@clemens Wow, i laughed so loud :-)

@clemens Re mail coming tomorrow, too tired as well.

I am not sure if guests can vote, but if they can: Can anybody vote for any of the possibilities on the linked page/survey? I want to check if the feature is broken: golatex.de/…

@cfr Either a bug or I am completely lost. I tried this:

\documentclass[border=5pt]{standalone}
\usepackage{forest}

\begin{document}
    \begin{forest}
        for tree={draw}
        [center
          [first,anchor=east,calign=child edge,
            [ABCmmmm,anchor=west
            ]
            [DFR
            ]
          ]
        ]
    \end{forest}
\end{document}

According to the description of calign=child edge this should align the parent anchor of the parent (i.e., the east anchor of first) and the child anchor of the primary child (i.e., the west anchor of ABCmmmm), but this is not the case.

@GonzaloMedina So what's the difference between calign=child and calign=child edge? I would expect calign=child to behave as you say. I'm not sure calign-child edge is meant to. But neither of them seem to do either what I expect or what you expect or, actually, much of anything. The manual says that the effect of anchor is observable only when a node has a sibling. But I assumed that parent anchor and child anchor should have an effect even in a tree like the one I posted?

@cfr Well, it was my fault after all. Look at this:

\documentclass[border=5pt]{standalone}
\usepackage{forest}

\begin{document}
    \begin{forest}
        for tree={draw,parent anchor=east,child anchor=west}
        [center
          [first,anchor=west,calign=child edge,
            [ABCmmmm,anchor=east
            ]
            [DFR
            ]
          ]
        ]
    \end{forest}
\end{document}

It aligns the parent anchor=east of first,with the child anchor=west of ABCmmmm. Using calign=child in the same example, it aligns the node anchor=west of first with the node anchor=east of ABCmmmm. This is the behaviour described in the documentation.

But, as I understand it, parent and child anchors should not require a second child for calign=child edge to work?

@GonzaloMedina ^^

@DavidCarlisle OK, making a start on it

@JosephWright thanks, as you may have noticed I decided to sleep:-)

@JosephWright I left a comment on the doctest thing I had thought before I should split that up (as I kept forgetting to switch it to test the new code :-)

@Johannes_B :)

@Johannes_B I can't vote

@clemens I am an idiot. I closed the thread ages ago, but you cannot vote on a closed-thread survey ...

@DavidCarlisle I've done some integration work: will do a bit more later on

@JosephWright yes saw thanks once I got here

@Johannes_B Have you seen the Daft Punk medley?

@daleif Yes, that's nice.

@Johannes_B that got so many hits on YT so fast

@daleif 145m hits, and i heard of it today. Nobody left me a note.

@Johannes_B I think the Daft Punk one got more than 20m hits in 30 days.

@daleif That is very fast. I wonder how this happens? Heavy sharing on twitter and Facebook?

@Johannes_B probably, once it goes viral...

@JosephWright should ltluatex have a bytecode allocator (I note that's in luatexbase's TODO file...) Then we could perhaps do this tug.org/mailman/htdig/luatex/2010-February/001215.html

@DavidCarlisle I wasn't sure what the status on that area was now they've added \luafunction

@JosephWright "unclear" ? (not an uncommon status for things mentioned in the manual:-)

@DavidCarlisle Yes, or perhaps 'less useful'

@DavidCarlisle I better deal with user callbacks first :-)

Hello, a new question. What's the most expl3 approach to a similar to \tl_case:NnTF but with dimen ranges? I mean, could something like \dim_range_case:NnTF \l_tmpa_dim { {1cm}{2cm}{Between 1 and 2 cm} {3cm}{4cm+\parskip}{Between 3cm and 4cm+\string\parskip} } { <true> } { <false> }` be implemented?

@JosephWright hm so you can \dump the luafunctions table? (looking to avoid loading a file in everyjob)

@JosephWright yes, anyway easier to debug stack trace if it hasn't been byte compiled probably

@DavidCarlisle That I don't know

@DavidCarlisle The core LuaTeX people are, unsurprisingly, not that bothered about dumping stuff into formats

@DavidCarlisle Doesn't look like it, so a bytecode allocator still needed

@Manuel How are you defining a range here?

