@Kurt Yes, we you can. … It must, of course, be outside of `<MiKTeX-portable_folder>, but on the same drive.

@Speravir I never tried it and a friend asked me about that. He has a little bit problems with it but does not know the reason ... I guess it is a failed installation ...

@Kurt Ask him. By any chance this was already a question in TeX.SE. The greatest problem could be, that the compilation must be done through the command prompt, one gets from the taskbar entry of MiKTeX portable, or otherwise with batch file(s) where the system path is set.

@Speravir Ah, I see. I will talk with him. Thanks!

@PauloCereda Very good. Now if someone can add bookmarksnumbered to the hyperref options, one could actually use the PGF manual without loosing one’s mind.

@moderators There seems to be some spam going on with questions about trips to China.

@PauloCereda Here's something for you. :)

enrico2013$ arara nofoot
  __ _ _ __ __ _ _ __ __ _
 / _` | '__/ _` | '__/ _` |
| (_| | | | (_| | | | (_| |
 \__,_|_|  \__,_|_|  \__,_|

Running PDFLaTeX... SUCCESS

@MarcvanDongen Do you have a link?

HELP!!!!!!!!!!!!!!!!

\def\multcent#1{\halign {\centerline{##}\cr #1}}
\def\multleft#1{\hbox to size{\vbox {\halign {\lft{##}\cr #1}}\hfill}\par}
\def\multright#1{\hbox to size{\vbox {\halign {\rt{##}\cr #1}}\hfill}\par}
\def\Mdot{\hbox{$\dot M$}}
\def\degmark{^\circ}
\def\boxit#1{\vbox{\hrule\hbox{\vrule\kern3pt\vbox{\kern3pt
          #1 \kern3pt}\kern3pt\vrule}\hrule}}
\font\big=cmr10 scaled\magstep2
\font\bigbf=cmbx10 scaled\magstep2
\font\bigit=cmti10 scaled\magstep2
%
%       Simple units
%
\def\cm{{\rm\thinspace cm}}

@tohecz latex or plain?

@DavidCarlisle latex

@tohecz In the new order when latex3 goes \let\def\@undefined then everything will be fine:-)

@DavidCarlisle /me not likes that/

@tohecz well you have a choice you either stop people using inappropriate primitives, or you accept that they will use them. You have a certain amount of pain whichever you choose.

@DavidCarlisle they'll use \newcommand. And if you forbid \hbox, they'll use \mbox. Usage of \def is not the pain here

@tohecz The idea is that you provide defined interfaces which can rely on the fact that the primitives are always correctly named for internal use

@JosephWright I understand the idea

I mean, you cannot stop people from doing bad things just by removing some interface. They'll be always doing nasty things

@tohecz no \def isn't so bad actually but \font\bigbf=cmbx10 scaled\magstep2 does rather blow a hole in LateX's font mechanisms.

@tohecz Yes, but normal users often don't know what they should avoid: \if seems to be a common one

@JosephWright you have a point

and now apologize me, I have to tame this fucked mutilated beast

@JosephWright and for xgalley anything like \vskip or \hbox` or \special would break the document in entertaining ways.

@DavidCarlisle Yes: another more general case is \box

@DavidCarlisle That's easy to correct: \def\bigbf{\normalfont\Large\bfseries}

@egreg \let\bigbf\relax, why should someone need to modify the class defaults?

@tohecz It depends where it's used.

and please, whoever writes a manual, start with "Do you know \label and \ref?"

@egreg yes but if the default definition of \font was \def\font{Please use plain Tex if you want to use commands from the TeXBook} It would be less likely to occur and need correcting.

@DavidCarlisle And \def\newfont#1{Are you \emph{really} sure?}

@DavidCarlisle You'd want \protected here

@JosephWright :-)

Of course the aim of XML was similarly to clean up the uncontrolled "Tag Soup" of HTML by having more defined interfaces and draconian error handling for people who didn't obey, the result was that (real) people just didn't switch: they like tag soup and sloppy markup.

