TeX, LaTeX and Friends - 2017-02-01 (page 1 of 2)

2:22 AM

Keeping up with biblatex changes is tough. My CV no longer compiles. Luckily I keep 4 years of previous versions of TeXLive around... Back to 2015 and all is well.

Joseph Wright

7:08 AM

@AlanMunn PLK made some non-compatible changes for good reasons, but it is something of a pain I agree. Hopefully with those done we will move toward a more stable situation.

Werner

7:48 AM

@Papiro What's with the abrupt exit?

@ChristianHupfer There must have been some sock-puppetry going on here, as no single user account could be responsible for the multiple upvotes at regular, consistent time-frames.

@samcarter That's true!

Joseph Wright

8:35 AM

@Werner Whilst I can't see the detail, I think I understand why action has been taken: I don't think a sock-puppet is involved

In very general terms: targetted voting is not allowed (either up- or downvoting). If a user is repeatedly picked up by the automated tools engaging in targetted voting, the Powers may intervene and invalidate all of their votes to the target.

Note that such action requires access to data unavailable to the site moderators

CarLaTeX

8:51 AM

@PauloCereda A colleague of mine wanted the .gif file of Prof. van Duck at work and he asked me "how have you done this?", I have introduced to him the wonderful world of LaTeX. He was really astonished. I'm doing proselitism!

Paulo Cereda

9:12 AM

@CarLaTeX ooooh ducks are taking over!

CarLaTeX

Yay!

May I accept this answer: tex.stackexchange.com/a/351519/101651 even if I can't completly test it?

@PauloCereda He tought it takes me ages to do it! Hahaha

Paulo Cereda

Oh Didier is now registered on TeX.sx!

CarLaTeX

@PauloCereda If you know him it means he is reliable, I'm going to accept the answer!

Papiro

Hi @Werner! There is no specific motivation for my decision beside the fact that it did not contribute to the group in those 1528 consecutive days of access. At this time, I think I should apologize to some people, specifically to you. Thanks to all ;)

CarLaTeX

@Papiro why? Please stay with us!

Paulo Cereda

9:21 AM

@CarLaTeX he's a well known person in the TeX world.

@Papiro being here is the best contribution you can give to the community, don't go!

CarLaTeX

@PauloCereda That's another reason to not let him go!

Paulo Cereda

@CarLaTeX also, we have ducks!

Christian Hupfer

@Werner: I don't think that it was sock-puppetry, otherwise more than just one vote for a particular post should have been removed in my case, since there are more 'identies/user accounts' to vote. This is not the case. One removed vote per answer or question, but 500 of them ;-)

Paulo Cereda

@ChristianHupfer the algorithm works on strange ways. Believe me, I've triggered it so many times that, by a simple observation, I managed to understand the underlying modus operandi. It's mostly about timestamps, very naïve rules based on how a normal user woud proceed when voting...

CarLaTeX

@PauloCereda Exactly! @Papiro where can you find so nice ducks elsewhere?

Joseph Wright

9:30 AM

@ChristianHupfer See my comment above

Christian Hupfer

@JosephWright I have seen that. I've posted my comment from the 'data' I have access to... just the vote count being visible on my answers panel

Paulo Cereda

/sigh Time to make my paper less "British"...

Joseph Wright

@PauloCereda Likely expects that votes are spread through the day and (being based on the SO pattern) are likely to be for posts from multiple users

Paulo Cereda

@JosephWright it's another way of seeing it. :)

David Carlisle

@PauloCereda s/u//g s/ise/ize

Paulo Cereda

9:31 AM

@DavidCarlisle That would be my first action. :)

Papiro

@CarLaTeX <3 it is a very long story :-) @PauloCereda I will be here as a gost following the contributions of all of you. Thanks all.

Joseph Wright

@ChristianHupfer Well you would know if what a sock-puppet!

Paulo Cereda

@Papiro oooh a ghost

I am afraid of no ghosts

GREAT THE SONG IS NOW PLAYING INSIDE MY HEAD

Christian Hupfer

@JosephWright How should I know that?

CarLaTeX

@Papiro I'm very sad :'(

Paulo Cereda

9:34 AM

@Papiro Take your time, pal. Remember that we will always be here, and your place will always be available.

Joseph Wright

@ChristianHupfer A sock-puppet is an account controlled by someone to vote for their 'main' account: people set up multiple accounts then use them to vote for one-another to gain rep, etc. Thus you would have to be in control of at least two accounts for there to be a sock-puppet situation.

@ChristianHupfer Note that this doesn't forbid having more than one account per se: it's the voting pattern that's important

Paulo Cereda

@JosephWright yes, that's the main gist of it.

emacs + evil makes that awful editor less awful.

For example, I can easily quit it! :)

Joseph Wright

@ChristianHupfer Key point is you can't accidentally get sock-puppet upvotes but you can get targetted upvotes without having anything to do with them

Christian Hupfer

@JosephWright Alright, if you restrict the sock-puppet to somebody whose only intention is setup 'fake' accounts in order to push his/her own reputation here (just speaking for the SE participation). I was thinking of somebody that uses more than one account to push reputation of other users than him/her

Paulo Cereda

CarLaTeX

9:43 AM

@PauloCereda Fabolous!

