« first day (105 days earlier)      last day (4385 days later) » 

R.M
12:10 AM
@acl You'll also need to supply either the VertexTextureCoordinates or TextureCoordinateFunction
 
acl
@RM how? (sorry to be so thick, but I can't work it out in this case)
well maybe I should post it as a question to give you some votes too
(of course it could be that that nasty yoda character answers and steals your votes, but it seems not to frequent these forums)
 
R.M
@acl hmm... I was trying to create an example, but not sure how to do it for a sphere using Graphics3d
 
acl
@RM yes I guess that was my actual question. but I can't get it right
 
R.M
Yeah... polygons take an additional option, but don't see any for sphere
Is your shape specified as polygons?
t = ExampleData[{"ColorTexture", "BerriesPattern"}];
vtc = {{0, 0}, {1, 0}, {1, 1}, {0, 1}};
coords = {{{0, 0, 0}, {0, 1, 0}, {1, 1, 0}, {1, 0, 0}}, {{0, 0,
     0}, {1, 0, 0}, {1, 0, 1}, {0, 0, 1}}, {{1, 0, 0}, {1, 1, 0}, {1,
     1, 1}, {1, 0, 1}}, {{1, 1, 0}, {0, 1, 0}, {0, 1, 1}, {1, 1,
     1}}, {{0, 1, 0}, {0, 0, 0}, {0, 0, 1}, {0, 1, 1}}, {{0, 0,
     1}, {1, 0, 1}, {1, 1, 1}, {0, 1, 1}}};
Graphics3D[{Texture[t],
  Polygon[coords, VertexTextureCoordinates -> Table[vtc, {6}]]}]
That's mostly from the docs
Here's an example of Texture in 2D, but I guess you already knew 2D
 
acl
@RM yes that's the problem, I'd like a (scaled) sphere, in 3d. polygons I can lift from the docs :)
I'll ask it tomorrow (it's late here)
 
R.M
12:38 AM
Oh hey, what do you know, it's already been asked (on day 1!)
10
Q: Texture mapping and resizing a sphere primitive in Mathematica

Mike BanteguiBackground Info In Mathematica, it's only possible to texture map a sphere through the use of SphericalPlot3D or ParametricPlot3D. image = Import["MyTexture.jpg"]; (* Pretend this is something you'd use *) sphere = SphericalPlot3D[1, {theta, 0, Pi}, {phi, 0, 2 Pi}, Mesh -> None, TextureCo...

bottom line: SphericalPlot3D is the way to go, but you can manipulate it to behave like a primitive
 
 
4 hours later…
4:10 AM
1
Q: Do we need all three tags "strings", "string-manipulation" and "text"?

VerbeiaShould strings be a synonym of text? I think string-manipulation has a good rationale for questions involving StringReplace and other transformations, but for processing, importing and exporting textual data in the form of a String, is there any real distinction between text and strings? And if ...

0
Q: Do we need tags "probability" AND "statistics"?

VerbeiaI would have thought that probability could be a synonym of statistics. But I can't create the synonym myself because I don't have the upvotes for it. What does everyone else think?

 
4:40 AM
0
Q: Is Wolfram Technical Support Officially Represented in this Forum?

StackExchangerJust curious if anyone employed by Wolfram is officially representing the company in this forum? Also is it reasonable to assume that these forums are thoroughly monitored by Wolfram's software testing team? If so, is there any need then for us to forward a copy of posts describing unexpected...

 
 
4 hours later…
acl
8:41 AM
@RM right, I was hoping to avoid that, but I guess I can't
 
 
2 hours later…
10:12 AM
@rcollyer these are stats available in the mods toolbox. Both visits and page views peaked yesterday. Visits was 80% above the previous high and views about 25%. They are both back to 'normal', but the day ain't over yet.
 
10:48 AM
@SjoerdCdeVries Beautiful. I was there about 10 years ago - on a much colder day! Completely breathtaking.
hi @rcollyer
 
@SjoerdCdeVries Ok, so we're only seeing averages/medians.
@SjoerdCdeVries beautiful. Where?
@Verbeia hola. Just popping in briefly.
 
@rcollyer It's Sagreda Familia in Barcelona
 
@Verbeia Thanks. Thought I recognized it. Impressive structure.
 
sagradafamilia.cat Didn't know Catalunya had its own toplevel domain :)
 
That is likely new.
Any way, have to run.
 