@DavidCarlisle I think at least for coding we are making some progress. At the document level, the idea is more to move the problem so users don't have to make a concious decision

@JosephWright yes but the number of people who code is sort of manageable, I think it remains to be seen how actual real people react. We tried removing \rm in 2e I suspect very few classes don't load the package to re-instate it, and I would guess that it won't take long for a latex3 addon that re-enables all the tex primitives to appear. we may anyway need something like that unless we are planning to re-code tikz and pstricks and any other existing code that people use. We'll see:-)

@StefanKottwitz Somebody removed the message a few minutes after I reported about it.

@MarcvanDongen You can flag the message as spam, it's better than reporting it here.

@DavidCarlisle We've been much clearer than in 2e that some things are simply 'not supported'. We can't enforce this, but I'd hope if we continue to build on our own tools and support for 'high-quality' code using them then we should get a reasonable number of 'well-trained' users. We also can be very clear that anything that breaks, for example, the :D-rule, is unsupported

Ooh I just got upgraded to IE10 I wonder if anything works.. (Not that I'm using IE at all normally but still...)

@DavidCarlisle what do you speak about? ;P

@PauloCereda Oh noes! The TikZ manual has acquired another 300 pages! :-) LOL

@DavidCarlisle @JosephWright : An interesting debate on a topic which has been worrying me ever since I joined up on TeX.SE. Most people seem to advocate solutions which I would call "excessively complex/intestinal for the end user". Shouldné we have a movement that rewards simple solutions that avoid Plain TeX and what I call "Divining the Entrails"?

Don't be offended, David, but xii.tex, while a work of great art and skill, makes my hair curl.

@DavidCarlisle Chrome or Firefox? Or, of course, emacs?

@Brent.Longborough It's simpler to live in a world where tikz has no manual

@DavidCarlisle ROTFL It's still smaller than the Bible

@Brent.Longborough well yes but same could be said about that:-)

@DavidCarlisle OK, then, it's still smaller than IBM's z/Architecture Principles of Operation

@Brent.Longborough Chrome is not my favourite browser at present (as you may have seen on g+)

@Brent.Longborough Oh that bible

@DavidCarlisle I switched back to Firefox about three weeks ago - I can't afford a PC with 10gig of ram

My preferred web browser is lynx under Cygwin. Is there a web-mode for emacs?

@DavidCarlisle Careful, you may offend the Jedi

@Brent.Longborough My new one has 16 GB. :)

@egreg Chrome, paraphrasing Hofstadter in Gödel,Escher, Bach -- I will need them all.

@Brent.Longborough I go for simple if possible

@JosephWright Yes. I'm sure you may have noticed that I do too (though I have my Hermetic moments and failures). I think simplification lowers the barrier to entry.

@Brent.Longborough w3-mode was my preferred web browser for a while (I think it's older than lynx as a text mode browser, doesn't seem to be installed by default any more)

@DavidCarlisle Text browsers were so much better. Javascript is a curse. And now, we can't do this chat on Lynx or w3-mode...

@Brent.Longborough could use the phone.

@DavidCarlisle Should I install Chrome?

@egreg 16 Gb è troppo poco

@egreg It's a good browser generally but we are "in discussion" with them over the fact that they added MathML support at 23 and removed it at 24 so personally I prefer firefox (which has been my default browser since before it was called firefox) (But I install them all anyway:-)

@egreg PSmith is never here when you have difficult questions to ask

@DavidCarlisle Google's getting really evil these days. Abandoning Reader was the last straw.

@DavidCarlisle I have a G+ account but rarely look at it. Except for here in the chat, my Intertubes life is based on RSS

Just stumbled on a site: latexsearch.com

@KannappanSampath This is worth an ad!

@tohecz I am afraid, I don't have enough fu to pull one off!

@KannappanSampath I'll try my best ;)

@tohecz Hah! Good Luck.

