@Werner OK. Thank you.

@Werner: Now I proceed to the physical aspect of the problem here.

@JosephWright: I think "3 phases transformer" should be changed to "3 phase transformer". :-)

@CodeMocker Done

In

are you supposed to upvote the relevant comment? That's what I did. Though I don't understand how this is supposed to work. It says India 317, but there only 8 upvotes for India.

Do the upvotes get remove and added to a running total, or something like that?

@FaheemMitha 317 is probably some country code ;) I find that one ridiculous, I didn't upvote my country

@FaheemMitha 317 is from data.stackexchange.com

@FaheemMitha More precisely data.stackexchange.com/tex/query/99208/world-cup-of-reputation See also tex.blogoverflow.com/2011/10/tex-sx-users-around-the-world

@DavidCarlisle Thanks that does the trick

@TorbjørnT. I see. So this would be a separate count?

@JosephWright Thank you for your comment. Yes, I don't call \maketitle. I want something that will be called as close to the beginning of the document as possible, that's all.

What does \AtBeginDocument{\insertname} or \AtBeginDocument{\maketitle} so and where is it defined?

@FaheemMitha \AtBeginDocument is a LaTeX kernel command

It inserts code as part of \begin{document}

@JosephWright Oh,

After LaTeX has read the .aux file

@JosephWright So, that would work for me, perhaps. So, I don't need xpatch?

I'll try that. Thanks.

@FaheemMitha Not for this simple case

@JosephWright \AtBeginDocument{\insertname} works, thanks. Do you want to make that an answer?

@FaheemMitha OK

@JosephWright Not really. If \AtBeginDocument{\insertname} is used along with \maketitle, a page is generated before the one with the title.

@egreg That does sound suboptimal. A separate page with just the filename?

@egreg Ah yes, I'd forgotten about that. I'd probably just force the issue (as I do in achemso) by adding \maketitle to the hook as well

@JosephWright So, how would that look?

@FaheemMitha Do you plan to use \maketitle in your document?

@FaheemMitha Yes, the 317 comes from what users have written in their profile I think, while the votes are just those who happen to see that poll, and have voted.

Hello, guys! I'm thinking about whether to follow old advice of David to stop looking for solution which would auto-size delimiters like parenthesis and to switch to manual placement of \bigl, \bigr, and alike ones. But for this to work i need environment where i somehow could instantly see the changes that makes for example the edit from \bigl to \biggl, maybe even without need to move my eyes (or sight, don't know how to say it in English properly).

By the last i meant that after edit delimiter after \bigl (becoming \biggl) becomes bigger in code.

So, what environments are you using? Do they provide something alike?

Maybe auctex can do this magic?

@egreg I might. I'd like to leave myself the option.

@AndrewZabavnikov I use emacs for everything: it does try to show some constructs such as headings and superscripts at different sizes (so could do the same for \big..) , although mainly I find that an annoying distraction and would turn it off if I was doing any serious latex document editing.

@egreg I'd like something that would work either way. Sorry for the bother.

@DavidCarlisle And after your long experience with (La)TeX i can assume that you see in your mind the resulting document before compilation? Or not? :-)

@egreg Can't your two methods be combined? If \maketitle present use one version, otherwise use the other version? Or is this too ugly?

@FaheemMitha I'll be off-line for a couple of hours. Be patient.

@AndrewZabavnikov oh I never look at the pdf output, as long as there are no errors reported in the log, I'm happy:-)

@DavidCarlisle Wow :-0

@DavidCarlisle That's not a joke, right?

@AndrewZabavnikov a friend once complained to me that a latex file I'd sent had errors and we had several email exchanges where I said I got no errors when I ran it. It finally turned out that by "error" she didn't mean a reported tex error she meant that the output was wrong (in fact the document truncated and there was just a single page number 1 printed and no other text). I hadn't checked that part:-)

Okay. What about others here? How do you deal with the problem (if you see any)?