10:51 AM
ok, see you
 
acl
11:26 AM
@SjoerdCdeVries I was there once, many years ago, and found the thing breathtaking
 
@SjoerdCdeVries Is it actually finished?
 
 
1 hour later…
12:54 PM
I never realised this, but apparently there are still people out there using Mathematica 3: mathematica.stackexchange.com/q/4996/46
 
1:16 PM
i'm curious why this was deleted?
 
@EliLansey I asked that same question yesterday. Apparently it was at the request of the OP.
 
@Heike weird. it was a good question, with lots of good answers
 
@EliLansey As far as I understood some of the content in the question was confidential and shouldn't have been made public
 
@Heike ok? could be re-writted to hide any proprietary stuff
 
Looks like Aarthi stepped in to delete it, too.
 
1:25 PM
seems to go against the whole point of SE if you ask a question, get good answers, and then delete it so no one else gets any benefit
 
seriously.
 
@EliLansey That was one of the suggestions made to the OP, but I suspect he/she hasn't got back to the mods about that yet.
 
@Heike not surprising. they deleted their question w/ answers, why do we expect them to turn up again?
 
There's nothing special about the code itself, so I'm betting it was the model that was proprietary.
 
@rcollyer yeah, seems likely. But, they could just have easily asked with a bunch of random geometrical shapes standing in for buildings
 
1:28 PM
True, but by posting the original it probably got them in trouble.
 
here's @yoda's clarification of the matter.
 
This looks like it was incorrectly moved here. Flagged it for mod attention.
 
@Heike seems that removing the "super top secret" elements of the question would make it ok
i doubt that user will come back. unless they have another pressing question, of course
 
Did either of you see this answer for the InkBlot question?
13
A: How to make an inkblot?

Simon WoodsThis approach is based on a random walk of a shrinking disk. Several of these are combined and a Gaussian filter is used to smooth it out. Optionally the smoothed image can be multiplied by the original to restore the tiny "droplets" that are wiped out by the smoothing. There is a streakiness par...

 
@rcollyer wow
 
1:36 PM
I made this from it:
It seems like it is very easy to make bugs:
 
random walks?
 
@Szabolcs I think it is ultimately based on random walks, yes.
 
I took a driving lesson today, it was a bit scary
 
@Szabolcs yeah. "This approach is based on a random walk of a shrinking disk."
 
@Szabolcs Yes, they are. My sympathies.
@Szabolcs automatic or standard transmission?
 
1:41 PM
@rcollyer It was fun, but didn't expect to go out on the road in traffic the first day
@rcollyer manual
 
@Szabolcs good fun!
 
@Szabolcs harder to learn (more coordination is required), but more fun to drive.
 
can't imagine learning on a manual, though. i learned on a automatic, and then later learned to drive stick
 
I started on a manual. Definitely an interesting experience.
 
It was a bit overwhelming to pay attention to all those things
 
1:42 PM
@rcollyer I think manual transmission is the default in most parts of Europe.
 
Well, if I don't learn manual, I won't be able to drive almost any car here in Europe (or at least in my country) :)
 
@Heike It used to be the same here, but most people drive automatics here, now. My wife has knee problems, so manual is out.
 
@rcollyer my wife doesn't know how to drive manual, so we are also stuck with automatic
 
@Szabolcs it's worthwhile learning, but as you said, there is a lot to pay attention to. Beware of tunnel vision.
 
but we have a car with what i like to call "semi-automatic"
where you can manually force gear changes, but it all happens electronically
 
1:44 PM
A friend of mine had to shift his car into neutral when stopped to keep it running.
 
@rcollyer That's normal for a manual car.
 
@Heike correct, but not for an automatic. it was just in really bad shape.
 
When I was taking driving lessons my instructor regularly complimented me with my control of the clutch by which he meant that maybe I should speed things up a bit when parking the car.
 
It seems we drove 7 km --> goo.gl/CVgDx
 
@Szabolcs You got to do a roundabout as well.
 
1:48 PM
well, I got a lot of help, but it was scary ...
 
@Szabolcs fun!
 
