Wolfram Mathematica

Stack Exchange

00:39

0

Q: What is delaying graduation of Mathematica.SE?

According to the About page: Public betas typically last from 60 to 90 days, although they can go on longer. Our site appears healthy yet it has now been in public beta for 117 days. Is there anything we can or should be doing to move the process along? Is moving the process along at this...

acl

00:49

@RM any comments on @Mr.W's question on meta?

R.M

@acl no.

acl

@RM short and to the point then

R.M

I upvoted it. SE's tactic of remaining silent gets annoying after a point, especially when there's a community that's vibrant and active. I can understand the silent treatment for most other betas out there...

acl

@RM there's an answer there now

it'll be funny if it gets shut down because not enough high-rep users exist

can't @MrWizard compensate for the lot of us?

although I notice he's back to more human levels after SO

R.M

I was about to edit my previous message to say that "I'm willing to be surprised" when you mentioned the answer... I'm not surprised that I wasn't surprised.

acl

01:06

@RM hm I just looked at webapps.se. I see your point

R.M

@acl They've been a fully graduated site since late 2010!

acl

@RM but do they actually have any mechanism in place to measure community vibrancy? I doubt it

R.M

Yes, I did; 'twas a bad choice of words on my part. I guess I should clarify: Mathematica.SE looks great - the only question is whether it's ready to graduate now or soon. The number of high-rep users is "worrying" only in the sense of "would we hamper your growth by graduating you now" - some of the folks who can and are editing or closing right now wouldn't be able to immediately after graduation, for instance. — Shog9 1 min ago

Yet another canned response... the people editing and closing now have enough rep to be editing and closing after graduation too.

acl

right, and there's another comment about a quality evaluation

so we're going to be compared to google for open questions. let's hope the influx of mindless code-dump questions recently doesn't hurt us

well, a decent evaluation would conclude that we're doing very well indeed (if I may so, and not due to my own work); but is this visible to someone who doesn't use mma daily?

2

R.M

They randomly sample questions and ask us to evaluate... that was the point of that. They're paid to do the evaluation, but outsource it as usual :P

rcollyer

01:18

@RM he is correct on some level. But, every site that graduates goes through that, and we hit that sort of bump when we went from private to public. As it stands, 39 users will retain their edit rights when we graduate, 29 will retain their open/close ability, and 4 non-mods will retain their access. Also, while I like the moderator tools, I've done just fine being a 5k+ user on SO.

acl

@RM ah, much better, I misunderstood

good

@rcollyer 39? really? I had no idea I was so privileged :)

rcollyer

@acl yeah, during public beta, it is 4k for access to everything. And, 39 is much better than a lot of sites.

acl

@rcollyer I meant that in a site on mathematica programming it's pretty funny that I end up amongst the top users. I am at best a fiddler

R.M

@acl well, ahem... I'm more of a fiddler than you are.

rcollyer

RPG has 56 2k+, so not that much greater than us, and 6 non-mod 10k+ users.

R.M

01:22

I probably have more time than you because of being lower in the academic hierarchy

rcollyer

@RM I did not realize either of you played violin.

R.M

@rcollyer Fiddler on the Chinese roof

acl

@rcollyer on roofs, too

R.M

@acl ha!

rcollyer

Brilliant!

acl

01:24

@RM I mean, this should have been obvious

19

A: Using Mathematica's Graph functions to cheat at Boggle

My solution is a recursive tree traversal algorithm which seeks and searches neighbouring vertices only if it will lead to a word (e.g., Something starting with ZQ is immediately disqualified), but it's faster than yours because I construct the adjacent vertices list from the adjacency matrix rat...

rcollyer

@acl From what I've seen: the term fiddler does not apply. Definitely a craftsman, maybe not master status (Leonid and Mr.W have that locked up), but up there.

acl

but actually it wasn't

rcollyer

@RM the same goes for you.

acl

@rcollyer come on, Heike is way up there. Leonid goes for crushing everything in front of him, Heike finesses it

(I am idol-worshipping for the first time in my life!)

rcollyer

@acl I completely agree, and not including her on that list was an oversight.

