« first day (2587 days earlier)      last day (2339 days later) » 

1:19 AM
@egreg -- that's what one gets in bermuda. gordon nearly ran off the road on such a roundabout -- not used to circulating clockwise.
@AGoldMan -- i remember driving in england and the weather was quite nice, but when i started across the bridge over the severn heading into wales, the most fearful hailstorm attacked; it stopped as soon as we got to the other side. most peculiar.
 
 
6 hours later…
7:52 AM
@barbarabeeton you can go round roundabouts in the UK anti-clockwise, but it helps if you have a few cars of armed police to organise the traffic.
 
Does the code provided in this answer still compile without errors?
 
8:10 AM
@TeXnician Seems not :(
 
@JosephWright Then I'll ask a follow-up ;)
 
8:59 AM
83
Q: I am an editor for a lousy paper and I found a better algorithm than theirs. Must I share it with them?

Matilda Martin SantosI am handling a paper as an associate editor that proposed an algorithm that I find to be weak. In fact, I was able to show that a very simple, brute-force approach actually has a better running time than their algorithm. Therefore, I will recommend rejecting this paper. Do I have an obligatio...

 
@PauloCereda interesting thing...
 
@Skillmon yep!
 
@PauloCereda and funny to find that brute force is faster :)
 
@Skillmon I thought the same! The editor claims this finding is provable, so I wonder how the original algorithm is!
 
@PauloCereda must be bad as f*** :) Like brute force in core with many NOPs in between :)
 
9:07 AM
@Skillmon :)
 
@PauloCereda Which reminds me of something the free microchip compiler for their 8-bit chips did. It tried to suggest people to use the paid version because it does better optimization. In reality most of the optimization was to leave out all those NOPs the compiler spilled out in free mode...
 
@Skillmon oh my
 
9:59 AM
Hello: small question regarding \left/ \right
Is it possible to enforce that different pairs of left/right have the same (eg the max of the bunch) size?
\left | \frac{1}{z} \right | + \left | \frac{x}{z^2} \right |
Because of the 1 in the left part, the abs bars are bigger on the left side than on the right side, which doesn't look good
Is is possible to get dynamic adjustment while saying 'the both pairs of abs bars should have the same size'
?
 
@nathdwek You can use \frac{\vphantom{1}x}{z^2} on the right.
 
@UlrikeFischer Ok this would solve it, but than that assumes I know in advance which symbol is responsible for setting the delimliter height
Ideally, do you know have any idea on how to do this without knowing in advance the sizes inside the delimiters?
I guess you could put the whole expression in vphantom?
 
10:34 AM
@nathdwek you can use \middle (answer about that the other day, I'll find a link)
1
A: Using `\mathopen` and friends with `\middle`

David Carlisle I think bbb does what you ask, but the automatic size is too big in any case so it looks better if you do ccc with manually chosen size, which also then makes it easy to use the same size in all cases. \documentclass{article} \begin{document} aaa \[ \left. \middle( \sum_i a_i \middle) \m...

@nathdwek as it says there though 99 times out of a 100 you are better using \bigl and friends than \left\right
 
@DavidCarlisle thank you for your answer, I'll look it up a bit later in the day
For the moment I realized the math was wrong anyway :p
 
@nathdwek who cares so long as it looks nice:-)
 
Eh I don't work in academia anymore :'(
 
10:49 AM
@DavidCarlisle that's what I hope for my thesis
 
11:37 AM
@PauloCereda You use Lucida, so you are on the safe side.
 
11:50 AM
@mickep although it is hard to differentiate a blank page of Lucida from a blank page of Comic Sans.
 
@DavidCarlisle Touché!
 
@mickep Yes! <3
@DavidCarlisle It's hard to differentiate an empty thesis from a thesis with \color{white}... :)
 
@PauloCereda we have ways... put a stock of pink paper in the printer....
 
@DavidCarlisle oh no
A nice detail from my terminal: when the previous command failed, the lambda becomes red. :)
 
you have a lambda as your shell prompt?
 