I don't have a feeling for the width of the car (which really bothers me), and also not a good feeling for how fast it turns (which is not a big deal, at the end it was already better than the beginning)
The bicycle feels almost like it's part of my body, turning is so natural. It seems this is not the case when the vehicle is controlled with a wheel.
 
Not at the beginning, but it can get there.
 
@Szabolcs we recently got a new car which is wider than our old one
im still not entirely comfortable with the new width
 
Hi @J.M. !
 
1:53 PM
Hi Szabolcs!
 
@JM Hola! Nice teapot!
 
Did you see those incorrectly migrated questions? At least two of them today
 
@rcollyer Hi! I'm slowly getting better with making fake objects in Mathematica... :)
@Szabolcs I sent one off to math.SE. The behavior at SO puzzles me...
 
@JM the real trick is to make real objects with mma ... but the 3D printer interface isn't available, yet. :)
 
@rcollyer That sounds like a long wait... :)
 
1:55 PM
@JM but, it would be fun.
 
@JM Ah, I see you deleted the other one
 
@JM Our you in a position to be back full-time, yet, or still intermittent?
 
I'd always wanted to do minimal surfaces with a 3D printer, but, they look flimsy...
@rcollyer I think I'm more or less here, for the forseeable future. :)
@Szabolcs It was already in math.SE, so I deleted it.
 
@JM The minimal surfaces probably would be, but the 3D printer tech is getting better, and it can be used to construct replacement parts.
 
(On another note, I also gave an answer to that fake inkblot question. It's a long way from Rorschach, but it's a start.)
 
1:58 PM
@JM Does everyone drive a car in the Philippines? Or can you get by without one? I guess many people can't afford one.
 
@JM I saw that, and upvoted it.
 
@Szabolcs Most of the middle and upper class do. I don't. :) Public transportation works fine for me.
@rcollyer Thanks. Perlin noise is extremely flexible. I colored the teapot with it, and now it does nicely for inkblots. :)
 
@JM I'm currently having fun with Simon Wood's alien bug maker version, though.
 
Yes, that was an amusing piece.
 
@JM just flagged a not-an-answer
 
2:06 PM
Let me look...
In the meantime: see this
On that note: I don't feel like merging accounts, seeing that cookie loss is very much likely.
 
@JM He said he does not want to register ... so no point in merging until he does
 
@JM hysterical.
 
@JM Did someone mention cookies?
 
@Heike I was talking about that guy who doesn't want to register... no point in merging if he'll lose his cookies again.
 
2:49 PM
quick question:
in Reap, Sow type situations, do you ever use the first entry you get from Reap? Or really, it's always the sown values...
 
@EliLansey in the times I used that pair, never.
 
@EliLansey The first list can be useful if you've transformed the sown values in any way. In my attempts to compete with the other answers to Andy's question, I needed both the first list and the second.
Of course, that was the only time they proved useful, and I never got close to the speed needed.
 
@JM i love this!
 
@EliLansey The frustration just oozes from the words... :D
 
@JM been there.
a reviewer once commented on my paper, "Figure 6 looks odd."
i REALLY wanted to respond
Well, Figure 6 thinks Reviewer 1 looks odd, as well.
 
2:58 PM
:D
He never elaborated on why it looks odd?
 
@JM nope, that was it
i made some changes to the plot and instead said "I hope this clears up any confusion reviewer 1 had"
 
So much for reviewers being helpful...
 
Reviewers are never confused. They just get distracted (sometimes) by the pile of lordly work in their desks
be respectful :)
or die
 
On the other hand, it's not uncommon to see a reviewer with an axe to grind...
 
3:21 PM
@JM Can you merge this and this ?
 
Hopefully the surgery worked nicely.
 
@EliLansey I once got back a 2 page review with 2 scientific comments and the rest being detailed proofreading remarks... "page 4, col 2, line 5, word 6 — 'data are', not 'data is' "
That's perhaps my biggest beef with reviewers and journals... the pettiness on the usage of "data"
 
@yoda I consider "data" plural myself, but there are better things to spend time on...
 
"data" by itself is plural if you consider strict latin, but "data", when used in that context is a collective noun and collective nouns always take the singular form of the verb
 
3:36 PM
Ah, you treat it as a collective noun. In that case...
(Relatedly: "hardwares" and "softwares" hurts my eyes.)
 