Paulo Cereda

@CarLaTeX The blue beetle is dead, if I recall the order. :)

Christian Hupfer

@PauloCereda Is it a Norwegian Blue Parrot named ~~Paul~~ Polly?;-)

Joseph Wright

@ChristianHupfer It's the control that makes it a 'sock'

Paulo Cereda

@ChristianHupfer ooh I forgot about him! Lovely bird, isn't it? :)

@ChristianHupfer wakey wakey!

CarLaTeX

@PauloCereda ... and Elvis is still alive...

Christian Hupfer

9:51 AM

@JosephWright Well, the targeted upvote was wrong and I wondered already recently, that's why I wrote the statement of the profile page, there have been occasional reversals but now everything has been reversed, invalidating normal votes, perhaps just one per day (if at all), from a time when the upvoter has not been casting a bunch of votes within one minute. So this part of reversal is doubtful, regardless who it is

Paulo Cereda

@CarLaTeX he's trapped inside an IO monad. :)

/Haskell humour... sigh

CarLaTeX

@PauloCereda I only know COBOL (and Sas, if we can call it a programming language) :):):)

Paulo Cereda

@CarLaTeX I know a little COBOL, but that's because I wrote a simple lexer for it. :)

@Guilherme: Ahoy! Long time no see, pal!

Joseph Wright

@ChristianHupfer Like I said earlier, once there's a pattern of targetted voting then the Powers reserve the right to invalidate all votes from the 'donor' to the target: one can't after all know what the motivation for any particular vote is. For this reason it's important that people don't engage in targetted voting.

@ChristianHupfer You can of course raise with the staff, but I think they'll say something along the lines 'The vote pattern was outside of the acceptable range: you are not disadvantaged (not 'involved') so there is no further action appropriate'

CarLaTeX

@PauloCereda I find the COBOL structure similar to LaTeX, because you have a preamble and the division with the actual code... :)

Christian Hupfer

10:00 AM

@JosephWright No, that would be wasting time. I am not begging for reversals of reversals.

Paulo Cereda

@CarLaTeX you probably heard of Grace Hopper?

Guilherme Z. Santos

@PauloCereda Oh captain my captain!

Paulo Cereda

@GuilhermeZ.Santos ooh I see what you did there. :)

Guilherme Z. Santos

I had an overflow of work and TeX.SX is already addictive without the chat!

But now I'm on vacation. Finally! :D

And yesterday night I started developing a small library for TikZ (kind of my dream: contribute to LaTeX)! =D

Papiro

@PauloCereda Have fun with your Thesis ;) Tex Exchange was a COD (Compulsive Obsessive Disorder) for me. It is time to stop :-) By the way, along all this time, I have one, and only one, account here, the account named Papiro.

Paulo Cereda

10:08 AM

@Papiro oh no, my thesis! I knew I was forgetting something important. :)

7

Papiro

@PauloCereda <3

Paulo Cereda

@Papiro <3

Guilherme Z. Santos

@Papiro If I may ask, why delete your account? Even without it you're still able to access the website. Is it like a gesture of leave, so things have closure?

CarLaTeX

@PauloCereda No, I didn't know her!

Paulo Cereda

@CarLaTeX If you know COBOL, it's nice to know about Hopper. :) Her life story is very impressive!

egreg

10:27 AM

@PauloCereda I'm sure you have \usepackage{fncychap,comicsans} in it, which would be a nice start.

Paulo Cereda

@egreg of course!

egreg

@PauloCereda Good boy!

Papiro

@GuilhermeZ.Santos It is a kind of "medical recommendation" :-) During 1528 days and nigths I have accessed tex exchange all time and from everywhere. Tex exchange is the first page of all my devices (laptop, pc, tablet, phone, brain). It is time to stop ;)

CarLaTeX

10:49 AM

@PauloCereda I've read on Wikipedia, very interesting!

Paulo Cereda

@CarLaTeX then watch this: youtube.com/watch?v=9eyFDBPk4Yw

Guilherme Z. Santos

11:06 AM

@Papiro I understand, it's absolutely addicting... I remember at first I only browsed sporadically for solutions, but after discovering TikZ it escalated very quickly. Now I find myself constantly opening the Unanswered/Newest/Featured tabs of TeX.SX. Tou suou

CarLaTeX

11:17 AM

@PauloCereda I'll watch it (I have also to work now and then)...

Paulo Cereda

@CarLaTeX oh no

CarLaTeX

@egreg May I post your fan club photo about comic sans?

Paulo Cereda

@CarLaTeX oooooh

@DavidCarlisle: ^^ do you have a fan club?

David Carlisle

@PauloCereda of course carlisleunited.co.uk/fans

Paulo Cereda

@DavidCarlisle oooooooooooh

CarLaTeX

11:24 AM

@Papiro don't delete your account, set an alarm that tells you to stop every 10 minutes!

Paulo Cereda

11:36 AM

@DavidCarlisle: I endorsed you in Logic. :)

egreg

@CarLaTeX I don't think I've ever seen it: I'm not registered to FB

Paulo Cereda

@egreg oh no

CarLaTeX

11:59 AM

@egreg Forgive me I can't resist the temptation to post it!

4

Paulo Cereda

@CarLaTeX LOLOLOLOLOLOLOL

CarLaTeX

@PauloCereda Translation: "Let's look at your thesis" Thesis: 101 reasons to love comic sans font

Paulo Cereda

@CarLaTeX Got it. My Italian is acceptable, I just cannot speak properly, but I am very good at reading/hearing. :)

CarLaTeX

@PauloCereda Hence LOLOLOLOLOLOLOLOL!

