« first day (8 days earlier)      last day (3198 days later) » 
00:00 - 18:0018:00 - 00:00

12:03 AM
@Szabolcs I'm also not familiar with reddit, I've never posted there.
good, it's a blackhole there.
@yoda You praise it too highly, good sir. ;P
(It is admittedly better than 4chan, tho.)
12:25 AM
Is there some easy way of extracting something that matches a pattern from an expression? For example, I've got some convoluted object with f[x] somewhere in it, and I want to get the x.
Something like f[x_]->IWantThis
Maybe use RuleDelayed instead of Rule?
Then I'll still get the whole thing with some replacement.
I don't want to replace x, I want to extract it.
f[x_] :> x might work?
Yeah, with Cases[]. Up to Infinity level, if need be.
Cases[(* stuff *), f[x_] :> x, Infinity]
12:30 AM
That did the trick, thanks
Didn't know Cases could do replacements
there's a function importing a letter to a graphics object
but it relies on the undocumented function FilledCurve
(or at least on an undocumented feature)
In addition, it's a very hacky solution. Can this somehow be done in a cleaner way?
12:53 AM
@yoda if you can trial it, try. I find it very convenient for writing longer things (it helps with indentation, structuring packages etc). it's not as dynamic as the frontend, but for larger projects it's very useful (or even not so large, for anything over a couple of hundred lines I use the workbench). it has a nice debugger too
FilledCurve is documented. I would argue that ImportString returning an undocumented form of FilledCurve is a (small) bug.
@acl Thanks, I will give it a try
1:18 AM
@Szabolcs Yes.
2:03 AM
@yoda: if you don't mind a personal question: your actual line of work is signal/image processing, no?
@JM It involves a fair bit of signal processing, yes, but I'm not a hardware/hardcore DSP chap. It's probably right to say that I use signal processing a good deal to get results out of my data/interpret them. Recently I've been getting into a bit more on the theoretical side of signal processing than applied and I don't find it as interesting :) Image processing is more of a recent hobby and I'm learning it as I go along
@yoda are you reading any specific text on image processing?
Anybody knows the latex code for the RightTeeArrow?
@magma No, I'm not. I just look around for little problems I can solve. I sometimes try working on the blog posts on the Mathworks (signa/image proc) and WRI in the other language and try to learn that way... sometimes, I attempt to answer image proc questions on SO and DSP and learn as I go
@magma detexify.com
...and I realize that may have been less than helpful. \mapsto or \longmapsto if you want an extended one.
2:18 AM
Just a thought, going back to the discussion about the site's future logos, if it is in anyway similar to the Mathematica icosahedron, we will have to get sanction from WRI just to avoid any trademark hangups.
add one more vertex, problem solved!
@DavidZaslavsky thanks, I just found it
@yoda it would have to go to court, or be noticeably different. Otherwise, we risk trampling on their trademark.
heh, I'm just kidding here
even vague similarities have resulted in court cases.
2:21 AM
pfft.. just say it's a mine from "minesweeper"
Or a stylized head of a mace.
Somebody should ParametricPlot an overflowing stack :-P
@Szabolcs verbeia.com/mathematica/A51scraper.nb It is a modification of your earlier notebook, using the API source for the user information.
There's a Lincoln Logs proposal!
@yoda @rcollyer I doubt WRI would object. This site supports its users and promotes the use of its products
2:24 AM
(Obligatory wiki link for the uninitiated.)
@magma true, and I doubt they would, also. But, we would need to ask.
@DavidZaslavsky :)
@Szabolcs several of them had followed the proposal but a couple seem to have arrived independently. e.g. Rainforest Frog has no other SE account.
@rcollyer sure some of us can ask. Actually some of us work there, so it should not be too difficult
@magma You'd think that, but corporations fiercely protect their trademarks even if it is used on a site to promote their products. WRI is no different. For a recent example, LEGO refused rights and had some initial issues with LEGO® Answers, which is why you now see an ® in the site title and everything...
@magma yeah, there are a couple of people here who work there. But, it was a random thought, and isn't relevant until we get there.
My personal feeling is that, even if there is no objection, it makes sense to have a user contest to design something unique and beautiful and signifying the community.
2:29 AM
@yoda Trademarks must be fiercely defended because if they aren't the company will lose them.