@yoda the only review I did was essentially devoid of scientific content as I didn't have enough experience with the particular sub-field. So, I reviewed it for clarity of presentation v. content.
 
either way, most (if not all) journals have a proofreading stage that comes after the review... it just irks me that the reviewer spent 2-3 months only to point out typos... makes ou wonder if he ever values his time
@rcollyer why did you agree to do the review?
 
@rcollyer you couldn't have asked the job to be passed to somebody else?
 
@yoda it was a conference proceeding, and it was assigned by similarity to what I had done. Unfortunately, I was the only one close to what they were doing.
 
@yoda eh, he probably had a grad student review
 
3:40 PM
@rcollyer ah I see.
 
@EliLansey That's the more likely scenario, yes.
 
@EliLansey very likely...
 
At best he glanced over the grad student's job and then gave his imprimatur.
 
btw, that letter was hilarious... also, if someone dumps X citations that you ought to include and all X are authored by Y, then with probability 1, Y is the reviewer
 
I'd flag it with the editor, then.
 
3:43 PM
It's an easy way to pad your citations... most authors would be like 'oh wtf, I'm not adding all X.. I'll maybe insert 3-4". No biggie... still 4 free citations
 
@yoda Reminds me of the Chaos, Fractals and Solitons debacle...
 
@JM don't recall that one.
@yoda gaming the system using psychology. :)
 
acl
@yoda consider yourself lucky. I once had an attempted PRL where the first ref concluded with "this is a well-written paper on a timely topic, and I enthusiastically recommend publication in PRL" (this is verbatim). The (eventual) fourth ref said it is wrong and gave enumerated 15 points that he thought were wrong (including, inter alia, accusing us of misunderstanding chemical potential, BCS theory etc). the other refs smoothly interpolated between these two
 
@rcollyer Have a look.
 
@JM I get a 404.
 
3:48 PM
Dang, that bookmark's down. Wait...
 
@acl That sounds like fun ...
 
acl
@rcollyer well, actually, it was absurd enough to be fun
 
@acl :)
In my case, I calmly ripped apart one of the reviewer's points, point by point, citing the existing literature in the process. :)
 
acl
@rcollyer sure, that is what we did too, but so what? in practice, all he has to do is insist
 
@rcollyer Tsk, the Wayback Machine is loading slowly here. Try looking at it there.
 
3:53 PM
Elsevier B.V. () is a publishing company which publishes medical and scientific literature. It is a part of the Reed Elsevier group. Based in Amsterdam, the company has operations in the United Kingdom, USA, and elsewhere. Elsevier took its name from the Dutch publishing house Elzevir, which, however, had no connection with the present company. The Elzevir family operated as booksellers and publishers in the Netherlands. Its founder, Lodewijk Elzevir (1542–1617), lived in Leiden and established the business in 1580. The modern company was founded in 1880. Leading products include...
 
@acl Ah, true.
@JM That apparently went dark in 2010. But, you're talking about the editor using his "discretion" to publish his own papers, right?
 
Anyway, as a summary: El Naschie put up this journal on Elsevier, which turned out to be more or less a vanity journal.
 
@verbeia @acl @heike @rcollyer Indeed the Sagrada Familia. I didn't have time this morning as I was about to go on a grueling Gaudi tour. My feet are now two sizes larger. The Sagrada is, apart from its beauty, interesting for its mathematical aspects.
 
@acl 4 reviewers eh? harsh. Also, the associate editor should step in and weed out absurd reviews
 
Apart from the self-citation, papers submitted to it were "subtly" encouraged to cite El Naschie's papers. So, there...
 
3:57 PM
@yoda PRL is tough to get into.
 
@SjoerdCdeVries Can you share pics?
 
acl
@yoda we then sent it to PRA where a fifth reviewer was appointed. he said yes immediately...
 
Gaudi changed from the classical gothic straights structures to parabolic and hyperbolic ones.
 
@SjoerdCdeVries it's a beautiful building.
 
@JM I'll try to upload some more, but it's rather clumsy as I have to go through Dropbox to get them of my iPhone.
 
3:59 PM
Any way, I have to get some work done today. So, bye all.
 
@SjoerdCdeVries Well, if it's inconvenient, you don't have to... :) only if you can.
 