R.M

01:27

@rcollyer As penance, you should write 10 For loops

acl

@RM with GoTos

rcollyer

"10 for loops with GoTos"

Done!

R.M

@acl That's the most viewed question on the site as of today... just 3k away from a gold badge for Pillsy

rcollyer

@acl it was trying to decipher some Fortran code that used if-then-goto (Fortran IV) that I discovered that mma has Goto and Label.

acl

@rcollyer thing is, I still find it easier to program in C or Python than mma, when it comes to coding up complicated things. I apparently think imperatively.

guess I'm stupid or something

I have had various mini-epiphanies while learning lisp and mma, but still I find imperative programming easier

rcollyer

01:31

@acl it took me a long time, and a lot of it spent with the pre-v6 docs, to start to come to grips with the functional paradigm. It helped that I was learning meta-programming in c++ at the time.

acl

even though I practically only program in mma nowadays

rcollyer

I highly recommend:

acl

@rcollyer but that's it really, I haven't touched C except for trivial things in something like 5 years!

rcollyer

amazon.com/Modern-Design-Generic-Programming-Patterns/dp/…

it changes the way you think.

acl

@rcollyer nooooo......

rcollyer

01:33

@acl :)

(Insert evil laugh)

acl

ok I confess: I loved this

amazon.com/…

rcollyer

never read it.

acl

it's pretty good. but, again, I am not primarily a programmer, and I think more in equations than programs

rcollyer

so, do I.

I responded to Shog9: meta.mathematica.stackexchange.com/questions/396/…

tkott

01:54

@rcollyer just saw that you're on. Did R.M. ever send you that stylesheet based on solarized (or whatever it's called). That was gorgeous! If you did get it, do you mind sending it my way?

rcollyer

@tkott he did. Do you have an email address I can send it to? If you don't want to post your real address, consider using spamgourmet.com

I'll understand.

tkott

Oh, sorry, thought it was visibile from my profile. It's a gmail address: "tkott.spam" :)

(I throw that one around willy nilly, but useful for cases like this...)

rcollyer

@tkott sent it. I took out @RM's info.

R.M

@tkott I'll post a "how to" as an answer and probably upload that stylesheet some place so other others can use it (will also keep me motivated to finish it fully)

@rcollyer thanks :)

tkott

@rcollyer thanks :)

rcollyer

02:03

Either of you attempt the pattern madness?

@RM Since you hadn't given me permission to spread your info around, I didn't.

tkott

@RM When you do, I'll be sure to upvote it. That is a very pretty looking stylesheet!

rcollyer

I thought I spotted something earlier, but now I'm just as confused as everyone else.

tkott

@rcollyer I'm just stumped period. It took me 5 minutes of staring to even understand what the point was!

rcollyer

Look at Rojo's pattern:
In[]:=FullForm[b_[PatternSequence[_, _]]]
Out[]:=Pattern[b,Blank[]][PatternSequence[Blank[],Blank[]]]
versus Heike's
b : _[PatternSequence[_, _]] // FullForm
Pattern[b,Blank[][PatternSequence[Blank[],Blank[]]]]

In Rojo's, it is interpreted as Pattern[b,_][_,_], Heike's: Pattern[b,Blank[][_,_]]. Interesting, huh?

Hi, @Verbeia!

Verbeia

02:19

Hi there.

rcollyer

Any insight into the pattern madness?

R.M

@rcollyer certainly something with precedence

rcollyer

@RM likely.

Verbeia

Regarding the meta question - I also added a comment. I've been logging who votes to close.

rcollyer

Obviously, this works without a message: head[b_[_, _]] := {b}

Verbeia

02:29

I should note that our traffic has stagnated. I haven't updated the graphs for a while but I will soon. But it seems like we've hit a stable/stagnant patch.

rcollyer

@Verbeia he's already responded.

@Verbeia are our user counts still growing?

Verbeia

@rcollyer Yes and we have had 12 new "avid users" in the last two weeks, 194 all up. But traffic is a bit stuck.

rcollyer

@Verbeia I see that. I thought it had stabilized at a new (higher) level after all of the promotion by RM, though.