Paulo Cereda

@CarLaTeX pretty much. :)

@CarLaTeX: There are a couple of keynotes of @egreg in the River Valley website, and I can understand him perfectly. His Italian is very easy to follow. :)

CarLaTeX

12:07 PM

@PauloCereda Don't forget he is top user also on Italian Language SE!

Paulo Cereda

@CarLaTeX ooh I thought it was @DavidCarlisle :)

CarLaTeX

@PauloCereda :):):)

David Carlisle

@PauloCereda Você deve mostrar mais respeito

Paulo Cereda

@DavidCarlisle sorry. :(

David Carlisle

@PauloCereda no you're not! (how is your thesis by the way)

Paulo Cereda

12:22 PM

@DavidCarlisle :) Writing chapter 1. :)

Christian Hupfer

@PauloCereda Chapter 1 of ... 100dec? ;-)

Paulo Cereda

@ChristianHupfer I expect a long thesis. :)

Christian Hupfer

@PauloCereda I know, you \expand ed it ;-) :-P

David Carlisle

@PauloCereda @egreg was recommending 12bp font size for that very reason the other day

Christian Hupfer

@DavidCarlisle: How about changing bk10.clo to bk100.clo and \documentclass[100pt]{book} together with \usepackage[textheight=2ex]{geometry} to make the thesis really big (in number of pages)

@PauloCereda: It will be a great thesis, of course!

Paulo Cereda

12:32 PM

@ChristianHupfer <3

@DavidCarlisle <3

David Carlisle

@ChristianHupfer he's going to write in in Portuguese then I'll translate to English, so what could go wrong?

egreg

@DavidCarlisle 24bp would be even better, on 64bp baseline skip

yo'

\linespread{8.5}

Christian Hupfer

@DavidCarlisle ~~everything~~ nothing, of course ;-)

Paulo Cereda

@egreg ooh I have a style similar to Kant...

egreg

12:40 PM

@PauloCereda And you can make a fairly long thesis this way. With the maximum number of paragraphs, I get 600+ pages. Add indices, chapter breaks, whatnot and you easily get beyond 650.

Paulo Cereda

@egreg ooh

yo'

@PauloCereda however, not enough nihilist:

Paulo Cereda

@yo' ooh

yo'

\documentclass{book}
\begin{document}
\centerline{This thesis is intentionally left empty.}
\end{document}

3

@PauloCereda ^^

Paulo Cereda

@yo' LOL

yo'

12:45 PM

@PauloCereda Imagine this gaining you a PhD in philosophy :D

Paulo Cereda

@yo' ooh

Christian Hupfer

@yo': Too long

\null
\bye

Paulo Cereda

I want a Nintendo Switch!

/sob

David Carlisle

@PauloCereda after you finish your...

Paulo Cereda

@DavidCarlisle but I want it now, please!

Get M one too. :)

Joseph Wright

1:25 PM

@yo' \centerline is not a LaTeX command!

yo'

@JosephWright well, I know people keep saying one shan't use it...

Joseph Wright

1:46 PM

Question: Is it sensible in siunitx to worry about the difference between 123 and 123.? I'm looking at the v3 code (the basics are all working), and and am working out how to handle the current add-decimal-zero option. That then depends on how one views integers v decimals ...

It's a lot easier if one doesn't parse them as different ...

Christian Hupfer

@JosephWright It is only about parsing, not about typesetting, so 123. will still be typeset as 123. then (if one really needs this)?

Joseph Wright

@ChristianHupfer Well no, that's the point: siunitx parses numbers to then reconstruct them with the requested formatting. Of course, one can turn off the parser entirely but that means something like 12345. wouldn't appear as 12\,345.

David Carlisle

@JosephWright so what do you do with 12345.0 ?

Christian Hupfer

@JosephWright Well, in this case I wouldn't parse them as different

David Carlisle

@ChristianHupfer most programming languages would (in fact the trailing . is probably a syntax error in several)

Christian Hupfer

1:57 PM

@DavidCarlisle Well, many programming languages would treat it as a integer value, including C etc.

You're facing that you do too much LaTeX when you try to program in C with \int \main(...) and \printf` ;-)

yo'

@ChristianHupfer C? Really?

Christian Hupfer

@yo' Yes, quick and dirty

Joseph Wright

@DavidCarlisle Oh, that's fine as it has a decimal digit

@DavidCarlisle My current approach is to parse as 12345.0 => {12345}{0} but that means 12345 => {12345}{} and 12345. => {12345}{}. The question is whether I need to track the decimal marker explicitly.

I'm trying to make the code faster where I can, and one obvious thing to do is used a know 'rigid' format for internal number storage

yo'

@JosephWright Yes. 12345. and 12345 could be different to me.

Note that for me, 0.5=1/2.0 and 1/2 is not at all the same thing.

Joseph Wright

@yo' OK, guess I need to think again

I have a feeling I'll also need to retain the difference between 123 and +123

yo'

2:10 PM

@JosephWright quite possibly yes

qualitative properties of numbers can be important!