@Manuel Would be \dim_case_range:nnTF or similar (note n not N)

@JosephWright just had an email from Lufthansa, apparently I'm flying to Germany soon...

@JosephWright Instead of the usual case functions which have pairs of arguments (first the case, second the output in case it's true), this one should have triplets { <inf> } { <sup> } { <output in case the dim or skip is in that range> }. If there's any other syntax that comes to your mind, it's welcome.

@Manuel Give me five mins

@DavidCarlisle :-)

\input expl3-generic
\ExplSyntaxOn
\cs_new:Npn \dim_case_range:nnF #1#2
  {
    \tex_romannumeral:D \exp_args:Nf
      \__dim_case_range:nnTF { \dim_eval:n {#1} } {#2} { }
  }
\cs_new:Npn \__dim_case_range:nnTF #1#2#3#4
  {
    \__dim_case_range:nw
      {#1} #2
      {#1} {#1} { }
      \q_mark {#3} \q_mark {#4} \q_stop
  }
\cs_new:Npn \__dim_case_range:nw #1#2#3#4
  {
    \dim_compare:nNnTF {#1} < {#2}
      { \__dim_case_range:nw {#1} }
      {
        \dim_compare:nNnTF {#1} > {#3}
          { \__dim_case_range:nw {#1} }

@Manuel ^^^

I'm not sure quite how you want the boundary conditions: I've gone for inclusive

@Manuel If you think this is useful I can add to expl3

@Manuel I've defined only the nnF version: others are easy following the same pattern (just need the set up)

@JosephWright That was fast :)

@JosephWright I'm just playing a little bit with a section formatting, and needed this. In any case, that might be useful in the kernel (or some other approach you come with), because for integers, it's okey to check if they are equal to some particular cases, but dimens/skips are much less probable to have a particular length, so may be it's more practical to have ranges.

@Manuel Integer ranges might be handy too: I'll raise on LaTeX-L

Another doubt I had with the lengths is that you are using \dimexpr primitive, rather than manual parsing. So, how is it planned to add \widthof{whatever} or \hbox_measure:n {whatever} so they can be included in those dimen expressions?

@JosephWright probably too many options for inclusive exclusive, don't we have a generic boolean switch that you can use {#1 >min && #1<max}{..true}{..false} ?

@DavidCarlisle Not at present

@JosephWright odd. we did in my head..

@JosephWright oh but we do have \bool_if:nTF{\int_compare_p:n{#1 > min} && \int_compare_p:n{#1 < min}} (or at least we have the N version and could make an n?)

@JosephWright I don't understand how the hell does that work!? Where do you define \__dim_case_end:nw. Ahm, already defined in the kernel, sorry. I see you make use of that already written code, that's nice.

@DavidCarlisle Yes, have \bool_if:nTF but that's not quite as clear as a case statement

@DavidCarlisle You are probably right that it's hard to provide a generic version

@Johannes_B Thanks for spotting that typo (babel question)

@JosephWright Now you got my upvote as well :-)

@Manuel This currently not addressed (we had a version of calc ported to expl3 syntax but it was little more than a straight replacement of the names of control sequences)

@Manuel Problem is that there are places that expressions are used that are expandable, so you don't really want to force a non-expansion syntax

@Manuel At present I'm afraid there's an assumption you'll save the length and use that. Interface decisions of this type really not addressed yet. (This is also a rather specialised one)

@Manuel I'm sure @DavidCarlisle has a view :-)

@Manuel I just modified \dim_case:nnF for the requirements here :-)

@DavidCarlisle That deals with 'is the value inside a given range' but not @Manuel's question of 'a series of ranges akin to \dim_case:nnF

@JosephWright @ArthurReutenauer any thoughts on:

@JosephWright true

\cs_new:Npn \dim_if_in_range:nnnTF #1#2#3
  {
    \bool_if:nTF
      { \dim_compare_p:nNn {#1} > {#2} && \dim_compare_p:nNn {#1} < {#3} }
  }
\dim_if_in_range:nnnTF { 5pt } { 1pt } { 10pt } \TRUE \FALSE

@JosephWright why isn't \bool_if:nTF in interface3.pdf :-)

@DavidCarlisle It is

@DavidCarlisle Page 41

\bool_if_p:n {<boolean expression>}
\bool_if:nTF {<boolean expression>} {<true code>} {<false code>}

@JosephWright correction why isn't where I was looking (where the N one is)

@JosephWright write an expandable widthof in lua?