Verbeia

215 Teacher badges now

@rcollyer It has stablised just under 1500 visits/day, according to Area 51.

rcollyer

@Verbeia that's what I meant. prior to the posts by RM on reddit, I thought we were hovering near 1200/day.

Verbeia

02:36

brb - need to eat

Oleksandr R.

@rcollyer I am curious as to where the integers come from. It looks like the presence of PatternSequence causes some numbering of the patterns appearing within an expression but the way this works appears quite well hidden inside the kernel.

rcollyer

@OleksandrR That is the part I don't understand, and Trace (or, TracePrint) aren't that helpful. According to the comments, though, @kugler might have a handle on it.

Oleksandr R.

I hope so, because I can't make sense of it. Especially not for patterns like (p : {a_, b_, c_})[PatternSequence[]] -> Null.

rcollyer

@OleksandrR wtf? that is interesting, in the scratch my head sense. Maybe it has to do with the function call? ([PatternSequence[]])

@Verbeia I added a comparison to Skeptics as our top user count compares very well to theirs.

Oleksandr R.

@rcollyer I get the impression that PatternSequence is somehow "magic", like OptionsPattern, and its presence affects the parsing of the entire pattern. But what is happening or why it goes wrong, I don't know. Hooking Pattern gives no useful information.

rcollyer

02:50

@OleksandrR I didn't think it would, so I didn't try a hook.

As far as I can tell, patterns seem inert until they're acted upon, such as inside of SetDelayed or RuleDelayed.

Oleksandr R.

@rcollyer all it really shows you is that the patterns (a_, b_ &c.) are in fact being walked through in the order that they're numbered in the message.

rcollyer

@OleksandrR try removing the p, the result is almost as interesting. not useful, though.

Oleksandr R.

@rcollyer indeed, yes. That was my first iteration. Adding p does give you a bit more useful structure though.

rcollyer

Removing the a, b, and c seem to indicate that it is adding a pattern to the head of something containing PatternSequence that is the problem.

belisarius

hi all

rcollyer

02:55

Hola.

belisarius

Hola rc!

Oleksandr R.

Hello, belisarius!

rcollyer

@belisarius Any thoughts on the pattern madness question?

belisarius

@rcollyer Ha! I tried to get at something ... and all I got is a headache

rcollyer

@belisarius I understand. Thought I had something, and it was all an illusion.

belisarius

03:00

@rcollyer De ilusiones también se vive

rcollyer

@belisarius My Spanish is very rusty, can you translate what the title into English?

belisarius

@rcollyer "Miracle on 34th Street" ? :D

rcollyer

@belisarius I saw that. I meant a literal translation. :P

Verbeia

@rcollyer Wow, no non-mod 10k users other than the mods over at webapps.SE. I never looked there before. That's an oops. I see your point @rcollyer about top users on Skeptics, too.

One thing I've noticed lately is that a lot of recent first questions have come from people who actually joined a couple of months ago.

rcollyer

@Verbeia interesting, so they came back. cool.

belisarius

03:08

@rcollyer It's something problematic, because it is a common expression. "Se vive" is difficult to translate. It can be "you may also nourish from your illusions" or "you may live on your illusions" or something in between, all simultaneously.

rcollyer

@belisarius Google translate was equally helpful: "miracle on live."

belisarius

@rcollyer Nahhhhhhhhh

rcollyer

@belisarius Although, I take your two meanings as being synonymous.

Verbeia

@rcollyer this guy is one example; this guy wasn't asking his first question, but has returned;

and this guy

belisarius

@rcollyer "live on" and "nourish from" are not necessarily synonyms!

@Verbeia I asked my first Q two days ago :P

rcollyer

03:12

@belisarius absolutely. But, enough so, either would work. Though "nourish on" is a bit creepier.

J. M.

@belisarius Maybe "nourished by" is a better phrasing.

Verbeia

@belisarius indeed, but you have answered some :)

J. M.

As rcollyer says, "nourish on" sounds a tad ghoulish...

belisarius

@JM Oh well, my English never was better than that

rcollyer

@belisarius you don't count! (at least not in russian)

Oleksandr R.