Tsssk. An octave script is not running.

Oh and BTW, it was from this place: nfillion.com/index.php/resources/latex-reference

The author is asking for comments on: plus.google.com/u/0/111458925431759799575/posts/bHvVFVXoPWN

I have left one... :)

Where's @PauloCereda when arara is questioned? ;-)

@egreg hiding

latexsearch.com

@tohecz Great!! And, real quick!

@KannappanSampath Copy-and-paste works for screenshots, too ;)

@tohecz I am yet to figure out, how I'd take a screenshot of a particular region..,

@KannappanSampath You have to use cut-and-paste, too, of course ;)

@tohecz Upvoted. :)

Would a distro-upgrade do anything to my files on ubuntu?

@KannappanSampath files probably not, but it might be a good idea to make a backup

@tohecz I have started one, now to 12.10... :)

@KannappanSampath you want to go 11 versions up? Then I suggest a clean install!

@tohecz I mean ubuntu 12.10 -- how's that 11 versions up?

oh I thought "one" means "Ubuntu 1" ;)

sorry, my mistake

Oh... OKAY. :)

Hi folks, is it a bug or feature?

Apparently pgffor's \foreach \t in {45,45,..,360}{\rput(4;\t){*}} works! But not for PSTricks' \psforeach.

Oh i missuderstood.

No problem indeed.

@Karl'sstudents Probably a missing check that the increment is non zero.

@egreg I misunderstood the second item in the list. I though it is the incremental step, apparently it is just the second item of the sequence from which the step will be calculated.

Thus no problem.

@Karl'sstudents Actually a check for wrong input should be added.

@egreg Yes. I agree with you!

Does anyone know how stable interface3 is? Just discovered it after defining my own (and probably worse) sequence / collection mechanism :(

I use biblatex and biber with style author-year, however in \citet I would like to remove the year, possible?

@Lonelyisland You can use \citeauthor to cite just the author.

perfect

@Nicola Your web resources have come in handy for me! :)

@KannappanSampath Thanks :-)

I have used them several times now. Especially, some things reg. hyperref.

can I prevent in authoryear citestyle the comma between author and year in \citep

huhu

how can I set prenote for \cite globally?

What did I miss? :)

@PauloCereda Have you seen tex.stackexchange.com/questions/103466/is-arara-safe-secure

@TorbjørnT. ooh! Let me check it! :)

@PauloCereda We missed you: São Paulo?

@PauloCereda Tomorrow we'll be enemies. ;-)

@egreg I miss you guys too! :) And yes, a good trip to São Paulo. :)

@egreg Oh no! :) Italy will win. :)

why is the spacing changed by adding {}?

@DominicMichaelis very roughly, some commands eat the space until they receive something.... and {} acts like something and leave the spaces after that in place.

Any idea how to make the footnote counter count continously without resetting

@DominicMichaelis Sorry i just saw it and going back to my misery with matlab

matlab is such a crap i am so sorry for you

i did need it for numerics and it made me so angry

instead of paying lots of money to epilation, women should solve semidefinite optimization problems in matlab. It's more effective. A minor side-effect is that it starts from the head to remove hair.

i guess anything doing with matlab removes hair from the head

@Lonelyisland What document class?

things like \setcounter{footnote}{100} have no effec

scrreprt

@Lonelyisland \usepackage{chngcntr} and \counterwithout{footnote}{chapter} should do.

I placed this code directly before \begin{document} without any effect

it resets every page

@percusse LOL.

Oh, and BTW, we ended up using Octave for the course. @percusse

@KannappanSampath Still the same pain but at least for free :)

From within the \foreach body, how to know the current index without using \pgfmathsetmacro?

@DominicMichaelis In text mode what @percusse said, inmath mode because it's an empty math atom so (in particular) allows mathbin atoms to get mathbin spacing. $+1$ then + is start of a math list so reverts to mathord spacing but ${}+1$ it is infix \mathbin there is a question on site with full details somewhere

egreg: any further idea?