@DavidCarlisle Can't be done

@JosephWright boo

@DavidCarlisle Or at least as far as I understand it

@DavidCarlisle Typesetting is still left to TeX and I don't think there's access from the Lua side

@DavidCarlisle For example there's no tex.hbox

@JosephWright I don't suppose a lua system call to another instance of tex counts:(

@DavidCarlisle Probably not

@JosephWright but for pure text can't luaotfload basically tell you what the underlying OTF machinery would do? (In principle at least)

@DavidCarlisle Probably one for one of the LuaTeX experts: I'm thinking @StephanLehmke will know about this area

@DavidCarlisle Quite possibly

texdoc.net won a sponsorship box. :)

lalalalalalalalalalalalalalalalala

@PauloCereda we are not talking to you now you are rich with DANTE money

@DavidCarlisle I wish, I do not earn British Pounds. :)

Das Money!

:)

@PauloCereda Euro: very popular in the news at present

@DavidCarlisle Even in the land of the Queen? :)

@PauloCereda as heard on the radio as I came in this morning: "What do you give a duck that can't speak?"

.... pen to quack

@DavidCarlisle oooooooh

@PauloCereda something to do with the land of the queen's husband, I hear.

@DavidCarlisle :)

@DavidCarlisle Denmark?

@DavidCarlisle Sorry, missed your message two hours ago. It’s definitely weird, but luaotfload seems to be the most likely candidate to be at fault here.

Also I’m curious about the hyphenation “språ-ket” (in Swedish it would be språk-et), but I can live with that if it’s supposed to be correct :-)

@JosephWright wrong one. (It's best to have several countries from which to pick a royal title)

@DavidCarlisle He regards himself as Danish so ...

@JosephWright I thought he regarded himself as British by now

@DavidCarlisle I suspect he does what he d*@! well likes

@JosephWright Oh no, Joseph said a no-no word!

Ooh, rep cap

@JosephWright I'm at 40:-)

@DavidCarlisle The babel question was good (although I have a feeling it's been asked in some way before, perhaps a plain user)

@JosephWright Earlier than me? Appalling. ;-)

@egreg Don't think so

@egreg More notable for me, though, since I am only slowly making my way to the magic 150 days at the cap (currently on 120)

@JosephWright You're already a legend for us. :)

@egreg :-)

I suspect the CTAN team will not enjoy the TUG2015 week: I have a feeling I'll be updating expl3 again :-)

@JosephWright Found a spurious space? ;-)

@egreg Well I've got a second talk to write and I think some of the code I'll speak about has issues

@egreg Also need to add the großes Eszett business for @Johannes_B

@JosephWright You don't have to ;-)

@Johannes_B Will make a good demo

@JosephWright We could create a world league for the abolition of the Eszett.

@Johannes_B Might happen on the train on the way to the airport

@egreg Swiss people?

@JosephWright They know better!

@egreg Or perhaps they just can't spell :-)

In any case I can confirm that the bug reported by @UlrikeFischer at github.com/lualatex/luaotfload/issues/265 is solved with the current master branch of luaotfload, while the one in Karl Ove’s question is not.

@GonzaloMedina Lo he reportado a Señor Bezos.

@LaRiFaRi CC to @ChristianHupfer's friend Señor Biggles. :)

@PauloCereda what the duck? Never heard of Mr. Biggles...

just have reported some maybe-error in the titlesec-manual.

see comments below tex.stackexchange.com/a/134033

@PauloCereda uhhh, Biggles is from Monty... shame on me :-)

@LaRiFaRi Yes, a very famous sketch. :)

Well, something for my upcoming evening...

@PauloCereda I fear, evening is still some hours away in Switzerland. But thanks.

@LaRiFaRi ;)

@LaRiFaRi Will you join the league against the eszett?

@egreg never

@egreg the most complicated short cut I am proud to know is Alt+225

@LaRiFaRi I just hit Alt-s, getting ß, it's much easier. ;-)

@egreg hmmm not for me. My Vimperator kicks me to the addressfield when doing so. And all the Win-Stuff I can test right now, does ignore that completely

@LaRiFaRi Of course I have a real OS on my machines. :P And I use real text editors as well.