@yoda Yeah, in fact it's required by trademark law that a trademark holder actively pursue any unauthorized use.
@ArnoudBuzing I definitely agree with that.
@rcollyer jinx!
@DavidZaslavsky buy me a beer.
@yoda anyway we can collectively come up with nice(r) MMA generated image in no time
2:31 AM
I like @ArnoudBuzing's idea of a contest. Should open that discussion up in about 45 - 60 days.
And, if the winner comes close to WRI's icosahedron, we be polite and ask.
@rcollyer yes , but let's start make the rules for the contest now (image size, type of image,ect), so when the contest actually starts people are ready with their ideas and we can declare a winner in a week or so.
@rcollyer Needs["Brewery"]; link=LinkLaunch["uubp rcollyer"]; LinkWrite[link, Beer[]]; LinkClose[link];
@DavidZaslavsky where did you ever find that? And, why hadn't I heard about it before?
@magma I think we need more input from the designer to do that. But, I'm bookmarking this for later.
There's this little place called The Internet... :-P seriously though: on a whim I Googled for 'beer transfer protocol'
I mean, it is the most logical way to get a beer from point A to point B if those points are sufficiently far apart
@DavidZaslavsky Needs["Brewery"]; link=LinkConnect["beer for me", LinkProtocol -> uubp]; Drink@LinkRead[link, Beer]; Belch[]; LinkClose[link];
2:42 AM
lol, most excellent ;-)
(I think you can get away without importing Brewery since you're only drinking the beer, not providing it)
This needs to pop up on the main site on April Fool's Day, by the way
True, but I figured they packaged together.
Ah, true
@All I just recently noticed that there is a red icon in my system tray (right side on Win-PC) which says Wolfram Mathematica 8.01. What is it for. It seems it can be used for quitting. Quitting what?
and the edit window has passed. Needs["BeerMug"] would have been better.
@magma TeXForm["\[RightTeeArrow]"]
2:45 AM
@JM in math what is a RightTeeArrow used for?
ok, I need to do actual work... later all
@DavidZaslavsky later.
@magma - could be from the browser plugin (cdf). Have you opened a notebook inside a browser recently (or not so recently)?
@rcollyer Usually a mapping. To do it in Mathematica though, you need to do something like x \[Function] x^2...
@JM I generally use a RightArrow for a mapping, but I can see the other.
2:47 AM
@rcollyer You forgot a backtick... :)
Unless you intended Needs["BeerMug.m"]
@JM I was rushed, and had beer flowing through my veins. Blame the alcohol.
Anyone care if I unpin yoda's comment about voting in meta?
Adding the info to my important discussions post on meta.
@ArnoudBuzing I had a CDF in Firefox yesterday and i thought it was related to it, but today I have no CDF and the icon was there again. I now quit it, but I do not see any difference. My MMA session is still on and working.
3:03 AM
@magma - It's hard to tell without a screen shot of that area of your desktop. But I highly suspect it is related to the CDF plugin putting an icon in that location. You could get by getting rid of it, then opening a notebook with your browser and seeing if it reappears.
3:19 AM
Dilemma: I have a question involving extracting measurements from some pictures I have. The problem I have is it is potentially publishable. Thoughts on how to proceed?
@rcollyer You mean, if you ask it here, somebody else might get wind of your ideas?
@JM no, have to give credit.
@ArnoudBuzing I just tried and indeed it is linked to the CDF page in teh browser
@rcollyer As in "co-author" credit, or "acknowledgements" credit?
@JM "co-author" as the paper would ultimately be on using the technique itself in a physics lab.
3:24 AM
On that note: could you make a not too trivial toy problem out of your actual problem?
I don't know.
Maybe you can ask that first, and if somebody bites, then you can have him/her collaborate on your actual problem.
I don't think so.
@rcollyer What kind of pictures? If i may ask
If you ask your actual problem here, you've essentially tipped your hand...
...also remember the CC license we have here. You'll want to check if it's compatible with what you want to happen.
3:28 AM
@JM I don't think the journal would have an issue with it, but that is a good point.
@magma I'm sorry, it would give to much away.
@magma - Ok, that icon is 'as designed'. I don't know of a way to turn it off.