@percusse Well, to be very honest, I haven't yet seen the pain... we have been using this with nmm toolbox and pretty much, the only thing we have done is learn to find roots.

The following does not compile. :-)

\documentclass[pstricks,border=12pt]{standalone}
\SpecialCoor
\makeatletter
\def\N{26}

\begin{document}
% speficy the angular distance between the 2 sets of alphabets
\def\offset{13}
\begin{pspicture}(-5,-5)(5,5)
    \psforeach{\r} {2,3,4}{\pscircle{\r}}
    \degrees[\N]
    \foreach \t [count=\i] in {A,...,Z}{%
        \psline(2;\i)(4;\i)
        \pstVerb{/angle {\i\space 0.5 add} bind def}%
        \rput{!angle 6.5 sub}(!3.5 angle \pst@angleunit PtoC){\t}
        \rput{!angle \offset\space add 6.5 sub}(!2.5 angle \offset\space add \pst@angleunit PtoC){\t}

@DavidCarlisle could you find it? i was going to write a question about it, don't want to make a duplicate

In other words, what indexing macro can we use inside the looping body?

@DominicMichaelis hmm there is this one but I thought there was one with an @egreg opus somewhere tex.stackexchange.com/questions/96862/…

@KannappanSampath Sure, numerical methods are not inherently efficient so doing it iteratively doesn't hurt so bad unless you are solving a horrible integral equation etc. (or if you are caring enough to check the error accumulation ...) It's the pain when you start using the advertised matrix computations. They are not as accurate when you really start using them. I mean out of the academic toy examples world.

@percusse I might never. :-P

@DominicMichaelis this one might be better tex.stackexchange.com/questions/86385/…

@KannappanSampath You really don't know how lucky you are :)

@Lonelyisland About footnotes? If they reset at every page you must have a package that does it. Or enabled some feature of KoMa-script to the effect.

@percusse there are alternative software choices:-)

I am amidst a million mosquitoes and it is 4 and I need to prove dominated convergence theorem.

@DavidCarlisle I've talked to my department about it. They seem casually interested. But they also seem to prefer a MOSEK license.

@percusse The advantages of being an aspiring pure mathematician ^^^

@KannappanSampath Didn't Lebesgue prove it 100 years ago? :P

@egreg He did; but I should prove it for me to understand, how he did it.

This is why I've wasted two days....

a>1 is a scalar inequality

when a is a scalar

but for some reason matlab wanted it to be a matrix

so a>1 becomes

a >[1 1]
   [1 1]

aaaaaaaarghhhh

@percusse there's a clue in the name of the software.

%$#%^#$%&#$^%#$%^#$%

@percusse as i said that software is totally bull... i am really sry for you

@DavidCarlisle the name is actually matrix laboratory

experimenting on humans

@percusse but scalar types are layered over the original idea that simplified the syntax for declaring/using matrices by making everything an array of doubles so 1 is a 1x1 double array rather than an integer, current versions move away from that view but it still shows through in places

@david but if you add a scalar to a matrix, the scalar will be interpreted as as matrix of the right size with every value the scalar

@DavidCarlisle The problem is that it doesn't expand to an identity matrix but to a ones matrix.

as it always do :D

@David I am fascinated by your versatility: you can talk about several things authoritatively! Hats off to you.

@percusse yes well all software has its quirks:-)

@DavidCarlisle I'll slap everyone in favor of MATLAB with a Golub/Van Loan book :)

@KannappanSampath But perhaps it's a skill of 25 years bluffing on the internet rather than actual knowledge.

@percusse In my current projects in the day job I have a choice of getting matlab or IE to do something sensible. I promise you there isn't really any comparison. Would you rather work with IE10?

what are the package of the pgf bundle that I must load in order to use the math functions?

to pick random numbers and things like that

of course tikz works, but I'm looking for something minimal

particularly Im using \pgfmathrandominteger

@DavidCarlisle Well, your command of software is something that I don't possess. I simply want to do engineering and not debugging for months just to get some academic satisfaction :)

@leo With the regular 2.10 version you need pgfkeys, pgfmath both (If i remember correctly).