11:56 AM
λ paulo [~] →
@DavidCarlisle ^^ It's a Z shell with a theme. :)
I use bash, but when I have to work on git projects, I like Z. :)
 
@PauloCereda I always use bash with a classic black and white theme:-)
 
@DavidCarlisle :)
 
@JosephWright I feel I need to blame someone.... From an email rhat I just received...
. Indeed, beamer redefines the macro \reset@color by
introducing a \unskip and then breaks colortbl and hhline features.
 
Oct 15 at 23:12, by Paulo Cereda
@JosephWright we can blame Frank any time. :)
 
@JosephWright so apparently I need to protect myself with a \kern otherwise you \unskip the coloured \leaders...
 
12:14 PM
 
@alhelal _ is a math mode subscript you want \_ to typeset _
 
I can't able to insert underscore(_).
@DavidCarlisle Thank you.
 
@alhelal I don't follow. Do you want LaTeX to retrieve the content from the website and typeset it?
 
@PauloCereda yes.
 
@alhelal IMHO it's unlikely to work. :(
 
@PauloCereda IMHO means?
 
@alhelal "in my humble opinion"
 
12:46 PM
@alhelal luatex can do this but classic tex has no net access (it is older than the internet:-)
@alhelal you can use --shell-escape then run wget or curl or some other such program to download the files but it's simpler just to do that first and get local copies of everything.
 
@DavidCarlisle How?
 
@alhelal how what
 
@DavidCarlisle I get your reply. OK
 
@DavidCarlisle who's on first
 
@PauloCereda who
 
12:48 PM
@DavidCarlisle or be wacky and try sockets directly. :)
@DavidCarlisle yes!
 
@PauloCereda er were you reading the latex team email yesterday?
 
@DavidCarlisle no I don't get L3 secret emails unless I am CC'ed. :)
 
@PauloCereda someone was posting luasocket code to make a post request
 
@DavidCarlisle That sounds suspiciously like Bruno. :)
 
@PauloCereda nope
 
12:50 PM
@DavidCarlisle uh-oh
@DavidCarlisle OH JOSEPH WANTS L3BUILD TO AUTOMATICALLY SUBMIT THINGS TO CTAN
 
@PauloCereda With iTerm2, you can get that information as a decoration of the terminal window, in front of the failed command, rather than the one after. (↓↓)
 
@HaraldHanche-Olsen Cool, from any shell? I might take a look at mine!
 
@PauloCereda If you have a non-standard shell, as I do, you may need to do some scripting.
 
@HaraldHanche-Olsen which one do you use?
 
@PauloCereda es. I know of exactly one other user. There may be a third, who knows?
 
1:02 PM
@HaraldHanche-Olsen ooh looks very geeky, let me check that!
 
@PauloCereda The main developer, Paul Haahr, quit quite abruptly in 1997, I think. I have found a few bug fixes here and there in the odd git repository since. My current version is based on the 0.9-beta2 release, made in December 2000!
I am reminded of some science fiction story in which software archaeologist is a recognised profession: Rather than writing code to solve problems, these people find solutions in old software, often thousands of years old.
 
@DavidCarlisle Quite possibly: with no notes it always hard to try to work out what Till was worrying about ...
 
@PauloCereda Hm. Don't you have to finish something? ;-)
 
1:17 PM
@UlrikeFischer I have. :) Working on it now. :)
 
1:55 PM
how can shift these?
 
@alhelal Two minipages, perhaps?
 
@alhelal two minipages or multicols
 
yo'
Modern UX: Important things tiny, empty space and rubbish huge.
 
@yo' ooh
 
yo'
@PauloCereda Re: new firefox, new google calendar, ... :-(
 
2:03 PM
@yo' thanks for these design principles. I'll try to use those on everything I do from now on
@yo' What's too small in new firefox?
 
@yo' oh. I actually enjoy the new FF.
I'd add the trendy buzzwords: modern, clean, beautiful, humanized, etc.
 
yo'
@Skillmon you can have only 2 lines of frequent websites on about:newtab. And the thumbnails are so small you can't see anything on them. In old firefox, I have three lines and the thumbnails are huge.
@PauloCereda you mean: shitty, empty-looking, bizzare, unusable?
 