@rcollyer Yeah, I think a chat with the editor might be in order...
@ArnoudBuzing well this time it went off as soon as I closed the cdf page. Also, the Quit command reset the cdf. So we can say that it works as designed
@JM it's an AIP publication and their copyright statement doesn't appear as permissive as APS. For APS, the author retains rights to reuse.
@rcollyer maybe it is just a generic problem of how to measure something in a picture?
3:32 AM
@rcollyer Sounds bothersome...
@magma - I think it is supposed to hang around for a little bit (30 seconds?), just in case a user opens another notebook in a browser. That is done to help with startup time and loading performance issues.
@rcollyer or is it very very specidfic?
@ArnoudBuzing ah..ok, thanks for the info.
@magma ultimately, I think I can simplify it down to line detection of barely visible lines.
Binarize gets it somewhat right.
I think I'm just being lazy, and should go through the image processing docs myself.
Night all.
@rcollyer - Good night
Night all
4:12 AM
I've updated my website to advertise the site a bit more explicitly. verbeia.com/mathematica
It still gets a fair bit of traffic, so hopefully a few people come in that way.
@Verbeia Oh, a question BTW: Ted no longer maintains his stuff, I gather?
@JM he doesn't even answer his email.
But it doesn't cost me anything to keep his stuff on the site, and I'm going to keep the site because I need to keep the email addresses.
I see, thanks. I heavily used his stuff many years ago for a personal project...
I keep saying that I'll revamp the site one day.
It was meant to be a post-PhD project but given I submitted my dissertation almost 10 years ago, that hasn't really gone according to plan
A couple of other people have also already mentioned it on their blogs
The new public beta, that is, not my site)
Good night people
4:30 AM
Here's a quick(ish) edit job for people that kept their edit capabilities. I'd do it for the rep, but it is all spacing.
And, that doesn't count ... grrr.
wow, Stan Wagon is on the site now. (but unregistered - should we suggest to unregistered users to register?)
@Verbeia Hah, yes! I hope he hangs around.
Who's Stan Wagon?
@Rojo Author of ...
4:40 AM
Did Stan wagon just answer your question? Stan Wagon is on this site? o_O
Wholly mother
And I would have downvoted him if I had the rep
Now I'll upvote
@yoda He's been hanging in math.SE and MO for quite a while. It was only a matter of time... :)
(...and I somehow just knew he'd see it when we got to public beta.)
Just some tracking I've been doing via the Area 51 page
This chat get's saved for the future? Can't you remove what you write when you realise you said stuff that you shoudln't because you're sleepy and have blurred vision?
@Rojo You need some mod assistance, mostly.
But within a short time period from posting your boo-boo, you can delete it yourself.
4:47 AM
I see, thanks
Yaroslav has found us, too.
off for a while
5:20 AM
@JM The more than X comments flag is a system flag... just letting you know.
@yoda Ah, thanks. Still, odd...
The SE software still has a lot of mysteries in it...
Oh boy. It's the source of many a meta.so/prog.se rant, most recently, this one
@JM not only that, the Community user auto flags if you write two answers verbatim, if you delete more than 5 posts, if you have two consecutive closed posts, etc...
and on beta sites, it bugs moderators to comment on posts that they've closed.
Well gee, SO is hella different in scale; makes sense they'd have different problems from "smaller" sites like math.SE, or this site for that matter.
Lots of behavior don't scale well, either up or down.
That's probably what annoys me about their cookie-cutter approach.
yeah... well, I suspect this site will have a lot of "more than 20 comments" flags... we were pretty loose with comments on the mma tag and have been warned more than once that we had to cut it out
well, it's just a flag anyway... it's upto the mod to make a decision. Most probably your math mods dismiss those auto flags as nonsense, whereas some on other sites take action
So far as I can tell that thing gets totally ignored in math.SE. :D Lots of people do the Socratic route...
5:28 AM
I'm worried about your math.se mods dumping all the old mma questions on us
@yoda I might write something in the meta there...
That's what they did to dsp.. they just nastily moved all crap from the signal-processing tag the day it opened up until I yelled and waved sticks at them
On the other hand I've seen the first SO-mma.SE migration today...