@egreg Out of curiosity, is Claudio Beccari an emacs user as well? I think I saw him using TeXShop.

@egreg hehehe. I am to bad in vim to fight for that.... and windows is not my private choice. But come on: Mac? Well, at least you have no "Alt" key which is one of the worst words for a Cologne-Guy like me.

@PauloCereda I don't think so. He adopted Mac OS only recently.

We prefer Kölsch

@egreg Oh. My presentation will use vim. :)

@LaRiFaRi Good beer! I have a friend living in Krefeld, his father used to bring us some cans of Alt when they came to Italy for holidays.

@PauloCereda Which presentation?

@ArthurReutenauer I would love to try but the makefile is driving me mad. I can't produce a tds-tree as the documentation doesn't compile.

@LaRiFaRi One for the GuIT meeting. :)

I need to practice my Italian. :)

ah ok

@LaRiFaRi it's about arara 4.0. :)

@LaRiFaRi What brand of Alt do you recommend?

@egreg I have to confess, I never tried it. The Cologne vs. Düsseldorf fight seems to have major influence on me...

@egreg as I sad, "Alt" is the absolute no-go for Kölsche Jungs...

@LaRiFaRi Go with Kölsch?

yes

Some Guys in Cologne replace their Alt-Key by a Kölsch-Key

@LaRiFaRi Good beer as well

@LaRiFaRi :)

@egreg puh, luckely we find something in common. :-) Yes, very good.

And for the brand question. I go for Mühlen-Kölsch at the moment... but that changes quickly.

@LaRiFaRi Unfortunately it's difficult to find either Kölsch or Alt here. :(

@egreg In Zürich they want 3 € for a bottle in the supermarket. Well... I can wait until next carneval....

@egreg If you go to Köln ones, take some cans with you to Italy. We have quite a lot of brands to taste... koelschfuehrer.de/koelner-bier/alle-koelsch-marken.html

@PauloCereda: Das böse Paulo

@egreg I doubt that Düsseldorf offers 87 beer brands :-)

@LaRiFaRi Just about 1000km.

@egreg then you better buy Pittermännchen than cans.

@PauloCereda Do you go to Italy for this?

@LaRiFaRi I wish I could. :(

@ChristianHupfer Quack! :)

@PauloCereda Do you know Meister Eckhardt and Quack

@PauloCereda when I have to learn some monty tonight, you could go for some cool music.

oh, its "Meister Eckhardt and Quackie"

almost...

@LaRiFaRi ooh

@PauloCereda not everyones taste...

I love it. Featuring 7 drummers... I have to love it...

@UlrikeFischer The way I did it was

1. Clone the repository from GitHub

2. Set TEXINPUTS to .../luaotfload/src///

3. Set LUAINPUTS to .../luaotfload/src///

Then run LuaLaTeX, and the bug you reported was indeed fixed.

(As you can see, I run TeX from the command line; if you’re using a specialised editor or an IDE I’m afraid I can’t help you.)

@ArthurReutenauer Don't worry, @UlrikeFischer is an expert at all TeX things :-)

@ArthurReutenauer I doubt that this replace really all of luaotfload. It wouldn't replace luaotfload-tool in miktex and imho during the build the files in the fontloader folder are renamed (in the installed luaotfload they all have a prefix "luaotfload").

@UlrikeFischer Yes, of course, it doesn’t replace what the Makefile is supposed to do, but the fact that the bug disappears when doing that is proof that something happened.

By contrast, it doesn’t necessarily prove that the newer bug isn’t fixed too, but it’s a strong hint.

By the way, if you have problems building the doc you can skip that part in the Makefile.

@ArthurReutenauer I tried to remove the doc building from the makefile, but as my "make skills" are quite rudimentary it didn't work. I would have to reread the docs first ;-).

I'm feeling better, with a glass of wine next to me. :)

@UlrikeFischer I'll try to reproduce that 2014/2015 pdf inclusion later, but I'll be offline for a few hours.

@UlrikeFischer We can have a look in Darmstadt. This definitely is an intriguing bug.

@ArthurReutenauer Will there be chocolate?