Joseph Wright

@yo' That's slightly easier as I already have a 'slot' for a sign

@yo' Yes, but what's harder to tell is whether they can both appear in the same context with different meanings: there are places I'd explicitly have +123 but at least in my work wouldn't have 123 and +123 at the same time with different meaning

yo'

@JosephWright not that I have a use case, but I do differentiate these things

Paulo Cereda

@DavidCarlisle: you got mail

yo'

E.g. in algebraic computer systems: 1+1=2 (integer) but 4^0.5=2.0 (float or alike)

Joseph Wright

@yo' Yes, I can see that but siunitx is a package for physical quantities

yo'

2:14 PM

@ChristianHupfer 2. and 2 are not the same thing in C:

#include<stdio.h>

int main(){
	printf("%f\n", (float)(5/2)); // 2.00000
	printf("%f\n", 5/2.); // 2.50000
	return 0;
}

Christian Hupfer

@yo' I did not claim that.

yo'

@ChristianHupfer then I don't understand this:

17 mins ago, by Christian Hupfer

@DavidCarlisle Well, many programming languages would treat it as a integer value, including C etc.

Obviously, C differentiates numbers with and wihout the trailing period

Christian Hupfer

@yo' I meant that 123 will be an integer and 123. will be interpreted as a floating point number, even without a trailing 0, i.e. C allows 123.

Joseph Wright

@yo' Indeed, but the question isn't whether in some abstract way they can be handled differently (clearly yes), but rather whether they need to be for some real applications in the context of the scope of siunitx. That I'm not sure about!

yo'

@ChristianHupfer ah ok, that's what I don't understand

@JosephWright note that siunitx is used way beyond the context of measuments are related stuff

I can imagine things like: That makes 8461648946894865 iterations of the algorithm, which will take 4862153.845 years

Joseph Wright

2:18 PM

@yo' Yes, I see that happens! People seem to miss the fact that the core competence is unit formatting: I spend ages on that and get basically zero feedback

yo'

@JosephWright I feel sorry :-(

Joseph Wright

@yo' One of the things I'm hoping to address is line breaking in large numbers, which this would fit in with nicely

@yo' Oh, it happens

yo'

I mean, I use it mostly for formatting plain numbers, recently money, and tables.

barbara beeton

@PauloCereda -- grace hopper -- my hero! however, i've always thought that the navy should never have changed the title of her position to "rear admiral"/ "commodore" was so much more suitable. it's well deserved that the ship named for her is fondly known as "amazing grace".

Joseph Wright

@yo' Still, I think the way that programming languages parse code fragments is likely out-of-scope: presumably one would use |2.| or \code{2.} or ... to make the point in your example

yo'

2:19 PM

Especially for tables, the (IMHO wrong) alignment of period-less numbers is troublemaking.

Christian Hupfer

@JosephWright Positive feedback: I switched all of my worksheets in Physics and Maths and beamer presentations to use siunitx if it is connected with units, since my old scheme was stupid

Joseph Wright

@ChristianHupfer :)

yo'

@JosephWright yeah, sure, but not anymore in the example with the iterations :-)

(btw, having 2 PCs in front of you with the chat open in both is rather confusing)

Joseph Wright

@yo' The tables part was done at the 'last minute' in v1 and I think that still shows in v2: one area to improve on

@yo' Indeed, but one wouldn't I think need to differentiate 8461648946894865 from 8461648946894865. (indeed, the latter would probably be 'normalised' to an integer)

yo'

@JosephWright honestly, far from "optimal", but still, accept my compliments as this thing must have been a nightmare!

Paulo Cereda

2:21 PM

@barbarabeeton I have several books about her and I am truly fascinated by her inspiring work. She's indeed true hero!

Joseph Wright

@yo' Example of what I've got wrong? Test cases are useful for new development!

yo'

give me a minute

the problem is exactly in the dotless numbers

\documentclass{article}

\usepackage{siunitx}

\begin{document}

\begin{tabular}{|S|}
{Hi}
\\
1234.56
\\
8 % I want 8(phantom period) -- aligned as the other numbers
\\
8. % I want 8.(nothing here)
\end{tabular}

\end{document}

@JosephWright ^^

Joseph Wright

@yo' That I know about: I'm unsure how best to handle it! Changing the code will change existing documents ...

@yo' Bottom line here is that the S column was never meant to handle numbers lacking a decimal part ...

yo'

@JosephWright well, preserve-periods=true would do ... (both I think)

Joseph Wright

@yo' table-phantom-period perhaps?

Paulo Cereda

2:26 PM

Ostrich algorithm

In computer science, the ostrich algorithm is a strategy of ignoring potential problems on the basis that they may be exceedingly rare. It is named for the ostrich effect which is defined as "to stick one's head in the sand and pretend there is no problem." It is used when it is more cost-effective to allow the problem to occur than to attempt its prevention. == Use with deadlocks == This approach may be used in dealing with deadlocks in concurrent programming if they are believed to be very rare and the cost of detection or prevention is high. For example if each PC deadlocks once per 10 years...

yo'

@JosephWright I know, but people still do that. I don't want to sound to philosophical, but to me it seems you can't get around the ways how people use your code...

@JosephWright perhaps

Joseph Wright

@yo' Still have my original question: why have 8. and not 8.0 in the input in the third data line

@yo' True and you are right that I should look at this: I'll log it for the next stable release (so v2.8)

yo'

@JosephWright not sure, will think about it. Well, my selfish opinion is: "Because I got it that way from the authors and I have no reason to change it nor the time to discuss it."