I think CVS version fixed it to pgfmath as a standalone product.

@percusse what is CSV version?

@leo It's the current development version, it's not released officially. Think of it as the beta version of the next release.

@percusse I see, thanks

@percusse Yes well I wanted to be a research pure mathematician but I grew up to write xii.tex, things don't always go the way you plan.

@DavidCarlisle I don't think I've come close to a xii.tex yet :)

Is there a TeX macro to print a character when its integer equivalence is given? For example, \print{65} then it prints A.

@Karl'sstudents If you have hex code \documentclass{article}\begin{document}\char^^41\end{document}`. If ASCII code

@percusse Nice. Thank you!

Windows 8 is Chicago Bulls...

@DavidCarlisle Hah, my plan too. Hopefully, all would go well...

@DominicMichaelis I just answered your qn but @egreg beat me too it by a minute:-)

@egreg could i chaged my example a bit it needs to be {}-\alpha and not -{} \alpha

@DominicMichaelis Of course

@KannappanSampath The following does not compile:

\documentclass[pstricks,border=12pt]{standalone}
\SpecialCoor
\makeatletter
\def\N{26}

\begin{document}
% speficy the angular distance between the 2 sets of alphabets
\def\offset{13}
\begin{pspicture}(-5,-5)(5,5)
    \psforeach{\r} {2,3,4}{\pscircle{\r}}
    \degrees[\N]
    \psforeach{\t}{41,..,66}{%
        \psline(2;\the\psLoopIndex)(4;\the\psLoopIndex)
        \pstVerb{/angle {\the\psLoopIndex\space 0.5 add} bind def}%
        \rput{!angle 6.5 sub}(!3.5 angle \pst@angleunit PtoC){\t}
        \rput{!angle \offset\space add 6.5 sub}(!2.5 angle \offset\space add \pst@angleunit PtoC){\char…

@DavidCarlisle please post it too :) i will surely upvote it

@Karl'sstudents I am not sure I can help. I think it was meant for @egreg, @David, @percusse et al.

@KannappanSampath I am sorry, I wrongly clicked the reply button.

@Karl'sstudents never mind.

@percusse The following does not compile.

\documentclass[pstricks,border=12pt]{standalone}
\SpecialCoor
\makeatletter
\def\N{26}

\begin{document}
% speficy the angular distance between the 2 sets of alphabets
\def\offset{13}
\begin{pspicture}(-5,-5)(5,5)
    \psforeach{\r} {2,3,4}{\pscircle{\r}}
    \degrees[\N]
    \psforeach{\t}{41,42,..,66}{%
        \psline(2;\the\psLoopIndex)(4;\the\psLoopIndex)
        \pstVerb{/angle {\the\psLoopIndex\space 0.5 add} bind def}%
        \rput{!angle 6.5 sub}(!3.5 angle \pst@angleunit PtoC){\t}
        \rput{!angle \offset\space add 6.5 sub}(!2.5 angle \offset\space add \pst@angleunit PtoC){\c…

@DominicMichaelis Actually, $-a{}-a$ is just the same as $-a-{}a$ and $-a-a$

@egreg yeah but $-{} a$ is different to ${}-a$

@DominicMichaelis Yes, of course: the first is "{Bin} Ord Ord", the second is "Ord Bin Ord".

for me it is a totally new world :D

The simplified example but does not compile:

\documentclass{article}
\usepackage{pgffor}
\begin{document}
	\foreach \t in {41,42,..,66}{\char^^\t\par}
\end{document}

@DominicMichaelis done:-)

@DavidCarlisle Asking TeX to show the atoms is cheating. Like filling a crossword puzzle the day after, with the solution in front of you. :)

\char number without ^^ works. DONE.