Maybe I should write it out explicitly.

I'm talking about line break between parens or braces or brackets.

@egreg No problem.

@AndrewZabavnikov you are talking about a previewer? I use evince, which auto-updates

there is also whizzytex, but i've had problems with that

@FaheemMitha With alt-tab, I guess?

There is a question here about previewers somewhere

@AndrewZabavnikov Just do C-c, C-c with auctex. Do you use emacs?

@AndrewZabavnikov yes more seriously it's a pain. If you are publishing database output automatically you either need to look harder for an automatic solution or just accept a slightly suboptimal size sometimes. If hand writing a real document though I don't think it's such a big issue, I'd just use some arbitrary size say \bigl on the first draft and then adjust if needed as part of the editing, may also have to adjust the position of the linebreaks and other things, so it's just part of the work.

@FaheemMitha Now I don't. I used for some time, but never for LaTeX.

@DavidCarlisle Maybe the thing is I just can't imagine what \bigl and others mean...

If they are for real just of some random increasing sizes, I just can't see any other solution than compile-view-compile-view-compile-view-......

And because it's for just brackets (and would be repeated for almost every one, i think), it's just unacceptable.

There was someone hating people answering the questions that should have been asked, right? :)

@AndrewZabavnikov well it's fairly easy to come up with a syntax that autosizes, but has syntax restrictions rather than expecting it to be able to cope with whatever weird tex primitives are there, so for your own document you can do that. for example \myleftright{a+b\\ frac{x}{y} } where \myleftright locally defines something like (untested:-) \def\myleftright{\setbox0\hbox{\let\\\relax$#1$}\def\\{\right.\originalcr\left.}\le‌​ft(\vphantom{\usebox0}#1vphantom{\usebox0}\right)}

btw, I wonder (or maybe I do not) how is \beameroriginal coded

@tohecz \show\beameroriginal?

@DavidCarlisle #1->\csname @orig\string #1\endcsname . so it seems that multiple \renewcommand on the same command would cause the system to crash :D

@tohecz Yup

@DavidCarlisle Seems to be perfect solution. Can you tell me, whether integrating with for example aligned would be possible?

@tohecz A lot of the beamer stuff is risky

@AndrewZabavnikov probably:-) If you can't get it to work mock up a complete example and ask on the main site (it's easier to discuss code there)

@JosephWright yep. IMHO there should be \renewcommandwithalt<> that would do exactly what I do in the answer. The problem is knowing the number of parameters etc., but if xpatch can do it, why not beamer, right? :p

@DavidCarlisle I'm asking because it seemed to me that you can see the problem with custom redefinition of `\\` instantly :-)

@AndrewZabavnikov note it gets harder depending how much you want to support. That sketch typesets the equation twice, once to measure and then normally , but setting things twice always has standard problems as for a full solution you need to make sure any \label etc don't get executed twice or counters incremented twice. there are fixes for that (ams alignments already set their body twice for example and have an \@ifmeasuring switch) but the details can be tiresome.

@DavidCarlisle btw, wouldn't something like \begingroup\let\global\relax\let\xdef\edef\let\gdef\def ... \endgroup be enough? :D

@tohecz \globaldefs?

@tohecz no:-)

@JosephWright ok, it's confirmed, I'm not a TeXpert :p

@tohecz firstly you can't use a local group if you want it to wrap just a couple of lines of an alignment (the ones that get the () ) but even if you do the whole alignment and could use a group, it isn't global definitions that are the problem so much as file writing to aux file and global counter increments and...)

@DavidCarlisle ok, I didn't think of file handling, but global counter increments should be fine with \let\global\relax, shouldn't they?

@JosephWright have you ever set that without breaking something:-) ?

!!/texdef -t latex globaldefs

Oh Psmith is sleeping :p @Pau

@tohecz but stuff breaks when you do that.