Joseph Wright

@yo' Yes, tricky

yo'

@JosephWright yeah, many things get tricky once it's not your own work you edit, and in an ideal world does not apply

Joseph Wright

2:30 PM

@yo' Extending the new code to store/track the decimal marker is of course the path that avoids any further issues at the cost of some speed

yo'

@JosephWright yeah, speed cost is one thing I'm aware of as the parsing gets more and more complicated. Maybe there could be a "draft" option that would e.g. almost convert S columns to c columns. I dunno, now I'm really just daydreaming....

Joseph Wright

@yo' github.com/josephwright/siunitx/issues/302

yo'

or an option to preparse a table into a file (like "siunitxexternalize"). The point is, I can imagine people having a 50-page table; then every millisecond is precious

@JosephWright code formatting seems wrong, other than that, looks good to me and has my point!

(btw, is there any way to star an issue?)

Joseph Wright

@yo' For tables, I'm planning to completely change how the code does alignment, based on some ideas from pgfplotstable: rather than have to use a property list by parts, I'm going to do a single formatting run and leave some 'markers' in the output of that to hook into for the tabular part

@yo' That will allow formatting based on a simple delimited approach, so should be considerably faster

I've also got an entirely new approach to the font business, again trying to avoid bottlenecks

I just need to finish v3!

yo'

@JosephWright sounds goot to me

@JosephWright oh versions!

(ah, there's "Subscribe to this thread")

anyway, gotta go visit a friend with a broken leg. And tomorrow, I'm of for cross-country skiing (@PauloCereda shall join me for that)! See you later!

Paulo Cereda

2:43 PM

@yo' yay! I wish a good recovery for your friend and a nice skiing for you, Tom!

David Carlisle

@JosephWright I'd expect 123. always to have the same value as 123.0 and them to be the same as 123 or not depending on conventions. (But do people ever write 123. ?)

Joseph Wright

@DavidCarlisle As @yo' has noted, people do all sorts of odd things

yo'

@DavidCarlisle People do all sorts of odd things.

Joseph Wright

@DavidCarlisle Problem is the v2 code does allow them to be treated in different ways and I therefore need to work out whether it's actually used

yo'

@JosephWright (now we time travel) :-)

David Carlisle

2:46 PM

@yo' you should have used dcolumn:-)

Joseph Wright

@DavidCarlisle I have had some examples from time to time where the output is expected to be 123., but that can be handled at a later stage: my question is about the internal data storage

@DavidCarlisle Yes, yes: it's logged now so will be fixed shortly!

yo'

@DavidCarlisle there's, IIRC, the problem with the headers (need to use \multicolumn)

David Carlisle

@yo' documented feature, not problem

yo'

@DavidCarlisle you and your features.

now I'm really off. Bye!

Guilherme Z. Santos

4:11 PM

Ok, so, hi everyone. I just finished writing a library to tikz, called patterns.extra, which offers more patterns and make it possible to customize them (change size/density and line width). I haven't written a manual just yet but I hope to do so in the next days. I just have some questions about uploading it to CTAN. It says that I should upload only one file, so how to I upload the manual afterwards? Are there classes for writing manuals?

And finally: are there other things one should worry about before uploading any library package to CTAN?

Christian Hupfer

@GuilhermeZ.Santos Upload a zip or .tar.gz file. You should have at least a README file and your package code

David Carlisle

@GuilhermeZ.Santos you need to zip up the directory layout as arranged for ctan and optionally also make a zip file in the installed "tds" layout. there are some notes on that in the ctan upload page

Paulo Cereda

@GuilhermeZ.Santos They are good people, if you miss something, they will contact you.

Christian Hupfer

@GuilhermeZ.Santos: Try ctanify if you're working under Linux

David Carlisle

@PauloCereda s/contact/track you down and harass you mercilessly until you provide the necessary details/

@ChristianHupfer or l3build on any platform

Christian Hupfer

4:14 PM

@DavidCarlisle Didn't use that so far...

David Carlisle

(sorry Petra, only joking ^^^ :-)

Paulo Cereda

@DavidCarlisle cry havoc and let slip the dogs of war

Guilherme Z. Santos

@ChristianHupfer yeah, I just re-read the upload text, first I've read "Note that some file names are blocked, like those ending in .zip, .tgz..."

David Carlisle

@ChristianHupfer @JosephWright will be disapointed in you.

Guilherme Z. Santos

Now I realized there's a not before ending

Christian Hupfer

4:16 PM

@DavidCarlisle Not more than he is disappointed of me usually, I think ;-)

Guilherme Z. Santos

@ChristianHupfer I'm on Windows, so that's a no... =/ but thank's either way.

How about the manuals class? Is that of our choosing or is there some guidelines to be follwed?

David Carlisle

@GuilhermeZ.Santos for a one-off package you don't really need a build tool just for packaging just package as described in ctan.org/help/upload-pkg (l3build would do that automatically and provide a way of running test documents etc)

Joseph Wright

@GuilhermeZ.Santos Up to you

David Carlisle

@GuilhermeZ.Santos anything. Most latex packages use ltxdoc if the manual is just the documented source file. If it is a separate document use whatever you want article, memoir, ....

Joseph Wright

@GuilhermeZ.Santos Unlike some other languages, there really are very little in the way of rules for CTAN uploads

Did people see ctan.org/markdown

Guilherme Z. Santos

4:21 PM

Alright, awesome. I'll start writing pronto. Thank's guys.

