Join 3,552 readers in helping fund MetaFilter (Hide)


Stephen Wolfram: A New Kind of Science
October 29, 2001 8:48 PM   Subscribe

Stephen Wolfram: A New Kind of Science - From the publisher's summary, "Starting from a collection of simple computer experiments — illustrated in the book by striking computer graphics — Wolfram shows how their unexpected results force a whole new way of looking at the operation of our universe." May be big. Thoughts?
posted by paladin (26 comments total) 1 user marked this as a favorite

 
I'm sorry, there seems to be some mistake, your link goes to an advertisement for a book.
posted by Doug at 9:10 PM on October 29, 2001


I read about this on slashdot. Looked pretty intresting. Does anyone know anything about how actualy correct this guy is?
posted by delmoi at 9:14 PM on October 29, 2001


I read about him in "Complexity" & "Frontiers of Complexity". I don't have the math to check his proofs, but he seems on the level ntuitively.

His concept of Classes is pretty fascinating, as is the Computational Theory of Life, The Universe and Everything.

Ah, where's Deep Thought when you need him?
posted by signal at 9:20 PM on October 29, 2001


A great resource:
The Santa Fe Intsitute
posted by signal at 9:22 PM on October 29, 2001


Here's a fairly good overview of Wolfram's ideas...and here is a previous mefi discussion on the matter.
posted by Kikkoman at 9:31 PM on October 29, 2001


Previous MeFi adventures in Wolfram: [1] [2]. The guy is an indefatigable hype artist, that's for sure.
posted by rodii at 9:35 PM on October 29, 2001


Whoops, timeslip!
posted by rodii at 9:36 PM on October 29, 2001


I want to believe.
posted by sudama at 10:34 PM on October 29, 2001


"I've come to believe," says Wolfram, "that natural selection is not all that important."

Great quote!
posted by signal at 10:43 PM on October 29, 2001


Notwithstanding this has been seen before on MeFi (though before my time — those discussions were before I was a member) one reason I post this is that the book is coming out soon: I've pre-ordered mine on Amazon for delivery in January. Yes, Doug, it's a link to an advertisement, but I'm really fascinated to see if any part of what he claims is true. And if it is, should make for some great conversation (and I mean in the real world — remember that?).
posted by paladin at 10:57 PM on October 29, 2001


Paladin: Like the tv show?
posted by Doug at 11:14 PM on October 29, 2001


I also believe that natural selection is not all that important (triple threat: Kauffman, Webster + Godwin, and (maybe) Kimura).

But, anyway, I urge you all to read the previous threads on this topic, mostly because I made a really hilarious comment in the first one that I feel is a little underappreciated.
posted by sylloge at 11:22 PM on October 29, 2001


Doug,

No, as in one of the 12 Peers of Charlemagne. I was into knighthood a little while back. : )
posted by paladin at 11:27 PM on October 29, 2001


extraordinary claims require extraordinary proof
posted by D.C. at 11:51 PM on October 29, 2001


I'll be happy to read this book, IF it ever actually comes out. I've only been reading about its imminent publication for TWO years now.

Of course, I'm also getting a little weary from having to contact Wolfram, Inc., for a new Mathematica password each time I do something major like daring to tweak a disk partition, or upgrade to a new version of Linux. Who has time to read Wolfram when you're on the phone to his inept customer support people all the time?
posted by fold_and_mutilate at 12:38 AM on October 30, 2001


The lattice of heaven? We'll see, if he ever lets the book go to print. I've played with CA, and to be honest (and perhaps to prove my own shortsightedness) I found it sort of boring, and curiously unsurprising. Whatever impact he hopes to have on the scientific community won't, I think, be shared by the lay community. I think there is an atavistic tendency to believe in a kind of ultimate reductionism. Its demonstration will merely confirm what has long been a suspicion, perhaps a wishful superstition, held by simplinistas the world over.

The guy really has a shitload of issues to resolve. Care to disprove relativity, anyone?. Hard to believe it can be done in one little book. And really, I'm surprised by the relative lack of hyperbole surrounding the whole thing. Sure, he's spinning it at light speed, and the occasional well-orchestrated, boy-reporter-meets-mad-scientist rendezvous doesn't hurt, but the few scientists who have some limited access to his work are not exactly going off like bottle rockets.

Regardless, the man is interesting, and I am very jealous of his freedom to pursue his very own white whale. May any of us who want it be granted a lifetime blue-sky endowment.

Where the hell is that spooky cadre of MeFi Masters? Surely some of you guys are up to the knees in this stuff.
posted by Opus Dark at 2:56 AM on October 30, 2001


Should be interesting, but then again I find paint drying interesting at times.. Oh, and Mathematica rocks, its password thing doesn't, grr... WPA style..
posted by Mossy at 3:14 AM on October 30, 2001


I've been hearing about this book for years, and every time the hype goes higher. "It will completely revolutionize our view of the universe!" "It'll completely change our notions of reality!" You can almost feel Stephen Wolfram spitting on you as he yells.

Will this book actually be as revolutionary as S. Wolfram wants us to think it is? I'd like to hear from people like Stephen Hawking, Roger Penrose, Richard Dawkins, et. al. before I rush out to buy a copy. Wolfram is a lot like John Forbes Nash -- brilliant, but erratic.
posted by mrmanley at 4:41 AM on October 30, 2001


Thanks for the link to the Forbes article, kikkoman -- I hadn't seen it (or the two earlier threads).

It's practically always frustrating, though, to read articles like that when they're written by people who don't know any mathematics themselves -- they never understand what it is. Everything Wolfram is doing is mathematics; if you can program a computer to derive the results of a process, then you can also describe that process mathematically.