I requested that. That was an exact dupe and Heike wanted to copy-paste her answer. I suggested waiting so that it can be moved here wholesale and then merged
@yoda Sorry about that; I had a part in it. There were two unanswered DSP questions in math.SE, and I thought it could get answered there if it were moved...
...they just acted on my flags.
5:31 AM
Heh, no problems. A couple of them weren't bad per se, but they were >1.5 yrs old and the OP had even deleted their account... so we had no clue what they were looking for
Man, Stan Wagon threw a wrench in my evening plans... now I have to study his code later. :D
I happened to chance upon a WRI demo that I had seen before and liked, and just noticed that it was based on your program. Pretty neat :)
there :)
You didn't have to, but okay... :)
If I really wanted to hide, I wouldn't have used initials.
5:41 AM
Heh, I didn't say which one on purpose, but since you asked, well...
But still, Manipulate[] gives me trouble. It'll take me a while to master...
5:59 AM
@JM Nice, it seems Stan Wagon got interested in your question.
I've always noticed than whenever I mention his name in math.SE, he appears to answer or comment a few hours later. So, I was sort of counting on that...
4 hours later…
9:58 AM
How on earth do this many users on The Great Outdoors get that much rep in one day? There were only 70-odd questions asked, so maximum 1200 points from accepts. That's only enough for the top four users' reps to be where they are, never mind the rest area51.stackexchange.com/proposals/3659/the-great-outdoors
10:36 AM
@Verbeia Hmm...
10:48 AM
@Verbeia maybe the rep got transferred from the old proposals when they were merged together.
Yeah, that's one possibility.
11:10 AM
Might as well idle here, too :-)
@robjohn Oh hey. Have you seen the questions here? Lots of nice stuff...
@JM Can I use my math.se login to join mma.se, or do I need to use google again?
@robjohn As long as you're logged in with math.SE, you could still use your credentials.
Just click "login".
@yoda Speaking of which, there's a question on SO that would definitely be considered off topic there. I've voted to close, but maybe a flag would be more appropriate to have it sent our way. Not that the user has noticed my questions, yet.
@JM Yeah, but I had to use the Google Open ID, then it linked to my other accounts.
11:20 AM
@robjohn Oh, I see. Apparently I haven't used the Google login that much. In my case it auto-associated since I was online on other sites...
@JM I am logged in at math.se, but to log in and create an account, I had to use Open ID again.
Strange. Well,at least you're in, apparently... :)
@JM I am.
BTW @everybody else: rob's the guy you thank for that MathJax on chat thingie...
Hi @robjohn
11:25 AM
@JM Does @everybody really work?
@Verbeia hello :-)
Incidentally, the JSON data from api.stackexchange.com shows new users before you can see them in mathematica.stackexchange.com/…
@robjohn It doesn't; it's just idiomatic... :)
@JM okay :-)
Ah, just considered that my gravatar is tied to my email address. Unless I change my email address, which is tied to my Open ID, I can't have a different, Mma generated, gravatar here.
Oh, right.
Leonid just broke 2k. Szabolcs is almost there. JM's a slacker and can't get there for a couple of days. :P
11:32 AM
@rcollyer Your point being? :P
Hell, sometimes you guys are too quick for me...
@JM just you don't seem to sleep as you're always here.
so, your score should be higher. :P
gotta go.
@rcollyer My computer's always on, but I can't just be thinking about math all the time you know... ;)
11:48 AM
Hello everyone
12:11 PM
Hi @Szabolcs
here is an update
good night all
Good night @Verbeia Looks great!
@Verbeia I see a number under your avatar It says 5742. what is it?
looks like a cumulative reputation
@magma It's her combined rep across all SE sites, apparently.
on all accounts, correct?
@JM yes , why it does not appear all the time. For ex. i do not see it for you
or myself
@magma The SE software is weird that way...
I'm not sure if it's too easy a question for main: is there a way to generate lattices as GraphicsComplex[] objects?
It's pretty easy to make a Table[] of Line[]s or Polygon[]s, but apparently producing GraphicsComplex[] objects is the rage these days for multiple-polygon entities...
12:21 PM
@JM a lattice of points or a grid of lines?
@Szabolcs A grid, yes.