This is the regular original design of part of the church
This is the parabolic design, note the pillars
 
Pretty!
(Now, how do we fake these in Mathematica...)
 
This is the final hyperbolic structure
That could be a nice question. There's more. Brb
 
@SjoerdCdeVries Those look almost like an Escher drawing.
 
4:06 PM
I wanna one!
 
@JM have you seen this wacky thing in a new open access journal mdpi.com/2075-1729/2/1/1
and a recent followup
note, not much else has been published in this journal since...
 
I haven't seen them yet.
I got twinges while reading the abstract. Not a good sign...
 
@EliLansey At my previous workplace we had a self-published booklet lying around in the common room. It was written by an Indian guy who claimed that Pi was wrong and should be some expression involving rationals and Sqrt[2]
 
@Heike: I haven't told you yet, so: I like the use of Perlin noise in your fake inkblot answer. :)
On the other hand, you used his old version: Perlin is using a quintic instead of a cubic these days as the interpolating polynomial.
 
@JM Thanks. I was inspired by your use of Perlin noise as a texture.
 
4:14 PM
@verbeia The following picture is an answer to your question whether it's finished. Actually, it isn't but the roof and the stained windows are done, so it's 'habitable' now. The main tower is still to be done. I have to get back in 20 years. My previous visit was in '91 or '92.
 
acl
@Heike I guess this happens whenever famous people work. In my former workplace, we had piles of booklets on various explanations of the "foundations of physics", sent by the authors
 
acl
literally piles, in the coffee room (next to the actual phd theses of previous members of the dept :) )
 
@EliLansey Good Jove, 105 pages? Do you have a gist?
 
@JM everything is the universe is made up of spirals + CRAPLOADS of jargon
@acl i occasionally get emails from crackpots who want me to support their submission to the arXiv
 
4:16 PM
There was this article in a medical journal where they "developed a mathematical model" to find the area under some data curve... in other words they "rediscovered" integration some 400 years after newton and leibniz! Makes you wonder why they don't teach elementary math in schools (or at least, the existence of higher math than 1+1), regardless of what discipline they major in later on...
5
 
@EliLansey a colleague pointed that one out to me a few weeks ago. Seems like a publication from bizarro-world... even if some of the points were valid, it looks like it would be almost impossible to tell due to the sea of neologisms!
 
@acl I used to get a lot of spam with alternative proofs for Fremat's last theorem.
 
at first i used to respond to 'em, but i've since given up and just ignore it
 
@yoda Ah, the rediscovery of the trapezoidal rule... nice one, that.
 
@JM lol, yup
then there's also the viXra...
that's usually good for a laugh
 
4:18 PM
While I'm waiting for my data to transfer from a remote computer, quick question: any objections to adding as a synonym to ?
 
@yoda I remember that one. I think the Author named the formula after himself.
 
found it... just in case someone wanted to lighten their day by reading the abstract and conclusions
4
 
@Heike Ha!
@yoda OMG
that is insane!
 
@yoda On the other hand, I don't laugh so much at it. The mathematics wasn't particularly deep when I was an undergraduate...
 
Gaudi used a lot of curved surfaces that are actually full of straight lines. Forgot the name of the following one. Anyone?
 
4:19 PM
@SjoerdCdeVries Ruled surfaces?
 
@EliLansey "We're too cool for the arXiv!!!1!"
 
@yoda that one was great. I recall the author/journal ended up getting pretty butthurt when everyone pointed out how ridiculous it is. I also like the way the abstract is written like a school science project.
 
@yoda I'm a bit concerned about the fact that this method was more accurate than "Other formulas widely applied by researchers". Makes me wonder what methods the author is referring to
 
@Heike yeah, i was wondering the same thing!
trying to tease that info out of the paper
 
4:21 PM
@Heike very good point.
 
@EliLansey Okay, I tried the first few pages. The author seems hell-bent on making himself not understood, while inserting recognizable words for a semblance of scholarship...
 
@Heike wow, I hadn't really looked at the paper until now (only the abstract). It really takes some gall to label an equation as "Tai's formula" instead of numbering it and the model as "Tai's model" when Tai is the author!
I always thought someone else had to recognize your work and if significant, it'll eventually get the "X's formula" moniker
 
good God - this paper has been CITED!!!
TONS of times
 
@SjoerdCdeVries Seems to be a (patch of a) hyperbolic paraboloid. You know, the saddle-like surface...
 
ok, that's it. I'm writing an article for this journal on matrix multiplication
or, linear regression
anyone want in on a co-authorship?
 