Joseph Wright

Apparently CTAN has the GitHub MarkDown implementation including code highlighting: I did not know that!

Paulo Cereda

@JosephWright /gasp

Christian Hupfer

@GuilhermeZ.Santos I use tcolorbox and its documentation library, which is tikz in the background` (I mean tcolorbox)

Paulo Cereda

@GuilhermeZ.Santos I use... /whispers memoir...

David Carlisle

it works:-)

Paulo Cereda

4:24 PM

@DavidCarlisle oooooooooh

egreg

4:39 PM

@JosephWright Try this ***is*** a test and see the italic continues.

Johannes_B

5:00 PM

@cfr vvvvv

%%
%% This is file `lstpatch.sty', generated manually.
%%
%% (w)(c) 2004 Carsten Heinz
%%
%% This file may be distributed under the terms of the LaTeX Project Public
%% License from CTAN archives in directory  macros/latex/base/lppl.txt.
%% Either version 1.0 or, at your option, any later version.
%%
%% Send comments and ideas on the package, error reports and additional
%% programming languages to <cheinz@gmx.de>.
%%
%% This patch file will remove the following bugs from the listings package.
%% Each item contains the bug finder with date of report and first bug fix

@cfr Garden template states:

\textbf{lstpatch.sty} -- this is a \LaTeX{} package required by this LaTeX template and is included as not all \TeX{} distributions have it installed by default. You     do not need to modify this file.

It was loaded in ecsthesis.cls but no comment why.

Joseph Wright

@DavidCarlisle Clever example of one of the TeX--XeT issues :)

Johannes_B

@cfr If you want to know more and haven't already read it: The history of the template

Paulo Cereda

Does anybody here use Isabelle/HOL?

Werner

5:31 PM

@ChristianHupfer Here's what I think one can do in this instance... Once one suspects voting fraud, send an email to team@stackexchange.com and inform them. They can look into it and take appropriate action (reference: Why didn't this obvious voting fraud get reversed?).

I recall seeing you mention something in your profile description about someone's voting behaviour; I doubt that has any effect.

Informing team@stackexchange.com would just avoid some future disappointment when the big red numbers appear. :-o

Joseph Wright

5:51 PM

@UlrikeFischer I see you've spotted that TeX--XeT is 'interesting'. As I've said on that question (and elsewhere), whilst it does work for some users, at a conceptual level it's problematic and I at least wish we could have the Omega model 'everywhere'

CarLaTeX

6:39 PM

Does anybody know which is the percentage of female users here on TeX.SE? The question arises from the comments here: tex.stackexchange.com/a/351462/101651. Thank you!

Werner

6:56 PM

@CarLaTeX That information - user's sex - is not available to the public. Not even in SEDE.

Paulo Cereda

@CarLaTeX I remember Vivi, Olga and... xport?

And Ulrike and Barbara, of course. :)

And cfr!

CarLaTeX

@PauloCereda I didn't know that Ulrike and cfr are women! I'm biased, too!

Stefan Kottwitz

@PauloCereda And folks using a duck as avatar image to hide their true gender.

barbara beeton

@CarLaTeX -- a quick look through the first 5 pages of users (180 total as shown on my screen) shows 7 definitely female persons (3.9%) and 3 possibles (cumulating to 5.6%). that's definitely not a huge proportion. but i see no reason to think that the statistics would be greatly changed by looking further. (tex, of course, is an equal opportunity endeavour. these statistics probably reflect the relative paucity of females in "stem" areas. too bad.)

CarLaTeX

@Werner I don't remember to have checked some "sex" flag when I signed up, maybe these info aren't available at all... but I saw a poll about female programmers on the main site blog...

@barbarabeeton Yes, too bad!

Johannes_B

7:22 PM

Asking a question and getting help ... does it matter if the OP or the helper is a man/woman/duck/horse? We all are human beings.

CarLaTeX

@Johannes_B Of course it doesn't matter, the strange thing is that we (I, too) think that all the user are men, till contrary evidence...

Emilio Pisanty

Anybody here know what to do with a mystery

> LaTeX Warning: No \author given.

Johannes_B

@CarLaTeX Personally, i just don't care :)

Emilio Pisanty

when it's obviously false?

Johannes_B

@EmilioPisanty Using \maketitle? Why?

Emilio Pisanty

7:27 PM

It happens on amsart but not on the standard article class

So I'm working on this template: github.com/episanty/TheYoungManTheStation

works fine under article

but if I switch it to amsart it complains about the \author thing

currently using fancyhdr, titlesec and titling packages, so maybe there's some conflict there

Werner

@EmilioPisanty And you're not using \author? Or are you?

Emilio Pisanty

@Werner I am, of course

but it doesn't display

It does display if I remove the titling package, but I still get the warning

Werner

@EmilioPisanty So you're mixing classes with packages that affect the titling... it's easier to know what's going on when you supply a minimal example.

Emilio Pisanty

@Werner The tex file is at github.com/episanty/TheYoungManTheStation/blob/master/…

I'm trying to minimalize the tex to a MWE but it's proving pretty finicky

Christian Hupfer

7:42 PM

@PauloCereda: Misusing tikzpeople for Physics: Evil ducks doing 'construction work' ;-)

barbara beeton

8:02 PM

@EmilioPisanty -- articles submitted for publication to the ams require an author. hence the warning.

