Hey guys, I was wondering -- do you use \usepackage[bottom]{footmisc} in your documents? Typographically, I consider default output looking better. Technically, it probably should be sticked to the page bottom, I guess. What do you think?
@GonzaloMedina I once had the issue also mentioned in one of linked answers: A flagged post was edited shortly after my flag. IIRC I got a declined flag.
@Vochmelka I intentionally ignore you as long you use this avatar (I commented on this you surely remember).
@Speravir I see. @Vochmelka Would you be so kind to change it then if it won't be a problem? I don't know what the issue is but yeah... Otherwise, we can ask what the moderation says and get over with it via some CW or voting type of thing I guess.
@Vochmelka No offense but it looks pretty cool... as much as it can be cool as an avatar. Change your email address to something else and you'll get a new one.
@JosephWright What do you think about this or perhaps even the Powers that be? Start reading with chat.stackexchange.com/transcript/41?m=13616412 and end with my comment added below for completeness.
And to your avatar: I’m sure there are people, who feel offended (me not, but I consider it as bad taste). It would be wise to change to something less eccentric. — SperavirJan 28 at 1:54
@Vochmelka This is a ordinary gravatar avatar. The only thing is, that some of them are similar each other. Here you have a point now …
@Speravir @Vochmelka Still, as I said before, you can generate whatever gravatar you want, just fiddling with the hex digits like I did above. Then you simply download it at then put it as your avatar, so you'll have a manually set-up gravatar, which is I think pretty fine.
@Speravir yep. You come to a traffic light. It's red, and suddenly it's blinking yellow ... but at the bottom!
in all honesty, traffic in Paris is one of the craziest things I've come across. Cycling here is a thrilling experience, but presumably quite dangerous
@percusse yep, that's the highway case: blue power-LEDs. On the clothing, it's usually a well-thought combination of colours, always including blue and never including green
Hi. I was earlier having a problem where a single document tikzpicture added a blank page before the picture. David Carlisle suggested adding \setlength{\textheight}{.8\maxdimen} (not sure what this does) which got rid of the first blank page, but now I'm trying to add more stuff after the picture, but nothing is appearing. Suggestions?
@Faheem: rubber is not bad per se, but it's not actively maintained AFAIK. And I think the code is quite complex to be extended. :( But it had some good ideas (some of them used by me in arara itself).
@Speravir That's great! I've heard back from Karl. He says:
TL on Windows comes with an "internal" Perl, precisely so that our scripts (install-tl, tlmgr, etc.) can run. Otherwise, it would be impossible, since most Windows users do not have Perl, and we don't want to maintain separate programs for Windows.
He also said they (TL) install a generic .exe "wrapper" for all Windows programs (the source is included with TL, of course).
So all Perl scripts should work on TL just by running the .exe.
@PauloCereda :-) Watch out for my datatool+datatooltk article in TUGboat ;-)
Hi. I was earlier having a problem where a single document tikzpicture added a blank page before the picture. David Carlisle suggested adding \setlength{\textheight}{.8\maxdimen} (not sure what this does) which got rid of the first blank page, but now I'm trying to add more stuff after the picture, but nothing is appearing. Suggestions?
@FaheemMitha It's the default in latex not the default in any documentclass designed to put anything on paper. \maxdimen is the largest length tex can handle
@FaheemMitha in article 10pt for example it depends on the specified paper size:
@DavidCarlisle Sorry, I don't follow "the default in latex not the default in any documentclass designed to put anything on paper". I'm currently using article as my document type. if I don't set \textheight i currently get three pages. if I set it to .5\maxdimen (or seemingly anything else) I get one page.
@DavidCarlisle Ok.
@DavidCarlisle probably the best way of handling this to to figure out why there is an extra blank page at the beginning and not set \textheight at all.
My guess is the tikzpicture is too long, and is forcing a page break or something.
@FaheemMitha the value (.5\maxdimen or .8 etc) doesn't really matter much you are basically just making textheight infinitely big so tex never breaks over a page. You could say \textheight=\maxdimen but then when latex tried to add a page header you would get bigger than maxdimen when strange things happen so make it just a bit smaller than that
@FaheemMitha No if you care about that at all you shouldn't be using that method. Basically that is just telling tex not to worry about it and you don't care if it overprints the specified area a bit. If
@DavidCarlisle Exactly! Was just reading that page in TeXbook :-)
@DavidCarlisle What distance \joinrel is to remove - that is the question, actually. Where in TeXbook is defined that it is \thinmuskip, also is the part of that question.
@AndrewZabavnikov the use of joinrel is to put two relations together eg to manufacture an == from = and = so you need to lose the space that \mathrel normally puts on either side of the =
@AndrewZabavnikov \joinrel is defined as \mathrel{\mskip-3mu} because TeX adds no space between consecutive Rel atoms. So \mathrel{-}\joinrel\rightarrow produces a longer arrow as a math relation, with a thin backup between the - and \rightarrow.
@AndrewZabavnikov Yes, \mathrel{<tokens>} builds a Rel atom; \rightarrow is already a Rel atom by itself, because it's defined by \mathchardef\rightarrow="3221
@egreg But why then -3mu is used? Is it some reliance on Knuth's fonts?
(I mean that fonts must be reasonable for -3mu to work)
Oh! Stop! It's the Knuth's macro!