03:13

So, is there a concensus that we want to graduate now? I'd been wondering myself: while we could stand amongst the other graduated sites easily enough, is our traffic high enough yet? Compared to startup business, money, or fitness (three others I would expect to graduate soon), it still seems a bit marginal. On the other hand I don't know if perhaps graduating would make our site seem more official and draw more traffic as a result.

R.M

@OleksandrR it's not our consensus that matters...

Verbeia

@OleksandrR my preference would be to get traffic a bit higher and consistently rising.

J. M.

Indeed; they have the hosting, they make the rules...

rcollyer

@OleksandrR I think we could live with being beta for another couple of months. But, personally, I'd like to graduate soon, just to continue to twist the knife.

@Verbeia I wholeheartedly agree.

J. M.

@rcollyer Whee vendettas! :D

Verbeia

03:15

That said, I'm glad Mr.Wizard asked that question on meta. I think it has prompted the staff to put us back front of mind.

Oleksandr R.

@rcollyer understandable! I agree as well, @Verbeia.

R.M

there will be a small increase in traffic when we graduate because it gets linked to from the footer of all the other graduated sites.

J. M.

@RM "There's a site for that? I didn't see it linked at the bottom..."

R.M

@Verbeia never underestimate the convincing and reassuring powers of canned responses :)

rcollyer

From our experience on Area51, the staff is very content not to communicate with us. So that, we are then forced to up the ante, and push for things ourselves. On the whole, that may not be a bad thing, but I'd like more communication, personally.

belisarius

03:16

I spent some time answering Q on beta sites that got wiped out afterwards. A nasty experience.

J. M.

@belisarius At least you were able to help a few people...

rcollyer

@belisarius I agree. At least mine were moved.

belisarius

@JM But I as didn't keep a personal log ... all my work just dissipated.

Verbeia

@belisarius It should be in the data dump

rcollyer

@JM yes, there is lingering resentment. but, I think we're better for it. we push harder.

Oleksandr R.

03:19

@rcollyer yes, the lack of feedback at times is certainly a concern. I think we're largely self-sufficient, but being at the mercy of others who won't tell (often, or clearly) you what they're thinking can be troubling.

belisarius

@Verbeia Thanks. Perhaps I get courage sometime and look for it!

rcollyer

@OleksandrR exactly, and well said.

@belisarius so were you posting things on sex.se?

belisarius

@rcollyer (blush!) How did you recognize me?

no, better don't tell

Verbeia

@rcollyer speaking of which ... here it is again

J. M.

@rcollyer I'm still laughing; I had once thought that that site would be the MOST POPULAR SITE EVAR!!!1!

rcollyer

03:22

@Verbeia hey, if they can do it, great. I don't think they can, though.

@JM and, we're still hotter than sex!

Oleksandr R.

@Verbeia what's wrong with Yahoo answers for those sorts of questions? ;)

belisarius

Everything You Always Wanted to Know About Sex* (*But Were Afraid to Ask)

Everything You Always Wanted to Know About Sex* (*But Were Afraid to Ask) (1972) is Woody Allen's fourth film, consisting of a series of short sequences loosely inspired by Dr. David Reuben's book of the same name. The film was an early smash for Allen, grossing over $18 million in the U.S. alone against a $2 million budget. Film structure The credits at the start and close of the film are played over a backdrop of a large mass of white rabbits, to the tune of Let's Misbehave by Cole Porter. The film is divided into seven vignettes, as follows: # Do Aphrodisiacs Work?A court jester (pl...

J. M.

@OleksandrR "How is babby formed? How girl get pragnent?" ;)

rcollyer

@OleksandrR have you read any yahoo answers before? Your lucky you don't get infected just from reading them, much less following their advice.

Hi, @BrettChampion. All set for tomorrow?

Oleksandr R.

@JM yep! At least 80% of all questions and answers on that site are sheer trolling.

J. M.

03:24

Hello, Brett.

Brett Champion

@rcollyer Eh? What's this about tomorrow?