@PauloCereda If someone brings some :-)

@ArthurReutenauer oooh I will fix this bug

@DavidCarlisle If I do \begin{tabulary}{0.5\textwidth}{@{}L@{}} text\\ \end{tabulary} why does the table take up less than half of the \textwidth?

@DavidCarlisle will post a question on the site

Can somebody please check the output of the following code?

% arara:lualatex
\documentclass{article}
\usepackage{fontspec}
\usepackage{libertine}
\begin{document}
\textsc{tt}
\end{document}

@DavidCarlisle @egreg @JosephWright or any other xelatex/driver connaisseur, could you please check if the workaround to the \tikzmark problem with xelatex that I used in tex.stackexchange.com/a/255476/3954 (under the "Update" title) sounds OK and won't have any potential drawbacks?

@Johannes_B All fine with speed of golatex.de and latex-community.org ?

@Johannes_B Doubling RAM and sessions and database max connections did not help earlier, actually there were heavy attacks from some IPs to the web server which also gobbled all database sessions no matter how much resources

@StefanKottwitz I noticed a few fallout during the day concerning the MySql database. Both sites were completely unreachable during that time. But i am not really the right guy to judge speed, my machine needs about 10 seconds to load google. I should really trash the whole system and set it up again.

@StefanKottwitz Oh, attakcs do not sound good.

@Johannes_B No small caps here.

@Johannes_B Yes, various attacks, especially to blog sites, for example massive calls to their XML-RPC interface and trying automated logins via standard URLs and so on

@Johannes_B so they overloaded also the database via wordpress calls

@GonzaloMedina Same here.

@Johannes_B such things happen all the time but today it was simply a factor too much

@StefanKottwitz That part of the world is a black box to me. I don't understand a word.

@Johannes_B Once a server is on the internet, it is constantly under attack from many sides, that's just good to know :-)

@Johannes_B programs scan all and attack for example the CMS they find

@Johannes_B Imagine it would be in real life: if one goes out of the house, people with axes and knives and guns attack from different sides. That's actually the dark side a web user doesn't see.

@StefanKottwitz Oh no, Zombiekalypse

@GonzaloMedina Not enough kknowledge to judge what to blame. But it does work with XeLaTeX. Maybe @joseph knows who to address.

@Johannes_B Zombies is the right word. The Internet is full of Zombies. Do you know bot nets? Just one example, "Storm", is/was a bot net of up to 50 million computers. And it's not the biggest one. welivesecurity.com/2015/02/25/nine-bad-botnets-damage

@Johannes_B One of today's attackers was an IP of a German school (Fachhochschule), so that one was taken :-o

Many don't know that their computers are Zombies. If your computer is slow, check the processes which are running ;-)

@StefanKottwitz I bet there is something going on in the background. Any reading recommendations how i can find out?

@Johannes_B "ps axu", "top", "netstat -anp" I don't know reading to recommend

@StefanKottwitz htop is the first program i start myself. :-) I have to say, that i am running firefox with many tabs, often youtube and soundcloud as well.

Hey is there anyone around who uses MikTeX? I'm trying to help someone out, and I don't know much about Windows and MikTeX. We're trying to get things set up to use one bibliography file. We followed the steps in this answer to set up a local texmffolder. But running BibTeX leads to "I didn't find a database entry" error. The exact same bib file works when put in the same folder as the tex file, though.

@AdamLiter Certainly I know MiKTeX

@AdamLiter Screenshot of settings?

@JosephWright The TeXMaker settings?

@AdamLiter The MiKTeX ones

@JosephWright Yes, which tab do you want to see? The "Roots" tab?

@AdamLiter Yup

@AndrewCashner it's a documented feature, there is a question on site with a patch for people who don't want that (stretching columns seems such an odd thing to do I think I assumed no one would ever want that)

@JosephWright It says C:\localtexmf. If you check the "Show MikTeX-maintained root directories, then it shows C:\ProgramData\MikTeX\2.9, then C:\localtexmf, then C:\Program Files\MikTeX\2.9

Is that okay? It'll take me a few minutes if you want an actual screenshot to get it from her computer to mine.

@AdamLiter OK, so what's the path to the .bib file?

