« first day (318 days earlier)      last day (4598 days later) » 

5:05 AM
Interestingly, simple, non-research-inclined answers stating well-known facts seem to gather most upvotes.
 
 
2 hours later…
6:46 AM
@AndreyVihrov May that people recognize that well-known facts is the right answer and therefore upvote
 
yes
maybe people are reluctant to actually try the answers if they're too hard, and therefore won't upvote if they cannot validate the answer
 
7:41 AM
@hendrikvogt: Could you tell me if you got this notification. Thanks.
I used a special moderator-only @@ syntax which is no longer visible.
 
@MartinScharrer You can call people not in chat into chat with that syntax?
 
@NN Yes, that's the idea.
 
Is it possible for non-moderator people to call others into chat?
 
Because the chat rooms are not totally connected with the main SX sites, it also works with non-TeX.SX users!
 
@MartinScharrer Nice feature. So they get a notification via the SX thing in the top left corner on every site?
 
7:46 AM
@Raphink You can notify people which where in the same chat room already not too long ago.
@Raphink There is also the "lets discuss that in the chat" feature for long 1-on-1 comment threads.
@NN Jo
 
where is that feature?
 
@Raphink If you and another person post several comments below a question or answer the system will think you need to have an discussion and automatically offer you a link to open a new chat room with the existing comments.
You can also open you own chat room and invite people. But I think they need to have a chat account, i.e. at least be in any chat room once.
 
ah ok
hmmm
created a new room, but I can't find how to invite people
 
8:20 AM
@Raphink: You can go to the chat user profile and click 'invite user to' there.
 
ah ok :)
 
 
1 hour later…
9:25 AM
@Raphink I have to postpone my LuaTeX work until tomorrow ...
 
9:47 AM
no problem @Patrick :-)
 
10:36 AM
I'm getting angry every time I see \begin{figure}[h].
It is against the float concept.
 
What about \begin{figure}[h!] and \begin{figure}[H]? Do they make you furious?
 
@MartinScharrer I think it's a design problem. For people, figure naturally associates with figures and they expect a figure when using this environment. But for LaTeX, figure just means float. But figures need not always float!
You can see that this is not well-designed when you observe you have to use another semantically unrelated environment and even load a package to get genuine non-floating figures.
 
@AndreyVihrov What package is used to get genuine non-floating figures?
 
@NN It's the package providing the \captionof command (there's at least two of them).
 
@egreg: This is probably a better place to ask: in your edit to your answer on the lambda calculus question you had \mathgroup0. What does this do?
 
10:50 AM
@AndrewStacey It's equivalent to \fam0, which chooses the math family 0, which always contains upright letters.
 
@egreg \mathgroup is neither in the TeXBook nor in the e-TeX manual. Where'd it come from?
 
@NN I usually say that if a figure has to be just there, then it doesn't need a caption.
3
 
@egreg Indeed
 
@AndreyVihrov LaTeX defines \mathgroup as equivalent to the primitive \fam.
 
@egreg Ohh
The figure's caption and number can well be significant in LoF and when referencing it, no matter if the figure is floating or not.
 
11:02 AM
@egreg Great insight. It's very common to find in here articles and documents full of images with their respective captions, but with no reference to them in any part of the text. I had a hard time to convince my advisor to accept best float placement and "Figure n" in the text instead of "figure must be here!" and references like "figure below, figure above, figure in the previous page".
 
@PauloCereda Not to mention how many times "figure above" refers to nothing, because in some revision the passage has been moved elsewhere.
 
@AndreyVihrov Yes, many beginners have difficulties to understand the concept of figure. It should be be coded in a way that {figure}[h] automatically turns it into a non-float.
@NN Basically :-)
@NN You can also simply that the \@captype macro to figure, which is what \captionof does. See my answer tex.stackexchange.com/questions/28412/….
There was also a very short nofloat package which did the same, but it didn't got added to CTAN.
 
@egreg So true! :-)
 
11:34 AM
@AndrewStacey You're missing a \makeatother in your quotation example code.
 
@AndrewStacey I don't mean to be nagging but did you notice my latest comment to tex.stackexchange.com/questions/27899/…?
 