@yo' Yes, but since we are hipster millenials, let's call that way. :)
 
yo'
@PauloCereda let's not. Seriously, this sort of stuff annoys the hell of me
 
@yo' it annoys me too.
 
yo'
2:06 PM
Anyway, I gotta go, I have a meeting at the other university downtown in 20 minutes.
but also the google calendar: why is the day number not next to the day name in the compact version?
 
@yo' I don't ever see about:newtab ...
@yo' don't use the google-calendar...
 
yo'
2:40 PM
@Skillmon it's a trend, these two were just specific examples
 
2:58 PM
@yo' I like the design of the new firefox, tbh. just one simple userChrome and it's fine :)
 
@JosephWright yes although injecting \unskip at the end of random groups where the colour changes seems a bit brave (but adding a defensive \kern after the \leader in colortbl probably makes sense anyway.
 
@DavidCarlisle Probably due to leaking spaces elsewhere: I could try taking it out ...
@DavidCarlisle I am meant to be doing a beamer release: I'll do it today then make the change for the next one, so we can test it out
 
@JosephWright Job creation scheme for @egreg?
 
3:25 PM
@UlrikeFischer :)
 
3:45 PM
@JosephWright is the missing definition for \@currname the correct solution here tex.stackexchange.com/q/403845/2388? Or is there something wrong in the expl3 changes?
 
@UlrikeFischer Blog post is right, I must have missed the \@currname setting in my earlier answer on the site
 
@JosephWright Thanks I added an answer to the new question.
@JosephWright btw: with older expl3 versions it worked without it, but now there is \@pushfilenameaux which obviously expects something like "two commands and an argument" behind it.
 
4:06 PM
@Skillmon see this
 
@UlrikeFischer Add that to the answer?
 
@Skillmon see the source
 
@JosephWright done.
 
4:33 PM
Waiting for the train after free day in Florence. Very good journey.
 
4:49 PM
@alhelal posting images of source is really not a good way to get debugging help you are better to post question as questions on the question and answer part of the site, with source being complete test document as code so people can reproduce the problem and test answers.
 
@DavidCarlisle thank you.
 
 
2 hours later…
6:32 PM
Earlier there was an issue where Dropbox and latex did not play well. Have anyone heard of similar issue with onedrive? I'm seen a Mac user demonstrate a situation where after compilation the aux file had size zero, after we disabled onedrive sync and recompiled the aux was back to normal. The Mac onedrive app did not seem to be particularly configurable.
 
@daleif I regularly use LaTeX and Dropbox, without problems with the aux file.
 
7:22 PM
@DavidCarlisle -- what have you done to amsmath? i can't get your answer to tex.stackexchange.com/q/403830 to compile.
 
8:13 PM
Hi, today @daleif pointed me to \tableum from siunitx. so far i was not aware about it (because i only read section 5.12 in manual .-( ). it works fine with simple number align. however i was lost in case at use of comparators. for example use \tablenum[table-format=2.2]{< .01} gives error: "no space reserved for comparator on line ...". google so far not help me much.
cont. : use of option for S column in \tablenum optoions also not working. can anyone tell me, where i can find more detailed information about use of \tablenum?
 
 
2 hours later…
10:01 PM
@barbarabeeton what error? (you don't even need amsmath for that)
 
-- okay, so i'm using "LaTeX2e <2016/03/31> patch level 2" (which may be the problem). error:
! Missing ) inserted for expression.
<to be read again>
(
l.9 Here is some math: $ \textrm{zzäöæømm}
\neq \textit{zzäöæømm} \neq
?
(indentations lost, of course. also, should be a linebreak before \neq.)
(oops! linebreaks are okay, but indentation is gone.)
 
10:26 PM
user image
2
i have joined the palindrome club
no awkward commas, how about that
lol
 
@barbarabeeton that makes no sense:-) That's an e-tex error from \numexpr ( 1 \relax or some such...
 
@DavidCarlisle -- i'll send you the file and the log offline (tomorrow -- my charioit awaits). very puzzled. was trying to see if \ddot (math accent) looks the same as the text accented letter, but didn't get that far.
 
@barbarabeeton thanks
 
10:45 PM
@AGoldMan one more bronze badge would be an improvement!
 

« first day (2587 days earlier)      last day (2339 days later) »