@JosephWright C:\localtexmf\bibtex\bib\test.bib

@AdamLiter You might need to refresh file name data base (RFND) or a similar name, but i am not sure.

@Johannes_B Do you know how to do that? I know basically nothing about Windows and MikTeX ... :p

@AdamLiter test is a very confusing name Can you rename the file to databasefile.bib and the entries to adamentry and friendnameentry? A very distinct name is very helpful.

@AdamLiter docs.miktex.org/manual/configuring.html

@AdamLiter OK, so you didn't refresh the FNDB after adding the file?

@JosephWright Presumably not, since I don't know what that is or how to do that ... :p

@AdamLiter It's in the general tab for the MiKTeX options

@AdamLiter It's in the instructions you linked to :-)

@AdamLiter "didn't find an entry " seems to indicate that bibtex did find a bib but that it didn't contain the entry, so probably some other bib was found.

@UlrikeFischer I was wondering about that too

@JosephWright @UlrikeFischer Me too, hence recommending a more detailed naming scheme for testing.

@JosephWright johannes is quite right imho: people who names their files "test" are bound to run into problems.

@UlrikeFischer I have a file permanently on my desktop called test.tex for throw-away things :-)

@JosephWright @UlrikeFischer Yes, that's one of the reasons I was confused. Now we're getting an error that says it couldn't find the database file database.bib ---line 4 of file test.aux : \bibdata{database : }, which seems to be this problem?

The .bib file is UTF-8, and I've tried both LF and CRLF line endings. Should it be something else?

@AdamLiter Put the bib in C:\localtexmf\bibtex\bib\ourbib\test.bib (there was once a problem with bib directly in bib, I don't have the time to check if it is still there), then run on the command line initexmf -u, then check with findtexmf databib.bib if it is found.

Moving database.bib to ....\bib\ourbib\database.bib and then running initexmf -u and findtexmf database.bib returns nothing.

initexmf --report does return C:\localtexmf.

Among other things.

@JosephWright Mine is called test-utf8.tex (I have also a test-ansinew.tex), it has thousands of line of old tests :-;

@AdamLiter Who added localtexmf as root? The user or the admin? In the second case try initexmf --admin -u

@UlrikeFischer Doing that returns:

I can try removing it from the root settings.

Okay, that seems to have worked. I'm still not entirely sure why it wasn't working before. Was it because the local texmf folder was only added as root but the user is presumably not running as root when it executes stuff? I know nothing about Windows ... :p

Thanks everyone for your help! Never let me near a Windows machine again ... :p

It does seem to be working now.

@AdamLiter Ah, you didn't say that

Hi

@JosephWright Yes, sorry. I honestly didn't even initially realize that there were two separate settings panels for root and non-root. Definitely my fault. :p

@AdamLiter Most problems are because users tends to think "better do something with the admin tools, there are more powerful" and then you get confusing rights. A user who wants to use a local root should use the user tools.

Just uploaded standalone v1.2 to CTAN.

@MartinScharrer CHANGES?

btw: bibtex/bib as folder should work too, I checked.

@JosephWright Mostly fixes which where in the trunk for quite a while.

@MartinScharrer Ah: just in time for TUG2015

Speaking of which: would it be a good idea to have a separate TUG2015 room set up say on Sunday?

@UlrikeFischer Yes, I should have realized that was the problem when I noticed that there were two different settings panels ...

But thanks again for the help, @UlrikeFischer @JosephWright @Johannes_B ! :)

@UlrikeFischer As you can see just a bit above, there were some attacks against golatex today. i'll remove the redundant messages and your final note, if it's ok with you.

Yeah, finally the images from Pluto ...

@Johannes_B Sure, go along.

@SeanAllred re your tweet about possible confusion: you're right about that although I think lokking or not looking at the terminal output/the lo file is not the issue: whoever looks there (or in the manual) would get the new name everyone else won't…

@UlrikeFischer I mislead you a bit on that pdf inclusion, sorry, just posted again to texlive list

@Johannes_B did you forget to post the code? ;) latex-community.org/forum/…

@clemens Yes. tired. Edited. Thanks :-)

@Johannes_B no worries :)

@clemens @SeanAllred Can any of you give me a pointer? This whole twitter stuff is confusing.