So, the three usual ones: square, triangular, and hexagonal. I can just perform affine transformations on points if need be.
 points = Tuples[Range[0, 1, 0.1], 2];
Graphics@GraphicsComplex[points, Point[Range@Length[points]]]
This is for points ^
Nice! I wonder how to group as polygons...
   Line@Thread[{points[[;; ;; n]], points[[n ;; ;; n]]}],
   Line@Thread[{points[[1 ;; n]], points[[-n ;; -1]]}]}]
@JM ^ I don't like it, it's too much work. I'd rather work with points directly.
@JM you can group objects anyway, you can e.g. do Point[{{0,0},{1,0}}] to have two points. It will be faster to render than two separate Point objects
Sorry, I forgot this:
n = 10;
points = Tuples[Range[0, 1, 1/(n - 1)], 2];
Okay, I'll see what I can do with that; thanks!
That leaves triangles and hexagons...
1:28 PM
@JM Go to ref/FilledCurve -> Applications -> Tesselation for a fun example
Interesting. :) I note that the Table[] approach is taken as well; that's how I'm currently generating lattices.
@JM yeah, there's your day job after all, some pseudo-science called chemistry! :P
@rcollyer Hah. :P
@magma The numbers show up when your post, or cumulative posts, is large enough. For a one liner, you'll get a mini-icon next to your name. A bit longer, it grows into a larger icon. Bit more, and it adds in the cumulative rep.
@JM I couldn't resist.
It's messy but fun. What's your big deal? :D :P
1:41 PM
A: With versus Function

Leonid ShifrinTo my mind, the differences are significant if obscure. The very big difference in evaluation was described already by @Sal. Here are several more subtle ones, which may however bite you. So, functions go first. Functions Can be in two forms, Function[x,x^2] or Function[#^2] (the last is equi...

Broken scoping is broken scoping, no matter if we use Function or With or := or :>
I think once could construct an example for any of these
f[x_] := g[With[{a = #}, x] &];
g[fn_] := Module[{h}, h[a_] := fn[a]; h[0]];
@JM In my undergrad, the physics dept shared the building with the chem dept, and there was always good natured ribbing going on. Like "I may be learning about the fundamental interactions in the universe, but you're doing something useful," and the like.
And the organic chem professor was a beanpole who needed to lay off the sugar and caffeine.
@rcollyer Ah, it was a bit like that at my university as well. :) Always with the pranks... good times.
@JM BTW, have you ever read any of Derek Lowe's Things I Won't Work With blog? My personal favorite: FOOF.
Yes, I do mean dioxegen diflouride.
@rcollyer A nasty piece of work, that.
Seriously, nasty.
1:49 PM
One generally tries to minimize the use of things where you have two highly electronegative atoms next to each other...
Still highly reactive at LN2 temps when it is a solid!
Can you help me find that SO post wihch showed how to export the 3D axes as vectior graphics, while rasterizing the rest of the 3D objects?
@Szabolcs I don't recall that one.
There was a 2D one I remember
@JM unless your goal is to blow things up, or rip a hole in the fabric of reality.
1:51 PM
Found it, finally
Q: Mathematica: Rasters in 3D graphics

jmlopezThere are times when exporting to a pdf image is simply troublesome. If the data you are plotting contains many points then your figure will be big in size and the pdf viewer of your choice will spend most of its time rendering this high quality image. We can thus export this image as a jpeg, png...

@Szabolcs now I recall.
@rcollyer Yeah, except for those...
@Heike the 2D one is here: stackoverflow.com/questions/8717608/…
Q: How to decrease file size of exported plots while keeping labels sharp

Eli LanseyWhen exporting rather complicated plots (especially ListDensityPlot) as a PDF or EPS (for publication, for example), the resulting file size can be quite large. For example: data = Flatten[Table[{f0, f, Exp[-(f - f0)^2/25^2]}, {f0, 500, 700, 5}, {f, 300, 900}], 1]; plot=ListDensityPlot[data,Pl...

the 3D one looks interesting
just noticed I'm at +230 rep for today. I wonder how that happened.
@Heike No, you just have 125 today.
2:00 PM
@Szabolcs Module is similarly leaky. But, in this problem, neither Block nor Internal`InheritedBlock are. Added as a comment.
@JM If I click on the little triangle next to my name at the top of the page it says reputation 230 in the today column. What does that number mean then?