Emilio Pisanty

@barbarabeeton well, I sort of do provide one

which is why it's odd

I think I've got it, though, it's the (i) the titling package being incompatible with amsart, and (ii) \@author not working under amsart

both of which, of course, are clearly documented in the corresponding package docs, and come with corresponding warnings about standard packages being incompatible with standard classes. Or almost.

barbara beeton

@EmilioPisanty -- anything that fiddles with an article's front matter using amsart is most likely incompatible. the amsart front matter is different from that of article. this is intentional, and is documented in a lot of places.

Emilio Pisanty

@barbarabeeton sorry, I just get grumpy on occasion with this sort of thing

I'm mostly trying to weigh the relative advantages of using article and amsart as a base

barbara beeton

@EmilioPisanty -- well, at ams we get grumpy when something is submitted that doesn't follow well-documented instructions. (not trying to be a pain, but when one is trying to keep a large production flow under control, one needs to maintain consistency in order to minimize costs. and different publishers do have different requirements. many authors just don't understand this.)

Emilio Pisanty

@barbarabeeton I don't intend to submit anything to ams

I'm trying to build something that I can use personally for arXiv submissions that I like the look of

I hope that's not an offensive proposition in the TeX world, but then I never know anymore

barbara beeton

8:10 PM

@EmilioPisanty -- amsart goes to a great deal of effort to maintain consistency with article, at least for the body, to allow an author who's not sure where s/he will submit to have as much flexibility as possible the front matter is a different kettle of fish, and it is documented to be different.

Emilio Pisanty

@barbarabeeton fair enough, but that leaves me with little information about which will impose less roadblocks later on

barbara beeton

@EmilioPisanty -- if it's for "general" use (even if just by yourself), i'd recommend using article as the base. (even if you like the look of amsart better.) i think arxiv accepts either.

Emilio Pisanty

@barbarabeeton is this because amsmath is more locked down?

I looked in its direction because article was proving weirdly limited

i.e. things like not handling short titles for running heads, or affiliations of any sort

barbara beeton

@EmilioPisanty -- the difference in styles has nothing to do with amsmath. amsmath is supposed to be reasonably neutral with respect to most document classes (although of course we haven't exhaustively tested it with everything out there). the resources available in article are those chosen by leslie lamport; in other words, what he needed. ams needs much more, hence the differences in how the top matter is defined.

Emilio Pisanty

@barbarabeeton OK, thanks.

barbara beeton

8:17 PM

@EmilioPisanty -- sorry to be discouraging. this is all rather an accident of history.

Emilio Pisanty

I don't mean to rail, but I hope you understand why someone might find this frustrating: the standard article class has "only the stuff some guy needed", and the class that tries to build a broader base intentionally locks itself down.

@barbarabeeton "accident of history" might just not cut it if we want to pretend that LaTeX is a flexible and powerful tool for the 21st century that people can actually pick up and use. These are really some serious structural issues, I think.

But then I'm a voice in the wild, so oh well.

barbara beeton

@EmilioPisanty -- honest, i'm sympathetic with this. ams originally sponsored the creation of ams-tex, which was the forerunner of amsmath. although it did meet the requirements of ams publications, ams-tex didn't have things like automatic numbering and cross-referencing, which authors really wanted (and need). and those things were present in latex. so amsmath was "extracted" from ams-tex, some syntax was changed to make it more latex-like, and the two diverged forever.

(cont'd) there is nothing like the "academie francaise" to keep things consistent in (la)tex. it's the wild west. just about every publisher has their own ideas of how to do things. and there's also the koma collection, also very different in many ways.

Emilio Pisanty

@barbarabeeton That's not really the problem. The problem is that there are some big, glaring holes in the existing infrastructure, the system doesn't provide anyone any incentive to fix those, and instead the community's response is to bury its head in the sand.

(well, it's also a problem, but not the only one)

David Carlisle

@JosephWright well it's not like it's hard to find one:-)

barbara beeton

@EmilioPisanty -- well, if you have suggestions for improving the ams document classes, please send them in. if they don't violate ams requirements, they'll get a review, and could be accepted, although i certainly can't promise how soon anything in that line might be acted on publicly. (there's a very long pending list ...)

Emilio Pisanty

8:30 PM

@barbarabeeton From what you mentioned, that looks orthogonal to what I'm saying. What's needed is a flexible standard article class. Something that will allow you to ::gasp:: have a title in sans serif, and also accept syntax like \author[J Doe and B Loe]{Jane Doe and Barry Loe} and actually know what do do with it w.r.t. e.g. fancyhdr.

David Carlisle

@EmilioPisanty it's not that it doesn't provide things it just doesn't provide high level markup for the front matter, which is pretty hard for such a general class anyway, basically if using article then, by design you either built up some frontmatter commands yourself, or (as documented in the latex book) you just treat \author as a container for any author information and put name, address and any other data into \author, separated by \\

Emilio Pisanty

@DavidCarlisle sure, but in article the \author command doesn't take an optional argument

David Carlisle

@EmilioPisanty changing article isn't really a possibility when you have 30 years worth, so millions of documents already using it

@EmilioPisanty so?

Emilio Pisanty

@DavidCarlisle precisely, hence the "serious structural problems".