@DavidCarlisle ok, well, it was a completely blind shot, somehow my way of learning some new LaTeX tricks and treats

btw, the hotel room is finally quite heated, so it's a quick shower time for me. Laters!

@tohecz the way to learn this stuff is to put out a package like tabularx that sets things an arbitrary number of times and then just collect 20 years worth of bug reports about things that need to be added to the code that makes things safe in that context

@DavidCarlisle oh yeah :D

@DavidCarlisle What if \myleftright{a + \myleftright{b}}?

Doesn't \box0 equal \hbox{$b$} at the end?

@PauloCereda I think our problem may be related to Java bug 122429.

@DavidCarlisle Sorry for not trying it, but just speculating on code.

@AndrewZabavnikov you can make that work (although the command only really makes sense with `\` in its argument of course)

@AndrewZabavnikov yes but the point is that if you save it in a box you can use its height to put the same size brackets at the start and end

@AndrewZabavnikov oh sorry you mean grouping, you can make that work by taking care to stack existing values (read the comments in blkarray package for some thoughts on that:-)

@DavidCarlisle Yes, that's clear. But it seems that it is either groups or something about stack of heights.

:-)

Little bit late.

@AndrewZabavnikov and it's a case of diminishing returns, the one line of code off the top of my head above almost works with a bit of tweaking and if it does 90% of the cases in a real document, but you have to use explicit \bigxx for a couple of complicated nested ones, is it worth a couple of weekends coding and 20 years of bug reports, trying to make a robust version that can do those couple automatically?

@JosephWright Thanks.

Please, does such a command exist in LaTeX2e or should I make it myself?

\def\strip@dot#1.#2{#1}

@tohecz I don't suppose you count \filename@parse{1.2} and then \filename@base being 1 ?

@NicolaTalbot Oh. :( I found an answer in SO about using BufferedReader with a thread attached to it, so it hides the input as it's being typed. Quite clumsy.

@DavidCarlisle looks too complicated. I basically need to retrieve the integer part of a length :)

just divide by \p@ and use the result as a count

@DavidCarlisle what? divide? that needs some packages, doesn't it?

@tohecz \divide TeX primitive or etex dimexpr /

@DavidCarlisle ah this one :)

@PauloCereda Yes, I found a tutorial which works by doing a backspace after each character. I might add it as a less secure alternative if there's no console.

@DavidCarlisle I think it is worth making it freely available for everyone to hack on :-)

@AndrewZabavnikov They don't hack it, they mail you asking you to hack it, forever;-)

@DavidCarlisle So you don't reveal your email and just publish it with comments ;-)

But writing comments is already a big deal :-)

@DavidCarlisle \divide is the guy for me :)

@DavidCarlisle Is it true that there are scientific journals that do not accept materials with custom macros?

And with some non-standard packages, of course?

@AndrewZabavnikov probably (or at least hold the right to reject such electronic submissions) many of them don't use tex for printing they just accept latex for author convenience and then convert to some in house proprietary system, so if the author has used lots of clever tex tricks and the conversion fails, they don't want to debug that.

@tohecz And sadly nobody vote your last comment. :-)

@DavidCarlisle So what is then the way to do graphs (like in "k5 graph") in LaTeX? I guess no TikZ?

@CodeMocker oh this is you :)

@AndrewZabavnikov the way is to use tikzexternalize first, and then simply \includegraphics the external TikZ figure

@tohecz oh this is you too. :-)

@CodeMocker in all honesty, I thought you come from the golfing SE site: all the people there have quite crazy names

@AndrewZabavnikov well tikz or pstricks or just include it as an image (which may have been made with tikz of course) basically it depends on the journal, don't spend hours on some fancy \catcode changing syntax parser if you are going to submit it to a journal that needs to convert it to word for their in house review tracking system....

@tohecz :-)

@tohecz Never imagined that there is such a thing, still could have.