18 Mortarboards awarded, thus far. Both Leonid and Szalbolcs seem to be racing for Epic, also.
@rcollyer Hitting the rep cap 50 times is crazy! I hope I will never get that addicted :P
@Szabolcs and that's just a silver badge
@Szabolcs You've capped 7 times thus far out of the 10 days you've earned rep on.
2:09 PM
That says I need to close this web page and get back to work ...
Leonid is in the same boat.
And both of you don't need to be present to do that! Your answers are being upvoted that often.
Last night I got almost no upvotes :-)
@Szabolcs Then you aren't trying hard enough. Get back to answering questions.
This seems to be a dupe of this.
Voted to close.
Gotta to get some work done. Later.
2:32 PM
@rcollyer thanks for explanation
@magma you're welcome.
@Heike That's peculiar...
@Heike the "Equal symbol aligned math cell" was migrated.
anything other than ?
2:47 PM
Gah, I hate this frequented tags autohighlighting... I'll bookmark things as I please, software.
@magma Re: "why it does not appear all the time. For ex. i do not see it for you
or myself" you have to post enough lines in one block for the rest of your information to have room to display. Notice that on one or two line posts a small avatar graphic is used. On three line posts the large one is used. On longer posts the large graphic and reputation store is used.
@MrWizard yes, @rcollyer already explained that, thank you
@JM I think it has to do with a question that was migrated from SO to here. Apparently the rep I got for my answer to that question was transferred as well but doesn't count towards the daily allowance
@rcollyer I noticed. And the poster of this has asked his question here again.
@Verbeia The first few hours are considered day 0, not day 1. You were the only one to rep cap here, but most likely a lot of folks zoomed to 200 in the first day and then some more the next
@rcollyer and everyone else: What do we do when people cross-post? @JM @yoda Do you know the SE policy on this? --> mathematica.stackexchange.com/questions/755/… --> stackoverflow.com/questions/9019294/analytical-model-fitting
3:01 PM
@Szabolcs cross-posting is not allowed. I say we swarm and close the SO one if it is a good question and close the mma.se one if it is poor =)
besides, cross-posting doesn't make sense because everyone that used to be there (well, almost) is now here
@robjohn You can, actually... If your email provider accepts + delimiters (since you say you use gmail, they surely do), you can set your email on mma.se to be xxxx+mma@gmail.com and then set a different gravatar for that email. The mails will still go to your xxxx@gmail.com inbox
@yoda voted to close on SO as off-topic.
OK, but I'm not sure I can solve his problem.
True, but someone here may be able to do it better than on SO.
Yes, that's right
So he has a 4 by 15 matrix of complex numbers which is mapped to an ordering or something (permutaton) and an extra real number? And need to find an analytic form for that that he can fit? Very vague.
It is, and likely to get murdered on SO. Here, less likely. But, as you said, vague. It might be a technique question.
3:16 PM
I mean, even if he says: "I have a function $f: Reals -> Reals$ and want to fit it to a model. What kind of model can I fit it to?" it'd be way too broad.
Besides, it is another user brought over from the Dark Side back to the cause of good. :P
@Szabolcs ask for more detail.
SO just added G+ sharing links in the last few mintues
How do you type a blackboard bold R in LaTeX?
That'll do it.
Ah, lots of people lost edit rights now? :(
wow, I've gone 9 days without ever downvoting? The site must be really good then
@Szabolcs There are 12 who can and about 2-3 well on their way there
3:24 PM
@Szabolcs yep.
I just approved Rob's edit, that's how I realized
woo hoo. 2 points! :)
173 to go ...
heh, this site is going to be a breeze to moderate... users will do all the closing and stuff — we all pretty much know what's acceptable and what's not
So we use for all fitting questions? Sounds reasonable.
yeah, we're extremely vigilant.
@Szabolcs sounds reasonable to me, also.
3:28 PM
oh, and at 4k (equivalent to 20k in a graduated site), users can also vote to delete answers if they have a negative score. Useful for those half-spam half-helpme posts (if a mod isn't around)
@yoda yeah, and in no time at all, we'll have 2 people there.
@Szabolcs with your update to your answer to 500's question, could you add an example image, also?
