@acl I use a custom syntax highlighting scheme and set mine in $BaseDirectory/FrontEnd/init.m. I'd like to name the colors in one place and reuse them instead of writing option -> RGBColor[x, y, z] each time... I'm not good at imagining what the RGB triplet color looks like
@RM why not do it via the Preferences? so you can transport it to different machines? and, can the whole scheme not be set from the kernel's init.m? (I've never done it and am too lazy to check)
@acl well, the preferences only change the $UserBaseDirectory/FE/init.m file, which also stores a bunch of other crud over time... I wanted a base customization file that I can save and version and use in other machines. I thought of using the kernel's init.m, but wouldn't that reset temporary changes to $FrontEndSession when I restart the kernel? I'll admit, I haven't tried it though...
an alternate is using stylesheets, but I don't know how (yet)... I'm now poking around to see if I can set the syntax highlighting using stylesheets instead of init.m
I only control the fonts, intending, borders and other stuff with the stylesheet
@rcollyer Looks like you'll overtake me this week in votes cast :)
one advantage of being a grad student, though, is that you're inevitably the main author, so that it's up to you to finish everything. you don't depend on collaborators who might slow things down, be left by their girlfriends, experience arbitrary crises etc
@RM "the reality of life catching up" is exactly right...
@RM by the way, I recently rewrote a large(ish) project I had in mma in matlab, in the hope of speeding it up. unfortunately, it was actually slower....
I believe this is due to my inexperience with matlab, at least to some extend, but I think it shows that using tools you know well is better than using tools you don't know so well but may be potentially faster. I'll stick to mma, mma with C or pure C from now on (probably pure mma to be honest)
@acl Yeah, I fully agree with that. Although I've felt that as a beginner, MATLAB forces you to know a bit about your DEs before you can even use it (which is a good thing!)... your choice of ode23/ode45, etc depends on that. On the other hand, NDSolve is more powerful and in general, fits in well with rest of mma (i.e., you can move to the next plotting/exploration step seamlessly)
@Szabolcs, @balpha: I found the bug. Some strange things seems to happen during the stripping of the js-code. Where is the best place to put an lengthy explanation? In my first question here
As most of the users may know, we developed a language extension for google-code-prettify to support the correct highlighting of Mathematica code on mathematica.stackexchange.com. Currently, this extension can be tested by installing it as local JavaScript-plugin into your browser. A detailed dis...
@acl Also, I recently found out about EquationTrekker when writing up this answer (actually I knew about it, but had never tried it out before). It's pretty neat and useful
@rcollyer looks unlikely that anybody will drop what they are doing, read the paper, implement what's described there, and write an answer based on that... so I guess he's not getting an answer if he doesn't
@RM If only there were some site where people could ask general questions about Mathematica... This isn't really "programming", so it doesn't fit on StackOverflow. And the set of Mathematica experts on SuperUsers is minuscule... Oh where oh where can we get solutions to all our interesting problems?!
Before going on, please read the comments here. Short story is that the syntax-highlighter for our Mathematica code does not work as expected since we included it as a (I hope the term is correct) server-side script. Basic example:
f[x_]:=x^2
Here the x_ should be highlighted in green because ...
@BrettChampion heh, this was for a quick sanity check to see if I was missing something obvious... I'll probably post only after trying it myself first to the best I can. Don't want to be a hypocrite when I leave comments and slap the close votes on no-effort questions :)
I've written a script that lets you insert dollar symbols automatically(via a keyboard shortcut or button) in the editor window on Stack Exchange sites that support MathJax. Feel free to check it out.
Installation instructions and details here
@Heike You are right, my solution was incorrect, so I had to delete it. I feel a little tired to rewrite it right now and there are other answers already. The user defined DistanceFunction, taking into account radii, is a good idea though.
I notice that if certain items are present, then rendering is done on the GPU. Is this generally faster? Is it possible to suppress this for better quality (e.g. antialiasing)? Is it possible to force using the GPU for other types of graphics as well?
ToBoxes is so slow. On my machine more than half of the time spent on showing graphics can be due to ToBoxes (at least in the examples I tried). If I do something like ... /. Dispatch[{Graphics -> GraphicsBox, Line -> LineBox}] and so on, and make sure that everything is numericized, I can cut the display time into half (as this is 10+ times faster than ToBoxes)
@MrWizard I'd like this and this and this merged together. The last one hasn't been migrated yet. That's strange because it's my own question, so I thought there would be no problems
@Fx I'll be frank that I have dismissed some of your comment flags. I feel that comments are history for the post, and often show interesting development for those who care to read them. I don't like to irrevocably destroy that. I do try to be reasonable.
@MrWizard I feel the same, but if you follow the discussion on this in Meta you see that this is not the only possible way of reasoning. I feel I have at least to represent something of a general view.
@MrWizard Yes, If I look closer I see they're different
@SjoerdCdeVries One way to handle it would be to only remove comments that are flagged by multiple people, but that's going to require keeping a big stack of flags making our job a lot harder.
Deleted comments cannot be undeleted, or viewed by normal users; I feel that because of this, despite the fact that comments are "meta content," it is in some ways more serious to delete comments than to delete posts.
Let's assume that I receive a flag on a comment added to an answer, asking for the removal of said comment because the suggested improvement has been used to update the answer and the comment is now moot. Should I delete this comment?
I would hesitate to do this, as I strongly believe in "credi...
@SjoerdCdeVries you're probably reading too much into it... it might've been more along the lines of "if you don't want to unilaterally delete it (for whatever reason), just close it and the 2k+ crowd will delete it"... with 36 2k+ users, there's plenty of them to do stuff round the clock :)
@RM is right. I've also noticed that there are more people participating in close votes, people who weren't part of the original crew like Ajasja.
This is a good thing. Also our visits traffic numbers are holding up and increased slightly again overnight (on the 14-day median shown on Area 51 and the site list on StackExchange).
So should I provide the last vote to close this question?