@DavidCarlisle Could you say that the off-top-of-your-head-macro is safe for submissions then?

@AndrewZabavnikov that's the way I do it. I as well put all my TikZ code in seperate files that I \input; this makes it even easier

@DavidCarlisle I, for one, would refuse an article containing the \catcode command, unless there was a very good reason for that.

:13373081it depends it boxes and measures stuff, that is hard for a convertor that doesn't have a typesetter to hand. Try it with latex2html or tex4ht or latexml convertors to html for example. It will most likely break them without specific customisation, and if a journal requires html conversion for the web or ebooks or whatever such customisation slows them down and costs them money. but a journal that is just going to typeset it with tex wouldn't even notice it was there, it would just work

@DavidCarlisle So, basically, you wouldn't design a package that would rely on \catcode changing?

@DavidCarlisle Really bad...

@AndrewZabavnikov Oh I have distributed lots of those (look at xmltex :-) but I don't necessarily expect everyone (in particular journals to use it) Look at it this way I was once told by someone who published math papers that they got an author submission, and removed the documentclass and added their in-house documentclass with the journal page size and system fonts, If the paper worked it went on to be processed, if it failed they spent no time..

.. looking at the macros, they just sent the pdf to india where it was rekeyed from scratch which seems bizare but that was a fixed (low) rate per page rather than paying a TeX consultant at US rates to figure out what went wrong.

@DavidCarlisle Sounds almost like a joke. Sad it is not.

@DavidCarlisle What can this converters do? Do they have at least TeX's mouth?

@AndrewZabavnikov latex2html is just a top level perl script that uses string replace on "knwon" latex syntax, so even the simplest macro breaks it unless you make a matching definition, tex4ht actually uses tex (it is a specialised dvi driver so it can cope with arbitrary macros but can be a beast to configure latexml uses perl but basically fully re-implements texs parser (well enough to cope with weird stuff like y xii.tex christmas carol) (tex4ht and latexml are the only ones that can do that) ..

@DavidCarlisle Anyway, there is a way out of this with using first TeX's mouth to offer us preprocessed .tex.

.. mathjax is javascript and like latex2html doesn't really cope automatically with any tex definitions, it needs matching javascript for eachsupported package

@AndrewZabavnikov no as keep coming back to the point that tex's execution is interwoven. You can't choose the size of your bracket (an expansion level \if) until you have set your trial box 9which requires the full tyesetting engine)

So a timelord has two hearts?!

@DavidCarlisle I'm proposing using full TeX to get extra information and then on later stages to use this information to modify original .tex. Can i rely on \vrule (strut) support?

@DavidCarlisle Sorry, meant to write "stomach".

Or, put it another way, what is the most intentional way of saying how delimiters' sizes should be chosen? Is it \vphantom after all? That would be sad because this way it requires many repetitions in some cases to get things done.

@PauloCereda Yes :)

@egreg Thank you for the detailed answer.

@PauloCereda doesn't everyone?

@DavidCarlisle :)

Hey guys, @Jake's cat ate a duck. /sob

@tohecz I think it is just someone trying to collect statistics about users per country. I don't really see the point, but it seems harmless. I'm surprised there are so many Indian TeX users. I don't know if any others personally.

There are of course those River Valley people in the South, but they are a business.

@DavidCarlisle I wonder why a journal would not use TeX internally, at least if they do scientific stuff.

@DavidCarlisle That's pretty weird. You mean retyped from scratch? Wouldn't the typesetter introduce mistakes? I dealt with one such person recently. He was pretty dumb.

@tohecz I hope you are feeling better.

Is there an easy way of saying in the preamble - please wrap the whole thing in a verbatim environment?

@FaheemMitha Very often this is circumvented by having it keyed in two or three times independently and doing an automated differential analysis. We once had a meeting with a company who would ship German loose-leaf collections to china by the cargo container to have them converted to SGML that way.

Interesting side fact: No need for the typing teams to know German when it's done that way.