@Johannes_B Here you go twitter.com/vermiculus/status/621415838197878784 (more confusion: he responded to my private account ;)

@clemens And it is about the chemmacros stuff. I thought the template discussion came up once more.

@Johannes_B Ah. No

We lost the flag collection in the documentation of minitoc. :(

@DavidCarlisle This sounds much more probable.

@UlrikeFischer yes sorry trying to do to much in a rush this week:-)

@DavidCarlisle Thanks - I wasn't so much interested in stretching tables as in providing a fixed-width box in which to put variable content in columns that would read left to right instead of top to bottom. Now I am trying with \vtop{\hsize=0.4\textwidth ... } instead of a table.

@cfr There's "...how the children are positioned..." and "The child alignment methods refer to the primary and the secondary child,..." (emphasis mine) which suggest that more than one child are required for those options.

On the other side, the use of singular in the bit "calign stands for child alignment" suggests that it should work without siblings. In conclusion, I don't know whether it's a feature or a bug. I guess Saso (sorry for the typo in the name) has the answer.

@AndrewCashner why the tex primitives? \parbox[t]{.4\textwidth}{.....

@clemens That's precisely the point (also @Johannes_B)

If the user never looks at the console output, they'll never know the package was deprecated or what has succeeded it.

@DavidCarlisle I finally decided to start learning LaTeX3 syntax. What resources do you suggest for a beginner?

@SeanAllred but then the name doesn't matter anyway :)

@GonzaloMedina texdoc expl3, interface3 are good resources, and texdoc xtemplate helped me get a grasp on why we're separating things out anyway (and is frankly well-written)

@clemens The user still has to say \usepackage{new-name} at some point...

@GonzaloMedina Beginning with the distinction between LaTeX3 and expl3 :)

@SeanAllred Ah, thank you very much! Any additional recommendations?

@SeanAllred exactly

@GonzaloMedina l3styleguide is also important (I think that's what it's called)

@clemens I'm very confused! :) How are users supposed to find out the package has been deprecated?

Gotta run -- be back on in a little bit :)

@GonzaloMedina come to darmstadt and sit in when @JosephWright explains it to me, see first latex3 item here: github.com/latex3/svn-mirror/wiki/TUG2015-discussion-points

@SeanAllred those who read manuals or log files find out. Some here it in forums or whereever. Everyone else still uses \documenstyle

@SeanAllred @GonzaloMedina it is mirrors.ctan.org/macros/latex/contrib/l3kernel/l3styleguide.pdf

@GonzaloMedina interface3 is a sort pf reference manual but frank has a paper giving a higher level overview somwehere, I'll see if I can find it

@GonzaloMedina this is also interesting: mirrors.ctan.org/macros/latex/contrib/l3kernel/expl3.pdf And IIRC @JosephWright has a blog article series about the basics

@DavidCarlisle Can you please record Joseph and send me the video? :)

@DavidCarlisle That'd be great! Thank you.

@clemens Thanks, Clemens.

@GonzaloMedina there are these slides but I think there is a bigger tugboat article on this layer view somewhere, but this sort of sets the scene. latex-project.org/papers/2013-10-24-latex3.pdf

@DavidCarlisle I'll check my TUGboats and look up for it.

@GonzaloMedina I sent him SZ an email earlier asking if it is a bug or intended and recommending that the documentation clarify the intended behaviour if intended it is. (If you can follow that, congratulations.)

@DavidCarlisle Ah, this image on slide 4 I remember... I've seen part of this before.

@cfr I think I got the intended message :) That was the most sensible thing to do here, I think.

@DavidCarlisle @GonzaloMedina vimeo.com/67504395

@clemens Nice collection of videos. Thanks, once again.

@GonzaloMedina If this is the intended behaviour, it would be really nice if it weren't ;).

@GonzaloMedina you're welcome :)

@clemens Btw. thank you for continuing supporting LC ! We are a bit short of answerers to all the questions.

@clemens Today, half of the day, I was fighting attacks to the server, mostly to hosted blogs but they brought down the database server for LC and golatex too. I deactivated my blog and removed attack vectors on further sites (such as wordpress login urls) to get lower on the attack sessions

@clemens Now it's fine again and I learned a bit more about securing against automated overload and securing specifically wordpress