@rcollyer That question needs one more close vote, is it really worth it?
@Szabolcs shrug. complete answer, and all.
but, it's up to you.
@yoda Thanks. Changing my email address doesn't mess with the Open ID that I use to log in, does it?
and the OP tends to still ask for a lot of hand holding after a large number of questions ...
if it were one of the first few questions ...
3:33 PM
that is very, very true.
Gah, that last answer took a bit to figure. Good thing old features die hard...
@robjohn nope, the open ID still remains the same.
@JM can you check the review queue?
@Szabolcs Uh, the fitting functions I'd tag under . So I have to disagree...
3:35 PM
Bye all, have to get some work done.
@JM @JM I think you're expected to use ImageResize instead nowadays.
@JM The docs themselves tend to use "fitting" much more commonly than "regression", perhaps that'd avoid some confusion? One should be a synonym of the other if we end up using them.
@Szabolcs Well, fitting can be a synonym. :)
@Heike I still like the fine-grained control, though...
Okay, maybe make "fitting" primary and "regression" the synonym...
@JM to each their own. I'm a fan of the new Image functions in mma8 so tend to use those.
Which question are you talking about @Heike?
3:43 PM
@Szabolcs the image tiling question.
@Heike s'okay. :)
@JM that'd be fitting :)
pun intended?
@yoda So you guys really prefer to be primary? :)
@Heike The alternative is not believable here. ;)
btw, was your gravatar created with mma too @Heike?
@yoda yes. It was my first attempt at creating a net of an arbitrary polyhedron
it's not very practical but looks pretty
3:49 PM
It certainly does
Indeed. :)
if you cut it out and glue everything together you should end up with a sphere
Okay, both and are extant. I'll leave synonymization to somebody else...
@JM that requires 5 answer upvotes in to add the synonym, or a moderator. So, we need some answers.
or a moderator
3:55 PM
also, a tag needs to exist in the system. Since doesn't, you can't add it as a synonym (even as a mod)
Well, obviously there's a way out — briefly add it to a question and then create the synonym
@yoda I made it, then removed the tag. So, it's in limbo now?
@JM yeah... will be auto pruned by the system when the batch job runs
added it to the list in meta.
4:41 PM
We just closed this on SO, but to be honest, it feels way too broad here as well --> mathematica.stackexchange.com/questions/755/… The answer could be a full-semester machine learning course
@Szabolcs Yeah, I got that impression too.
5:00 PM
It sounds like he's trying to find what exactly is going on, but is floundering. I don't envy him.
Question: and/or ? And does it make sense to have distinct from ?
I vote "and" :-) Two matrix questions are not necessarily linear algebra, and if someone asks about SVD, it might be strange tagged matrix ... on the other hand it's true that it's better to have less tags ... don't know what to say
Yeah, I think we can keep both here. Not all matrix use in Mathematica is necessarily linear algebraic...
Done for today. Later, y'all.
5:22 PM
@MrWizard None of these options changed anything. :-(
Do you guys know whether I can reset all options for the current session (!!!) to their default values?
That printing problem has been bugging me for quite some time now and I'd really like to fix it, since using Latex every time and copying stuff over manually isn't that enjoyable
@David I don't know, but you figure out how to reset programmatically, you could look into $FrontEndSession ... or just backup the config files, and delete them ..
@David Thanks for the upload, but I can not uncompress tar.gz. I am not using my own computer atm.
(So I don't want to install a program that can uncompress)
I'm trouble with the image uploader palette. Any help would be most appreciated:
Q: How to work around Column cutting of a pixel from images?

SzabolcsI have a problem where Column will occasionally cut off a single pixel from the top of images. To reproduce the problem, evaluate this: image = Import["http://i.stack.imgur.com/2dEty.png"] Column[{"text", Image[image, Magnification -> 1], "text"}] The output I get looks like this in the fr...

I can't let this problem stay, even if it doesn't affect the actually uploaded image. Users will think that the upper row is really cut off, and will cancel the upload.
5:47 PM
@David You're right, it's awful. I'll test for you on Linux when I get the chance.
The input text is terrible too
@Szabolcs Compared to the floating text the input is almost perfect.
00:00 - 18:0018:00 - 00:00

« first day (8 days earlier)      last day (3198 days later) »