Verbeia

@JM Interestingly the other site that went to beta just before us was poker.SE. I hadn't looked at its stats lately ... sad

Brett Champion

;-)

rcollyer

@BrettChampion ah, another procrastinator. ;)

J. M.

@BrettChampion Cocky, cocky. :)

rcollyer

03:25

Since you're here, @brett any thoughts on this question?

J. M.

@Verbeia Well, they really ought to shuffle things up a bit...

Brett Champion

@Verbeia I saw this evening that it was back...

Verbeia

Hi @BrettChampion - is there going to be a video to watch later of the Q&A - it's on at about 3a.m. here. Staying up that late for a Mathematica thing would definitely get my into trouble with my beloved.

rcollyer

@JM that's just wrong, that joke should be flushed.

belisarius

@rcollyer I used yahoo answers to make clear to my kid why he should not trust everything he reads in the Internet. He understood very quickly.

3

Brett Champion

03:26

@Verbeia I believe there should be a video available afterwards. Don't know how soon. (The video stuff is an entire

rcollyer

@belisarius that is funny, and scary. glad it worked.

J. M.

@belisarius You are a good man, bel. :)

Oleksandr R.

Hm, it's getting light. I should go to bed. Bye all.

Brett Champion

@JM We did a test run this morning... And a partial one Friday...

rcollyer

@OleksandrR night.

@BrettChampion I'll have to make sure I'm in a building tomorrow that doesn't drop my connection every 30 secs like the one I was in today. :|

J. M.

03:28

I'll wait for the video; my connection these days craps out even on YouTube...

Verbeia

anyway, back to work - see you all

rcollyer

@Verbeia bye.

Brett Champion

@rcollyer I have no idea about the PatternSequence question. I stay far away from the actual implementation of the pattern matcher...

rcollyer

@BrettChampion that sounds like a good thing to do.

J. M.

@BrettChampion That hairy?

Brett Champion

03:33

@JM One of the nice things about the pattern matcher is that it doesn't change very often, so doesn't need changing often.

It's the sort of thing where any change you make is likely to cause some subtle change in behavior somewhere else in the system.

rcollyer

In other words, it is very deep in the system, and literally everything else ultimately depends on how it functions. So, don't touch.

Brett Champion

Yup :-) At least not without a very good reason.

J. M.

Ah, just like the Lorenz equations... :)

rcollyer

PatternSequence is relatively new (v.6), so maybe there's something subtle here that didn't quite go right when it was added.

Repeated was also modified in v.6.

J. M.

Speaking of new stuff, I'm wondering why nobody thought of Except[] earlier myself.

rcollyer

03:37

Point, but it is v5.1 vintage, so of middling age.

belisarius

What is this? Does anybody understand it?

0

Q: --W. Mathematica Application Structure

I had such expression in "applicative form", like : Y [X][X][Z][X][Y [X][M][Z][X]][Y [X][F][Z][X]] In these structure we have different types: Y,X, Z, F . And I want to index such structure, in a way to rewrite this expresion like ..we have : Y [X][X][Z][X][Y [X][M][Z][X]][Y [X][F][Z][X]] ...

index mathematica expression-trees

Brett Champion

I like Except, but it can be devilish at times...

rcollyer

@belisarius not a clue. maybe a y-combinator type problem? but, that's just guessing.

Brett Champion

@belisarius Haven't looked at it closely, I figured it was related to the earlier question about combinators.

(And what's '--W'?)

rcollyer

@BrettChampion it works Except when you need it to. :)

@BrettChampion a bad attempt to get by some of the title filters?

belisarius

03:39

@BrettChampion The inverse of W++?

J. M.

" Mathematica ++"... hmm, I wonder.

rcollyer

@belisarius not exactly, that's a pre-decrement, W++ is a post-increment.

belisarius

@rcollyer I should ponder about it :)

rcollyer

@belisarius it is all fully understandable, see this wiki article. =)

Brett Champion

@rcollyer Pick[{1, 2, 3}, {1, 0, -1}, Except[0]]

Without evaluating, what's the result?

rcollyer

03:43

@BrettChampion sorry, already evaluated it. mostly because I had to look up how Pick operates. And, that is certainly not Kosher.