David Carlisle

@EmilioPisanty er what?????

barbara beeton

8:34 PM

@EmilioPisanty -- while that would be nice, what ams needs to provide to independent indexers and aggregators isn't easily parsed from the "combined" author model. it's much more reliable to handle authors independently as done by amsart et al. it was all tested quite rigorously before being "frozen". you don't want the subscription prices of journals to rise just because of the problems involved in handling a less functional approach, do you?

Emilio Pisanty

@DavidCarlisle you say the basic infrastructure is indeed missing crucial ingredients, but it can't be changed because too many people use it. I say that that's a serious structural problem of the infrastructure and its interaction with the community that uses it. But that's obviously a subjective statement.

@DavidCarlisle so.... if I want an abbreviated head I need to roll my own commands? so more fragmentation?

David Carlisle

@EmilioPisanty \author and the error message you quoted come from the format not from article class:

\def\author#1{\gdef\@author{#1}}
\def\@author{\@latex@warning@no@line{No \noexpand\author given}}

Emilio Pisanty

@DavidCarlisle This is not about the error message quoted above.

@DavidCarlisle It's the fact that under article, the standard syntax \author[J Doe]{Jane Doe} will return ! LaTeX Error: Missing \begin{document}..

David Carlisle

@EmilioPisanty no but the point is that \author in the format is just the simplest thing possible, the intention is that classes design their frontmatter commands

@EmilioPisanty yes but why would anyone do that? you can't just take a random latex command and guess that there are some undocumented optional arguments.

Emilio Pisanty

@DavidCarlisle I'll have to disagree here - the syntax in question is quite standard. I understand your point of view but I don't think you understand mine.

David Carlisle

8:41 PM

@EmilioPisanty but the syntax in question is wrong. It could have been correct (had we had any more token memory available in 1993) but it just isn't. There is no universal rule that says you can use an optional argument as the first argument of a latex command.

Emilio Pisanty

@DavidCarlisle Sorry, but what you're saying only reinforces the case for serious structural problems. You're saying most classes in the wild use syntax that's wrong for one of their most basic commands?

David Carlisle

@EmilioPisanty why is that any different from \mbox[2in]{hello} or \begin[this][figure} or any other plausible but wrong guesses of command syntax?

@EmilioPisanty no, I didn't say that. I said that most classes define the syntax of their frontmatter commands. if you are using a class that defines \author to have an optional argument then obviously \author[J Doe]{Jane Doe} is correct input for that class.

Emilio Pisanty

@DavidCarlisle This is sort of missing the point a bit. Is there any functionality in the article class to specify which \@author gets fed to running heads like that in fancyhdr, but keeps the long author name for the front page?

David Carlisle

@EmilioPisanty no, the only defined heading types in article just use section heads.

Emilio Pisanty

@DavidCarlisle Then what I am saying is that article is underpowered, and that it's a shame that there isn't anything comparable that isn't underpowered.

I also find it frustrating that the community sees this and other similar issues and just shrugs, but that's a separate issue.

David Carlisle

8:46 PM

@EmilioPisanty "underpowered2 or "intentionally minimal" comes to the same thing.

Emilio Pisanty

@DavidCarlisle then I think it's a shame that there isn't any suitably flexible, suitably non-intentionally-minimal class for one of the basic use cases for the software.

David Carlisle

@EmilioPisanty no they don't shrug, they do as designed, leave article as a very basic starter class and produce classes with different features. koma or memoir as big generic classes, or journal classes such as amsart, revtex etc

Emilio Pisanty

@DavidCarlisle All of which are very tightly locked down, incompatible with a bunch of packages, and not even remotely interoperable.

I'm sure it's all nice and dandy for someone who's been working with the software for decades, but I would want a tool that I can use without spending all of my life to it.

Maybe I'm asking too much, though.

David Carlisle

@EmilioPisanty I think you are vastly over-rating the usefulness of such a class. For most basic unpublished documents something simple like article is fine (although some of the design decisions are a bit eccentric) and any journal using latex for publishing would just make their own class as now in any case.

Emilio Pisanty

@DavidCarlisle I think you're vastly underestimating the potential of what LaTeX could be.

David Carlisle

8:51 PM

@EmilioPisanty quite funny statement really considering.

Emilio Pisanty

aaaaaanyways

is there some relatively simple to define \shortauthor so that it will define itself?

i.e. so I can just type \author{Jane Doe} \shortauthor{J Doe}

David Carlisle

@EmilioPisanty yes just copy the line I put above that defines \author and write shortauthor where it has author

\def\shortauthor#1{\gdef\@shortauthor{#1}}
\def\@shortauthor{\@author}

@EmilioPisanty ^^

Emilio Pisanty

@DavidCarlisle hmmmm, yeah, I tried that

it works fine in the text

not so much in a fancyhdr header

David Carlisle

@EmilioPisanty well it just defines a macro called \@shortauthor nothing will happen unless you use that somewhere

Emilio Pisanty

@DavidCarlisle sure. but then \fancyhead[LE]{ \@shortauthor} kills it

(inside \makeatletter and \makeatother of course)

oh

or does that also need to go?

David Carlisle

8:58 PM

@EmilioPisanty don't see why it should, you can always make a MWE and ask a question on site

@EmilioPisanty does what need to go where?

Emilio Pisanty

@DavidCarlisle as in, it's not required to make \@shortauthor work (as opposed to \@author)

Transcript for

$TeX - LaTeX$ TeX, LaTeX and Friends