4:23 PM
basic addition, they may not get that you're making fun of them.
 
acl
seems there were comments on it, and there is also an indignant reply
 
it's been cited 160 times!
 
@EliLansey Well the trapezoidal rule is quite useful, crude as it is... ;)
 
acl
bloody hell, just look at the comments in the pdf I linked
 
acl
4:26 PM
and the author seems particularly proud of working the "model" out for herself
seriously, I usually hate it when people are made fun of, but this is ridiculous
 
@JM yeah, but citing this paper?!?
 
I like this phrase: "The precision of Tai's model. Because Tai's model is based on the calculations of individual squares and triangles, its pre- cision is obviously absolute."
 
yah
im serious, is there an article in a medical journal on linear regression?
"Although we do not have a
first reference, it is our understanding
that the trapezoidal rule was known to
Isaac Newton in the 17th century."
 
@EliLansey Pretty sure there's one somewhere...
 
@JM there's a dft package (density functional theory) that re-implements Simpson's rule in 5 or 6 different locations throughout the code ... :P
 
4:28 PM
@JM dunno. if it took 'em 400 years to learn the trapezoid rule, i wouldn't count it out
2
 
(On another note, I used to cut curves on paper myself, so maybe the rediscovery of the trapezoidal rule isn't terribly bad...)
@rcollyer Ah, they haven't heard of Gaussian quadrature I suspect...
 
@JM I remember that, and you weighed them. Interesting solution to a problem.
 
@JM i like that! that's a fantastic hands-on integral method
 
@JM probably not. But, it's a tight mesh, so Simpson's is pretty good. Although in other places, they use 4th order Lagrange interpolation, so maybe they don't consider it good enough for other things ...
 
@rcollyer I know, but Gaussian gives more bang for the same number of function evaluations... oh well.
 
4:32 PM
Ah, found it. That paper on the trapezium formula reminded me of this cartoon:
user image
5
 
@EliLansey The last time I checked a certain engineering handbook, that method was mentioned first before a short mention of the trapezoidal rule and Simpson's. No Gaussian.
 
@Heike love it!
@JM it's really clever
 
@JM true. Personally, I'd love to use this method at some point.
 
@rcollyer yeah, it's a nice one. I remember sharing with you Trefethen's paper on that...
 
@JM I have vague memories of that. My memory is a little fuzzy due to continuous sleep deprivation.
 
4:36 PM
(Also, it's still cheaper to do a DCT than to solve a symmetric eigenproblem, last I checked.)
 
I think you're right. Now, if we can convert the eigenproblem into a DCT, then we're talking.
 
(For those who don't know what rcollyer and me were referring to, see this.)
 
I don't recall, is Simpson's a Newton-Cotes rule?
 
@rcollyer Yes, the quadratic one.
 
@JM that's what I thought.
@JM As I understand numerical integration (and approximation of functionals, in general), the idea is to exactly match the functional over some subspace (i.e. annihilate) of the entire problem. To me, this is analogous to interpolation. Is there a method that is analogous to a least squares fit?
 
4:49 PM
@rcollyer perhaps integration by dust accumulation (monte carlo, afaik :))
 
@rcollyer Yeah, that's how I always thought of numerical integration: make a fake function that behaves like the function you're actually interested in, and work with that.
I've seen the use of smoothing methods before integration before.
 
Well, at least I'm not off-base in my interpretation.
Gotta run.
 
Moving averages, Savitzky-Golay... smoothing splines were the most complicated thing I saw. Though the problem of greater concern was differentiating them and not integrating them, which was almost always well-behaved...
 
R.M
If I'm not wrong, isn't 1 //. x_ :> x+1 a proof of Turing completeness as this question asks
of course, set MaxIterations -> Infinity
 
Hmm, too bad the guy who was asking about numerical inverse Laplace transforms didn't follow up. I just found the old Mathematica code I wrote...
 
 
5 hours later…
acl
10:19 PM
@JM why don't you post it, I'd like to see it :)
 

« first day (105 days earlier)      last day (4385 days later) »