all of them?
At least this does not give an error:
\DeclareFontFamily{OMS}{cmsssy}{\skewchar\font48 }
\DeclareFontShape{OMS}{cmsssy}{m}{n}{%
<5-10>gen*cmsssy%
<10->cmsssy10%
}{}
\DeclareFontShape{OMS}{cmsssy}{b}{n}{%
<5-10>gen*cmssbsy
<10->cmssbsy10%
}{}
 
@daleif I didn't check which fonts are available (just edited the lines from thee existing fd) but <5-10>gen*cmsssy% ought to be wrong as if you ask for a 5.5pt font it will (I think) try to load cmssy5.5.tfm, which is why I split it up into ranges (cf the entries in fix-cm.sty for the standard cm shapes. But what works, works:-)
 
@DavidCarlisle ahh, that makes sense. did not know the syntax (and my test file isn't that large.
Isn't there also an extra syntax one can add to these to scale the font? alpha, beta etc almost matches the font size for the text
I've been looking through a few packages to see how the scaled option works. But not many provides it.
Just noticed one last issue that I does not have a grip of either. As mentioned I'm substituting all the letters and numbers to come from a different font. How do we get \bm to with with that? and \mathbf for that matter.
I've been using
\SetSymbolFont{letters}{bold}{T1}{ffu}{b}{n}
where ffu is my private font
 
11:34 AM
@daleif (texdoc fontenc has the syntax 4.4
@daleif for bm you need to set up the bold math version (then bm uses the fonts from there in teh normal math version
@daleif cf \SetSymbolFont{letters} {bold}{OML}{cmm} {b}{it} oh that's exact;y what you did, bm should work then (if loaded after that, and you haave enough slots left)
 
This is a big mess, but if we can make it work...
BTW: texdoc fontenc just gives me the encguide, which does not explain much
One thing I like about sansmathfonts is that it fixes \mathrm in beamer (probably just sets it to \mathsf. I;ve never understod why this was not the default.
 
@daleif do what I meant not what I wrote. texdoc fntguide
 
11:52 AM
@marmot Oh I just saw, there are DokumENTENeinstellungen ;-)
 
@DavidCarlisle that made more sense. Now lunch
 
@JosephWright Thanks for the link to the duck shop in Amsterdam. The following duck is actually from this shop - it reminded me of the former incarnation of @PauloCereda as bee duck :)
 
12:22 PM
@DavidCarlisle the only thing wrong with this one was a type. The second font is called cmssbsy, then it works just fine
 
1:20 PM
@DavidCarlisle I'm wondering, would it make sense to email the maintainer of sansmathfonts (I can see that beamer already loads sansmathaccent of the same person) and have those .fs files changed. It does not really seem necessary to use discrete sizes when we are dealing with non-mf fonts.
 
2:15 PM
@mickep Thank you! Your suggestion worked beautifully.
 
2:34 PM
@daleif sounds sensible to me, we only keep the cm ones at discrete sizes for very basic article class compat from the beginning of time it doesn't make sense for new fonts at all really, even if metafont has to generate a few bitmaps disk space isn't quite as it was in 1985.... and for scalable fonts it makes no sense at all
 
@samcarter ooh a bee duck
 
@DavidCarlisle another question about the broken tabu/colortbl tex.stackexchange.com/questions/480673/… ;-(
 
3:16 PM
@UlrikeFischer I did promise Frank I'd get a fix for that into tl2019, not had alot of time in recent weeks...
 
@DavidCarlisle don't worry. As you saw I'm telling people not to use tabu ...
 
@UlrikeFischer that works if it's people if it is the back end typesetting part of generators such as doxygen, harder for end users to avoid...
 
3:35 PM
@DavidCarlisle ew doxygen
 
@DavidCarlisle well yes, but it is a bit naive from such a tool to use a package like that. Since 6 or 7 years it has the "don't use, unmaintained" sign, and now every one is surprised that it broke.
 
3:53 PM
@UlrikeFischer no one reads documentation, I once thought my subtle abstract at the top of the blkarray doc would discourage people from using it, but apparently not.
 
@DavidCarlisle ooh an abstract
 
4:06 PM
@DavidCarlisle ;-). tagpdf says at the begin "This package is not meant for normal document production.". And the last feedback I got was from a guy who made a 200 page document with it.
 
@UlrikeFischer oh my
 
@UlrikeFischer You could argue that most documents are shorter than 200 pages and thus a 200 page document is not a normal document :)
 
4:21 PM
@DavidCarlisle just read on the winedt mailing list: "If the filename contains unicode characters [other than those that can be translated in your default code page] than this is a known limitation with SumatraPDF and Synctex." Did anyone ever tested how synctex reacts to non-ascii files names?
 
4:54 PM
@DavidCarlisle I've sent an email to the apparent author. After playing with this all day, I don't think the approach of using sansmathfonts and then attempting to substitute the letters and 0-9 is going to work (note this has to work with pdflatex). I'm using \DeclareMathSymbol to redefine a-zA-Z0-9, but this does not affect \mathbf, \bm, \mathit, etc. Setting the letters font, does not work as it is a text font I'm using, so no greek.
Perhaps figuring out how to do a virtual font might be easier to work with.
 
5:05 PM
What is the most modern and coolest way to create indexes with LuaLaTeX?
 
5:18 PM
@daleif what are you trying to do? Do you want to make it compatible with another sans serif font?
 
5:31 PM
New palindrome for me:
user image
4
 
6:27 PM
@Kurt You will run out of palindromes very soon unless you find a way to increase your score by 1 mod 5...
 
@marmot Yes I know :-( I hope to see the 399993 ;-)
 
6:42 PM
@Kurt Yes, 6 digits are better than 5.
 
@marmot Ops, sorry, should be 39993, but you are right, then @DavidCarlisle can not mention the wrong comma ;-)
 
@Kurt 9 digits will be even better because then there will be an even number of commas.
 
 
2 hours later…
8:50 PM
@Kurt @marmot you should follow @CarLaTeX's example
yesterday, by David Carlisle
user image
4
 
@DavidCarlisle LOL The image would be better if the badges numbers with 5, 55, 555 :-)
 
 
1 hour later…
10:00 PM
Hmm, xindy does not seem to like something like \index{A~B}. It dies reading entries in .ind file.
 
10:47 PM
@Kurt Thou shalt not lead @DavidCarlisle into temptation -- and 33, 333, 3333 would look even better.
3
 

« first day (3063 days earlier)      last day (35 days later) »