@AlanMunn Good catch. Thanks. (Obligatory link: tex.stackexchange.com/q/11750/86)
@NN Nag away! (Does N.N. stand for Nags iNcessantly?) I did notice it, but Tuesdays are busy days (I lecture on Wednesday mornings). The problem with passing through more than one point is twofold: 1) we're working with cubic beziers here, and once you've specified the end points then you don't have much freedom left, 2) we're working with cubic beziers here and getting them through a particular point involves solving cubic equations. (ctd)
 
@AndrewStacey Finally someone got the acronym right!
 
I chickened out a bit with that answer by making the bezier particularly nice so that the equations weren't difficult to solve. Making it pass through several points might not be possible with the "special" bezier that I used. I guess that one extra point would be possible, so long as it wasn't opposite the one given ("opposite" taken with respect to the line between the start and end).
 
11:46 AM
@AndrewStacey I see (from what I comprehend of these technicalities). Maybe your answer is the closest one can get in solving the problem with automation.
@AndrewStacey Really appreciate the effort
 
11:57 AM
Would it make sense to add a bounty as Werner suggests in his comment? tex.stackexchange.com/questions/27899/…
 
12:17 PM
@NN Not convinced. I think that there's a fair bit of mathematics behind this one that needs to be solved first. Also, the problem may need to be made a bit clearer. For the original problem (of paths avoiding nodes) then as I said in a comment, there are huge programs like graphviz that solve this and reimplementing this in TeX seems ... difficult.
So that's why I chose to solve a slightly different problem: of a path going through a point. Generalising this could be tractable (one way would be to split the path in to multiple beziers between the points), but technically it is a different question. So I think that rather than putting on a bounty, you could ask a new question (linked to the old one) that asked:
given a list of points, find a path that goes through these points in as "smooth" a fashion as possible.
@CharlesStewart: Thanks for your edit on the TikZ/PGF meta question. It's hard to keep track of all the places where these packages have been mentioned.
 
@AndrewStacey ok
 
@AndrewStacey Look at the \makeordinary macro for solving the problem of \DeclareMathSymbol.
 
12:39 PM
@egreg Great! I took the liberty of adding a screenshot. Hope that's okay.
 
 
2 hours later…
2:40 PM
9
Q: Is this rotating cube interface user-friendy?

Peter Of The CornI'm working on a prototype for an innovative form interface, where different parts of the form are shown on different sides of a cube. The cube rotates, and the user can fill it out as the cube spins. I have a working example. To me, it seems like a pretty robust solution (before, our forms took...

The first comment in this question is priceless: "You are trolling. You've gotta be trolling. Please tell me you're trolling."
 
2:53 PM
@PauloCereda hilarious
 
haha
love the selected answer, too
the guy actually took the time to fill that in?
or did he edit the screenshot?
or maybe he modified the javascript on the fly to fill it in
 
I had no idea. The cube keeps spinning nonstop for me. =P
 
he cheated in some way
it's just not possible to fill in this form
 
3:08 PM
I like this answer ux.stackexchange.com/questions/11229/… "To improve it, you could make it not be a rotating cube. [...] It's like trying to play guitar with handcuffs on."
 
So true. :-)
 
 
3 hours later…
5:50 PM
Yay! I finally got the "Disciplined" badge.
 
\o/
Time to go for the peer pressure badge!
 
That would be tricky. I don't remember exactly how many down-votes I've garnered on this site, but I think it's still in single figures so they'd have to be all on one post for me to be able to do that1
(Oops, "that!". Forgot that I had my keyboard in TeX-mode with 1 and ! swapped.)
 
aye, peer pressure seems hard to get without doing something bad like offending the community
 
6:05 PM
@NN For high-reputation users, a greater challenge would be earning the "Tumbleweed" badge. (@Werner got it at the begin of his career at tex.sx.)
 