So he relied on his fonts, and then nobody thought about this issue?
@egreg Do you know why at normal zoom levels (like 100% or 200%) I see some glitches in the place where lower bars (while not the upper ones) of equal sign join with this code: $$ =\joinrel= $$?
@AndrewZabavnikov but in general, no they are not specified anywhere particularly accessible. computer modern, even if these days is mostly used in its type1 form derives from the metafont as egreg says so you can look at the sources, but otherwise you'd have to look at the font in some kind of font editor and figure out how much wider the specified character width is than the actual width covered with non-white pixels.
@DavidCarlisle Thank you very much! It works now. Do I always have to specify a fixed width when using 'p'? Is there a way to let 'p' columns expand as 'c' columns do?
@ComFreek yes. some bloke wrote tabularx and tabulary packages that give variable width parbox columns in tabulars (but you need to specify the total table width in that case) (and several other ristrictions)
@ComFreek Although tabularx sort or works it isn't letting p expand "like c" it instead resets the entire table multiple times trying different widths until the specified total width is achieved. It is deeply built in to the entire TeX system the distinction between horizontal mode things that have a natural width, but never break over a line and vertical mode things where line breaking happens, but requires a specified width in advance.
@DavidCarlisle "several other restrictions", does that mean that tabularx is not recommendable? I've just tried my table with tabularx, and it works except for the misaligned list (it has got a greater top margin).
@egreg That's essentially to be expected as Lua doesn't care/know about tokens in a TeX sense. I'm primarily interested in things at a 'typesetting' level, so 'characters' (probably not even LICR).
@Johannes_B Does this problem extend to case-changing? A quick test suggests not but I may be wrong
I'm making a family tree. All person correspond to a rectangle with a name in it. Male and female persons are currently only distinguished by colors, but these don't show up on black and white. Any suggestions how to mark these? I thought of male/female symbols.
@ComFreek well it's not for me to recommend it or otherwise, but the restrictions are due to the fact that the environment body is evaluated several times, which means you have to be careful with side effects like writing to files, if you only want it to happen once, and verbatim doesn't work (as it never works in a macro argument)
@JosephWright Sorry, i was afk. As far as i remember the à charackter ends in ao just like the nbsp (U+00a0), so every char ending in a0 should behave this way. But i am by far not on this topic, i was just around that day. I am also reading the luatex mailiing list, and i can remember that taco fixed the issue. But i couldn't find it in that moment. I'll have a look again.
For many years I've been using a program called "Family Tree 2.0", which came on a CD with PC Format magazine of July 2001. This software is pretty much obsolete... it won't work on operating systems beyond Windows XP (which is also nearing its end of life), and there seems to be no mention of it...
@FaheemMitha well, "clear" -> "clarity" -> "clarify", "beta" -> "betity" (the state of being beta-like) -> "betify". Well, we can ask @Alan, he's just arrived :)
Is there a correct term to refer to a decimal number where the decimal point is a full stop and there are no group separators and no scientific notation? (As in the number format required by commands like \setlength.)
@JosephWright Do many people use Y&Y nowadays? I worked with a publisher who used it, but it seems outdated to me. Maybe only because I'm used to the modern workflows.
@tohecz Hmm. -ify isn't really a very productive suffix in English, so to add it to make new words is tricky. But I think I'm with @FaheemMitha 's intuition here, that 'betafied' is more transparent than 'betified'.
@tohecz It's hard to make the the argument though, because of two facts about English: first, we don't have that many words like 'beta' that end in vowels, I think (relative to words that end in consonants) and that's the real test for what happens to the 'i' in -ify. When a word ends in a consonant, it's clear that the suffix is '-ify' but when the word ends in a vowel, what should happen, (since two vowels don't like to be together).
But looking at a bunch of existing -ify words, most of their roots (which are largely Latin based) end in consonants.
@egreg Yes, I was thinking also of that kind (although my example was not as tasty: 'agendify', (not 'agendafy'). So this provides some support for @tohecz 's version of the spelling. But I have only one reservation about that, and it has to do with the number of syllables of both your and my examples. My intuition is that there has to be enough of the root "left over" to recover it, so for three syllable roots, it's ok, but for words like 'beta' not so good.
@AlanMunn yeah, I get it I think :) Well, Czech is not so versatile in making new words as English is, but I still make a lot of new words. We have the very nice thing with the prefixes, you can make new words this way quite easily :)
i was using myopenid as my openid, but they say the service will be turned off Feb 1, though at the moment it is still accepting my password. Does anyone here have suggestions for the least awful such service that (a) won't invade my privacy or (b) disappear suddenly?
Still working at the moment, but probably only a matter of time before they get around to turning it off.
@JosephWright Well, your university can become an OpenID authority etc., the problem is that these things cost a lot (like CTAN would if not supported by people from academia and by enthusiasts like Stefan), and you need someone to pay it
@tohecz I can see that there are costs, certainly, but the basic premise of OpenID seems to be underminded if you end up having to trust Google/Yahoo/... to get one
@FaheemMitha You have probably millions of requests a day in the OpenID system. If each of them were 1kB of traffic, it's solely 1GB of traffic a day. Not a large number, but you need some good hardware and good connectibility to manage it
@FaheemMitha We are talking about one node, which is much smaller. But that all needs to be considered. If OpenID should become a leading authorization method, you need to have strong options