Thanks @percusse!

@Karl'sstudents my mistake it's \char`^^ actually

@Karl'sstudents \char^^\t is illegal. And char`^^\t either.

@egreg \char \t works.

@Karl'sstudents Yes, if \t is a number. Actually \char`^^\t doesn't give errors, but by chance.

@egreg \char`^^\t does not work in my case.

@egreg I'd never do that of course, I just suggest it as a possibility for beginners:-)

@Karl'sstudents Of course: \char`^^\ is equivalent to say \char28, because 28+64=92 which is the character code of the backslash. So with cmr10 you get, from \char`^^\t, the output "øt"

The control sequence \t is never seen.

@Karl'sstudents Note "does not give errors" is not contradictory with "does not work"

@DavidCarlisle So both statements are equal?

being not contradictory is not equivalent to being equivalent

@Karl'sstudents not equal, just not contradictory. egreg said it did not give errors, that is the syntax is legal, and TeX does something so it is working as designed. you said "it did not work" but that is to be read as "did not work the way I wanted" which is a different thing all together.

@DavidCarlisle Oh. We might be using different code fragment during the test. I used this code in my test. :-)

\documentclass[pstricks,border=12pt]{standalone}
\SpecialCoor
\makeatletter
\def\N{26}


\begin{document}
% speficy the angular distance between the 2 sets of alphabets
\def\offset{13}
\begin{pspicture}(-5,-5)(5,5)
    \psforeach{\r}{2,3,4}{\pscircle{\r}}
    \degrees[\N]
    \psforeach{\t}{65,66,..,90}{%
        \psline(2;\the\psLoopIndex)(4;\the\psLoopIndex)
        \pstVerb{/angle {\the\psLoopIndex\space 0.5 add} bind def}%
        \rput{!angle 6.5 sub}(!3.5 angle \pst@angleunit PtoC){\char`^^\t}

@Karl'sstudents I don't get it; why should you want the ^^ notation to begin with?

don't do this {\char'^^\t} it can't possibly do anything useful, even though it does something

@egreg I just copied from percusse's comment below.

@Karl'sstudents It's a very different thing. TeX transforms ^^41 into a character before starting macro expansion.

@DavidCarlisle PSTricks' pipe line does not like the illegal character.

@egreg I thought \char^^can be appended with a variable \t (which is the \foreach's variable).

@Karl'sstudents So when it sees ^^\t it follows its rules and passes to the next stages the tokens <character 28>t and there's no control sequence \t there.

@Karl'sstudents That's a different thing clearly if you put random unexpected tokens into a complicated macro it fails somewhere down the line.

OK. I think it is my misconception. I thought \char^^ is a macro that input a number and output a character. :D

But now \char <number> seems to work like my misconception.

:-)

@Karl'sstudents If you want to cycle over the letters, \char\t is what you need. Why looking for complications? \char wants a number after it, so it expands \t which, in your loop, takes the values from 65 to 90.

@egreg it has been implemented.

My question is actually like this:

@Karl'sstudents ^^ is before tokenisation you can use it even mid cs-name try ^^5c^^66box{a}

@Karl'sstudents There is exactly \char65\relax. Better adding \relax because many things could go wrong.

@egreg Good idea. Noted.

what packages are there to make while loops?

@leo Interesting question!

@leo lots but latex had \loop (more or less from plain tex) and \@whilenum, \@whilesw and @whiledim without any packages

I'm aware of the \whiledo provided by ifthen, but some timed I used it and crashed inside a tabular

@DavidCarlisle what I want to do is do something while the value of a counter is less than or equal to a number

@leo ah well yes if you are trying to generate tabular rows, you need to take care,:-)

@leo TeX's not that consistent, the answer depends on what the "something" is that you want to do.

@DavidCarlisle but \whiledo can be used inside an individual cell right?

@leo most likely,

@leo check out