@StephanLehmke That sounds rather inefficient. Plus it must be insanely boring for the people doing it, expecially if they don't even know the language they are typing in.

@FaheemMitha Uh, not wanting this to come over wrongly, but I assume there are a lot of jobs which are insanely boring (for instance processing mandarin sections).

@FaheemMitha it sounds inefficient but look at how many tex experts procrastinate here rather than fixing stuff, and just re-keying it at a pre-determined cost per page is a known cost and that's what people in suits like to hear.

@DavidCarlisle: youtube.com/watch?v=Gs56vyBd160

@StephanLehmke That's certainly true. I'd go as far as to say most jobs are insanely boring.

@PauloCereda after work, probably (I don't trust your youtube clips:-)

@DavidCarlisle Are you procrastinating? I thought you just enjoy answering questions. :-)

@DavidCarlisle safe for work, I promise. :)

@FaheemMitha We're are a lucky bunch, probably, for even being able to ask the question.

@PauloCereda but would it look like I was working?

@DavidCarlisle good question... research, perhaps? :)

@StephanLehmke Probably. That's why people do research, even though it is brutal. Because the alternatives are so horrifically braindead.

@DavidCarlisle Seems that for MathJax we cannot rely even on \vrules

@AndrewZabavnikov not surprising, \vrule is a tex primitive not expected in a latex document. the full list for mathjax is here docs.mathjax.org/en/latest/tex.html#supported-latex-commands

@DavidCarlisle Yeah, that's what i looked in for.

@DavidCarlisle Is it right that \vrule is the only primitive that could help me?

Anyway, is there a way to produce a file with TeX where only some of the original file's macros were expanded?

But that question is not critical at all - we can always get back to something external for further processing.

@AndrewZabavnikov well for mathjx you could use \mmlToken to directly code teh mathm in tex syntax and specify the size in differemt ways (but you can not measure things as you can in TeX unless you want to really hack the javascript)

@DavidCarlisle No, no, no. The intent is the following now. Get the original document (with my packages' macros), pass it through TeX once, get it to write the needed info (the biggest material inside each pair of bracket) into some helper file, then process the document yet again removing import of my package and replacing my macros with needed \vphantoms.

And I'd like to do it in renderer-independent way.

@AndrewZabavnikov good luck (that's a lot harder than it sounds:-)

@DavidCarlisle I agree. (And @AndrewZabavnikov, I recommend you to trust him, he's used auxiliary information many times in his packages ;) )

@DavidCarlisle Even hypothetically, there is a problem in user macros involving my macros...

@DavidCarlisle You recall I was talking about Lamport's interview, where he talked about something better than TeX's macro expansion. So, you think that he might have been wrong about finding something better?

@FaheemMitha Can you drop here a link to that interview?

@AndrewZabavnikov Sure. One sec, let me look.

@AndrewZabavnikov research.microsoft.com/en-us/um/people/lamport/pubs/…

Actually the second google hit for "Leslie Lamport interview".

The first interview is also interesting, but seems to have more to do with CS

@FaheemMitha Oops.

So, is there some way to declare an entire file verbatim? I haven't found anything yet.

@FaheemMitha \verbatiminput{file} ?

@DavidCarlisle Would this work inside the file itself?

@FaheemMitha if you are careful \verbatiminput{\jobname}\endinput but it would show that verbatim of course. What do you want to do really?

@DavidCarlisle I have a text file, and I want to latex it, but make it verbatim. So, instead of wrapping the text with a verbatim environment, I thought it should be possible to somehow say in the preamble - this entire file is verbatim.

I thought the em measure of length was a measure relative to the page size. Is that not the case?

@FaheemMitha Nope. Relative to the size of the current font.

@StephanLehmke Ok. But if the font is the same, then it reflects the same length?

@FaheemMitha Yep. For TeX's definition of "font".