J. M.

@BrettChampion Holy crap...

belisarius

@rcollyer Many years ago, when I was learning C, I thought it was kind of pedantic to set a syntax exception just to earn one character in source code. It took me a few days to understand that the real thing was the difference between ++c and c++

J. M.

In a similar vein: Pick[{1, 2, 3}, {1, 0, -1}, _?(# =!= 0 &)]

(Pick[{1, 2, 3}, {1, 0, -1}, _?(# != 0 &)] works as expected, of course.)

rcollyer

@belisarius yep, and it still fubars people to this day.

Brett Champion

@JM Yup. It's incredibly annoying that Except[pat] isn't the negation of pat.

J. M.

03:48

@BrettChampion I know, but I'm surprised that piece of code I gave has the same malfunction...

belisarius

@rcollyer If I remember well (and it could be an implanted memory) it came from and antique Assembler (VAX , PDP11, Perkin Elmer ... one of those)

J. M.

I was about to say "I'm glad this works..." until I tested it.

rcollyer

@belisarius I know just enough assembly to understand that I need to run away and hide.

J. M.

@rcollyer You've never tried assembly language programming? :) It's fun! (for certain perverse definitions of "fun")

belisarius

@rcollyer I was my vice when I was tender and young

rcollyer

03:51

Can anyone tell me why this only prints one statement?

Internal`InheritedBlock[{UnsameQ, inQ = True},
 Unprotect[UnsameQ];
 s : UnsameQ[a_, b_] /; inQ :=
  Block[{inQ = False, res}, res = s; Print[res];  res];
 Pick[{1, 2, 3}, {1, 0, -1}, _?(# =!= 0 &)]
 ]

Brett Champion

Have it print a...

(And that's a lot of effort to do Pick[{1, 2, 3}, {1, 0, -1}, _?((Print[# =!= 0]; # =!= 0) &)]

rcollyer

@BrettChampion why yes it is, and it only prints once there, also.

Brett Champion

@rcollyer Correct. :-)

rcollyer

@JM I learned postscript, instead. Of course, I used it to test a bug in the transfer from the network controller to the main controller (i.e. print engine) of a printer. The bug involved a file that was slow to transfer. Mine took a half hour per page, as it was generating a page covered with 2pt by 2pt blocks with random colors. These print outs could then be used to give the compression algorithm a heart attack when scanning it in. :)

J. M.

@rcollyer Okay, you get a pass... thing's tricky.

Brett Champion

03:56

@rcollyer Try Pick[{1, 2, 3}, {1, 0, -1}, _?((Print[#]; # =!= 0) &)] and then remember that we're dealing with a ...Q function...

rcollyer

@BrettChampion you'll have to explain why being a Q function effects the behavior.

J. M.

...and the difference between that and Pick[{1, 2, 3}, {1, 0, -1}, _?((Print[#]; # != 0) &)] is illuminating indeed.

rcollyer

ah, I see. It is recursive decent. sob. that would be frustrating to try to get right.

Brett Champion

Time to go. 'Night all!

J. M.

But Cases[{1, 0, -1}, _?((Print[# =!= 0]; # =!= 0) &)] works nicely. It really is something with Pick[]...

See you, Brett!

rcollyer

04:01

@BrettChampion night. talk to you tomorrow.

@JM yep. As I said: recursive decent. The top level matched, so it didn't look any deeper.

@JM do you ever feel that we're just amateurs at this?

J. M.

@rcollyer I always assume I'm an amateur unless proven otherwise. :)

(...and I've been screwing around with the program for nearly two decades.)

rcollyer

@JM I thought I was good at this, and then something like Pick or PatternSequence comes along and forces me to re-evaluate my standing.

@JM The first version I bought was v3. My university had the one prior to that.

J. M.

Oh, version 2.2 was its own brand of fun. :)

rcollyer

Anyway, it is very late here, and I need sleep. Night.

belisarius

@JM I love your "one liner" aphorism

J. M.