@lockstep the tumbleweed badge seems hard to get for anyone at tex.sx because the traffic is low compared to other sx sites and because of the fact that many tex.sx users are careful to check out every new question
it seems much easier to get tumbleweed on sites with more traffic, e.g. I got one without effort at au askubuntu.com/badges/40/tumbleweed?userid=19490
also the Deputy badge was easy to get on AU since there are lots of first answers
 
 
1 hour later…
7:44 PM
@NN I could write an answer saying "Use LibreOffice, TeX sucks"; but probably everybody would know it's a joke for getting that badge! Unfortunately the Tumbleweed badge is beyond my possibilities, as I have 0 questions until now. :)
2
 
@egreg I actually thought about you having 0 questions. Never been a question you thought of posting?
 
@NN Usually I try to work out my own answers to my own problems. Who knows what can happen in the future?
Too much into answering, also: I've no time left for posing problems to myself. :)
 
@egreg of course but if there's some answer others might benefit from it makes sense (and it's justified) to post a question and answer it yourself
 
@NN Plus I've also the Italian TeX forum, where I've 10000+ messages in 7 years, which makes almost 4 each day. :)
 
@egreg You're a machine!
 
7:54 PM
@NN But that is a classical forum, so there can be many messages in the same thread.
 
8:07 PM
egreg can divide by zero. =)
 
@egreg That's more than Robin manages on c.t.t, and he does the CTAN announcements!
 
(Based on Jon Skeet facts):
Q: Can egreg ask a question that even egreg can't answer?
A: Yes. And he can answer it, too.
 
@egreg Ever posted a question in that forum?
 
8:31 PM
@PauloCereda That reminds me of a local joke; the word we use in our dialect to denote a dumb guy is mona. Not precisely "dumb"; the definition might be: "He's so mona that at the world championship of mona people he classified second".
 
8:41 PM
@egreg haha nice one. I think we have a similar joke in Portuguese, with the same context. :-)
 
@PauloCereda I should mention that mona has another meaning, that I won't report here. Suffice it to say that only a male can be mona in the sense described above. ;-)
Wow! Two starred comments by me and three about me!
 
@egreg hehe don't worry. I once saw this word and asked my mom if she knew the meaning. Thankfully, she had no idea and spotted the possibility of being a slang. I checked the dialect and found two meanings. Needless to say, I only explained the "clean" one to her. =)
 
9:05 PM
haha, three equivalent answers in 1 minute tex.stackexchange.com/questions/28465/…
I'll delete mine
 
9:37 PM
@NN I've had this image for a while
 
 
1 hour later…
10:53 PM
0
Q: Who would like to become a member of the TeX User Group, sponsored by Stack Exchange?

Stefan KottwitzThe initiative on meta Do we want Stack Exchange to become an institutional member of the TUG? got a great response. It quickly became the highest voted question on meta. This fact and the well thought out concept convinced Stack Exchange to join TUG: Stack Exchange is institutional member ...

 
@StefanKottwitz I'm already a member.
 
@egreg that's good! some are, the poll on meta showed at least 7, I remember
 
11:18 PM
@Patrick Here is one of my (unfinished) books typeset with the latest version of the new impnattypo package in debug mode, with colored ties colored in magenta and turquoise when inserted for (respectively) single letters and short last lines. I had to adjust some words still to prevent homeoarchies.
 
@egreg Ok, we've simultaneously proposed similar solutions to the example continuation problem. Is there a reason to prefer yours? (Apart from the fact that if I were a betting man, I would.)
 
Others are welcome to comment, too :-)
 
@AlanMunn I started with a more general approach, but in the end decided not to go too deep. Your solution has the "weakness" that it uses the built-in method for an attribution, which is something I wouldn't do.
 
@egreg You mean the use of [continued]?
 
@AlanMunn Yes.
 
11:32 PM
@egreg I assume that's more a matter of taste, then, or are there other reasons to avoid using it?
 
@AlanMunn Not many, in this case, I guess. But a submission to a journal might have strange effects on it, if they change that mechanism.
 
@egreg True, but they may not like a continuation environment in the first place. But I get the idea.
 
11:50 PM
@AlanMunn There's Gonzalo's answer that cuts any discussion. I should learn thmtools some day.
 

« first day (318 days earlier)      last day (4598 days later) »