@StephanLehmke Hmm, that doesn't seem to be the case here. Maybe I'm moving my margins around.

@FaheemMitha Where's "here"?

@FaheemMitha it's so much simpler to have another file that includes it with \verbatiminput otehrwise once you make everything verbatim (easy) you have to stop and finish up the file by hooking into etex's \everyeof hook and stop verbatim there etc

@StephanLehmke I'm changing the font size. That is probably doing it.

@FaheemMitha Exactly. Different fonts for TeX.

@DavidCarlisle OK, but having an extra file for every file i want to verbatimize this way is quite a lot of overhead.

@FaheemMitha you don't have to have another file, you can type the wrapper on the commandline if you wish

but why use tex at all:-)

@DavidCarlisle Oh. How would that look?

pdflatex "\documentclass{article}\usepackage{verbatim}\begin{document}\verbatiminput{file‌​}\end{document}"

possibly with backslashes doubled, depending on your commandline shell

@DavidCarlisle Oh, I see. Intereresting.

@FaheemMitha but every file would be called article.pdf as jobname would be set on first file found which is article here

@DavidCarlisle Yes, not ideal.

@DavidCarlisle I could write a rule that does this, renames stuff and bakes us a carrot cake. :)

@PauloCereda you could write a rule that sends the text file straight to the printer and skips the latex step

@DavidCarlisle lpr -r myfile

@PauloCereda too easy, you need to wrap it up in some yaml and java processing nonsense

@DavidCarlisle ooh we could also send via some network protocol with an XML interface. :)

meanwhile: the news @PauloCereda's been waiting for:

bbc.co.uk/sport/0/cricket/25874455

@DavidCarlisle WOOOHOOO

@DavidCarlisle: we need to finish this project before xor.

@PauloCereda Ta-da! datatooltk version 1.1.1 now works with arara. I've added --noconsole-action switch which can be one of: stdin (use System.in), gui (use a dialog box) or error (just give an error if there's no console). The datatooltk.yaml sets gui as the default, so even if arara is used without --verbose, it will still prompt for a password.

@NicolaTalbot Oh my! <3

Typesetting in the 80s. ^

You gotta love when your class gives you this error:

! Error: Missing \maketitle.


Type  H <return>  for immediate help.
 ...

l.48 \maketitle

@PauloCereda I am happy with my new car ;-)

@MarcoDaniel ooh! Which one?

@PauloCereda I will send you a picture ;-)

@PauloCereda Now I just need to find a way to stop vim from auto wrapping lines that start with %arara: :-)

@NicolaTalbot Oh no! How long is this line? :P

@PauloCereda More than 80 characters :-P

% arara: datatooltk: { output: products-sql.dbtex, sqluser: datatool, sqldb: datatooltk, sql: "SELECT * FROM products" }

(It's just a sample file. The sql statement could potentially be a lot longer.)

@tohecz:

\documentclass{article}

\usepackage{expl3,l3regex}

\makeatletter
\ExplSyntaxOn
\renewcommand\title[1]{%
\regex_match:nnTF { ^[a-zA-Z] } { #1 }
{ \def\@title{#1} }
{ \def\@title{ Dude, ~ start ~ your ~ title ~ with ~ a ~ letter! } }
}
\ExplSyntaxOff
\makeatother

\title{Ahoj}

\begin{document}

\maketitle

\end{document}

Is it too ugly?

@NicolaTalbot <3

Can't wait for the L3 tigers eat my heart.

@PauloCereda well, nice, but l3 is too fancy for scientists you know :-/

@tohecz <3 How are feeling today? :)

@PauloCereda I'm tempted to say that it's better, I can even hear the chat gong ;) just now it's not exactly painful, but I have a feeling like that someone superglued my left ear shut :(

anyways, gotta go for dinner, they close in 1 hour and I really don't want to miss it

@tohecz Have a nice meal, Tom. :)

@PauloCereda thanks. later!

@PauloCereda Way easier than my try... :)

@cgnieder My lack of TeX skills doesn't allow me to understand your answer. :)

I'm stoopid. :)

@PauloCereda I don't believe that :)

@cgnieder Pretty accurate. :)

Do you think it's worth me replying to the OP's comment to say that I guessed the OP was fairly new to LaTeX which was why I added links to help create a MWE?

@NicolaTalbot who knows? Maybe he/she gets it after a while that there's help behind that link. But maybe not...

@cgnieder Yes, I think I'll leave it. I don't want to sound overbearing. :-)

@NicolaTalbot I added a comment :)

@cgnieder Thanks :-)

Would anybody mind voicing their typesetting opinions real quick?

Do you think the one on the left looks too spaced out on the left margin? (As opposed to the right, where it might possibly be too cramped?)

@SeanAllred I like the one in the left. :)

@SeanAllred I like the left one better. I think there's too little margin in the right version

@Sean: no duck enthusiast? I'm disappointed. :(

@PauloCereda @cgnieder Thanks!

In German, Entexpert. :) @cgnieder: right? :)

@PauloCereda I thought about putting the faintest shadow of the Dolan you made for Pops as a TikZ background picture, but I feared someone might actually see it XD

@PauloCereda It would be “Entenexperte”. Or “Entenenthusiast” :)

@SeanAllred ooh it's dirty, let's do it! :)

@cgnieder ooh it's way better than I expected. :)

Ich bin ein Entenexperte!

@PauloCereda oh Lord…… :)

@PauloCereda :)

@SeanAllred the one on the left is better. Top-align the experience groups with the first items, I would say

Yay left power! Oh that sounds so wrong.

@PauloCereda XD XD XD

@tohecz As in have the tops of the description and the data aligned? I think I can actually do that… I'm pretty proud of this class file.

@SeanAllred yes, I'm not a fan of vertical centering of table cells (this is basically a table structure), one gets lots in what belongs to what ;)

@tohecz I didn't realize how much of a table structure it was… I have them set up as minipages. Introducing a \vfill definitely did not do what I expected (but did exactly what should)

Do you think it would be better to typeset it as a tabular?

And last two things that come to my mind, more suggestions than anything stronger: 1) I'm not sure how the sf, rm and sc fonts go together if it's zoomed in or printed, on the zoomed-out one it looks a bit "overdesigned", but as I say, that's a personal feeling and only on the priview

@SeanAllred just use [t] with the minipages ;)

The printed version looks fine to my untrained eye :)

2) isn't bullet a bit boring? what about \rule[0.1ex]{1ex}{1ex} or something else, maybe just try couple variants, you might find something else more pleasant or get back to bullets ;)

@SeanAllred ok :)

@tohecz I did think about that—off the top of your head, can that be changed with enumitem (I'd imagine)

Bullets are boring :)

@SeanAllred yes, something like \setlist[itemize,1]{label={\rule[0.1ex]{1ex}{1ex}}}

@SeanAllred without enumitem: \renewcommand\labelitemi{...} (IIRC)

@cgnieder yep, should be just right

@cgnieder btw: Hi Clemens (I hope I spell your name correctly ;) )

@tohecz @cgnieder Thanks!!

@tohecz Hello :) Yes, that's right

@cgnieder Thanks for the answer. Currently we're not sure what the policy will be. The Editor-in-charge would prefer titles of pure text and no math at all, I would prefer only enforcing the 1st thing to be a real word.

I think that along with [utf8]{inputenc} support, one would use the support of the good old commands like \' too, and then it would be fine

!!/texdef -t latex \'

@tohecz ooh wait that's my cue!

@PauloCereda well, quite :)

!!/help

@PauloCereda Psmith, the TeX bot: Good night! I'm Psmith, the friendly TeX bot - the p in my name is silent, as in pshrimp. I'm here as a companion to our fellow users in the typographic land. As you probably noticed, I always reply under Paulo's account, but do not despair, I say, my replies are always preceded by my own name. Enjoy your stay at TeX.sx! If you need any help, just ask our chat residents. Cheerio!

@tohecz: ready. :)

!!/texdef -t latex \v

Psmith, the TeX bot, in fixed font mode: Here's the output from texdef:
**Psmith, the TeX bot:** I'm terribly sorry, but I think `texdef` didn't like your input. Could you take a look at the parameters?

Psmith, the TeX bot, in fixed font mode: Here's the output from texdef:
**Psmith, the TeX bot:** I'm terribly sorry, but I think `texdef` didn't like your input. Could you take a look at the parameters?

@tohecz I haven't edited the answer, yet, but testing for \UTFviii@two@octets would be doable:

\makeatletter
\newcommand\ifletter[1]{%
  \ifcat#1a\@empty
    \expandafter\@firstoftwo
  \else
    \expandafter\@secondoftwo
  \fi
}

\newcommand\ifaccented[1]{%
  \expandafter\ifx\expandafter\UTFviii@two@octets#1\unskip
    \expandafter\@firstoftwo
  \else
    \expandafter\@secondoftwo
  \fi
}

\newcommand\iffirstletter[1]{\@iffirstletter#1\@empty\@empty\q@stop}
\def\@iffirstletter#1#2#3\q@stop{%
  \ifaccented{#1}
    {\@firstoftwo}% starts with \UTFviii@two@octets
    {\ifletter{#1}}%
}
\makeatother

@tohecz: sorry, apparently this construct is invalid for Psmith (is valid for texdef). :(

Psmith, the TeX bot, in fixed font mode: Here's the output from texdef:

\v:
macro:->\OT1-cmd \v \OT1\v

!!/texdef -t latex "\'"

Psmith, the TeX bot, in fixed font mode: Here's the output from texdef:

\\:
macro:->\x@protect \\\protect \\


\\ :
\long macro:->\let \reserved@e \relax \let \reserved@f \relax \@ifstar {\let \reserved@e \vadjust \let \reserved@f \nobreak \@xnewline }\@xnewline

Ouch.

@PauloCereda lol :)

@tohecz Too much cricket. :)

anyways, I work in T1, how do I teach that to texdef?

\iffirstletter{\'exx}{true}{false} still fails...

!!/texdef -t latex -p [T1]inputenc \v

Psmith, the TeX bot, in fixed font mode: Here's the output from texdef:

:
Compile error: LaTeX Error: File `T1.def' not found.

@PauloCereda ouch squared :)

@tohecz Hey stop trying to kill my bot! <3

> \v=macro:
->\T1-cmd \v \T1\v .

@cgnieder ^^ so a test for this crazily-named macro \T1-cmd should be enough

@PauloCereda I don't kill him, I only make him crazy step by step :)

@tohecz LOL

@cgn there's a leakage of the 2nd byte of the unicode char

@tohecz that should work

@tohecz yes I just realized that

@cgnieder one needs to gobble that part somehow, with something along my \ONETWO#1#2\ENDONETWO -> #1

@tohecz shouldn't be too complicated but I'm too tired right now. I can't concentrate any more...

@cgnieder well, not a big deal, I'll look into it later myself :)

!!/fortune

@PauloCereda Psmith, the TeX bot: Here is your fortune: Confucius say you have heart as big as Texas.

!!/fortune

@PauloCereda Psmith, the TeX bot: Here is your fortune: Confucius say: if you think we are going to sum up your whole life on this little bit of paper, you are crazy.

Damn you Confucius!

!!/fortune

@PauloCereda Psmith, the TeX bot: Here is your fortune: Someone has googled you recently.

o.O

@PauloCereda It wasn't me, I swear!

@cgnieder <3

@PauloCereda /reluctantly raises hand But it wasn't google, it was linkedin!!!