As for evolution, clearly these mathematical constraints must inform our understanding of evolution, but how do they present an argument against it? So, it turns out there are some limits on the possible forms that individuals of a species might hold. The species still must evolve into whatever form it assumes, right? And it turns out that there are some very efficient rules for creating those forms. So isn't that an argument for evolution, that a species can ultimately utilize those rules?
posted by mattpfeff at 6:39 AM on October 30, 2001


more wolfram links

newscientist interview w/ james lipton -- Are you the Isaac Newton of the 21st century?

and a much better computers, science, and extraterrestrials: an interview with stephen wolfram from hal's legacy (mit press)
posted by kliuless at 8:12 AM on October 30, 2001


From the publisher's summary, Wolfram "made a series of discoveries which launched the field of complex systems research."

When the publisher has to resort to blatant lies to push the book, I say no thanks. I'll pass on this one. Did they get this "bio" from Wolfram himself?
posted by anewc2 at 12:44 PM on October 30, 2001


Probably. I think he's been slaving away on the book for the past ten years because he simply cannot get past his own sense of self-importance. Just take a look at this quote from the Q&A section of the website:
In the early 1980s I published some of the results from my early computer experiments, particularly ones on some systems called cellular automata. They created quite a stir, and over the last fifteen years they have led to many books and thousands of scientific papers. But I was never satisfied, for I had always thought that what I had discovered was just the beginning of something much bigger.
Well maybe that feeling was wrong, and people have understood the implications of CA just fine.

Really, what does he expect people to do? Toss out everything we know about physics, chemistry, and biology and instead attempt to describe all phenomena in terms of CA rules? Does he want statements like "say, this looks kinda-sorta like rule 136" to be the limit of our understanding of the world? Yeah, right.

He's so obssesed with finding a new, revolutionary way of doing things, he is attempting to forcibly create one where it doesn't really exist. I blame the fact that late 20th century education is so heavily steeped in the rhetoric of "paradigm shift"s, it has a tendency to produce people whose only purpose in life is an endless search for the "next big thing".
posted by Potsy at 5:08 AM on October 31, 2001


Completely in accord with what Postsy, matt and other say here, but: we shouldn't downplay Wolfram's work either. He did come up with those classes of CAs, and there are close ties to the "edge of chaos" work that Langton and others have done.

I wonder sometimes whether there's some jealousy involved. Way back, still fairly early in his career, Wolfram made the decision to go commercial and develop Mathematica. Mathematica hasn't taken over the world the way he hoped, but it's been pretty sucessful and I'm sure it's made him a rich man. But in doing that he essentially threw away his career as a scientist, and I'm sure it irks him to see all the attention paid to the SFI, Langton, Farmer, Kauffman et al. I'm sure he sits in his castle somewhere, Mad Thinker-style, and says "Those fools! It should be me they worship, not that tool Norman Packard! [mad laughter]" Thus the desperate self-aggrandizement of his book. But the fact is that, besides the good work they did, Langton, Farmer, Packard, Kauffman, and all the other pop heroes of complexity have interesting personal stories, which makes them good vehicles for pop science, and/or are big-picture visionaries; Wolfram is a grad-student-turned-mogul. He just doesn't make good copy, unless he makes these huge claims.

It's my impression that at root, Wolfram's gripe with modern science is that it uses differential equations instead of discrete math. Many of the achievements of physical science have involved ever-more sophisticated uses of, essentially, calculus for modelling natural phenomena. It's been known for a long time that this is a hard approach--no closed-form solutions of most DEs, etc.--and so people have increasingly been turning to purely numerical discrete approaches like Finite Element Analysis, which depend on huge amounts of computer power. This has been characterized as a shift in the way science is done, away from the Galilean style of modelling everything with equations.

It seems to me Wolfram is advocating returning to a style in which theoretical math is fundamental, only the math is discrete math rather than calculus. Whether CAs are the right path or not remains to be seen, but we know that it's in principle possible to model just about anything with a CA.

As to the idea that nature itself is discrete, and should be modelled in a discrete way, it's hardly new. It's most famously associated with the work of Edward Fredkin (although he hasn't published much, preferring to live on his private island) and the late Information Mechanics Group at MIT.
posted by rodii at 5:43 AM on October 31, 2001


It seems to me Wolfram is advocating returning to a style in which theoretical math is fundamental, only the math is discrete math rather than calculus.

Excellent point, rodii.

It's too bad everyone needs to make so much hay (pardon the horse pun) about this being some cosmic shift in the way to do science. The apparent fact that discrete mathematics has such strong explanatory power in areas where analysis becomes a very clumsy tool is very interesting -- and, accordingly, it seems very likely that the most elegant explanation of the nature of the world will employ some discrete mathematics. But, for all that, it's rather unlikely that it won't employ calculus, too.

So Wolfram is developing a new tool for the physicist's belt; that is excellent, and worth discussing in itself. But all this hype about it replacing all the other tools -- that's horseshit.
posted by mattpfeff at 6:01 AM on October 31, 2001


"Excellent point, rodii."

I want that on a t-shirt too.
posted by rodii at 5:17 PM on November 1, 2001


yes, I've been designing lots of t-shirts lately.

Ironically, I'm an unemployed copy writer. Sort of. (Too many job descrips to count, I guess.) Any of you guys willing to make a donation?
posted by mattpfeff at 6:42 PM on November 1, 2001


« Older StreamCast Networks (Morpheus, MusicCity) chooses ...  |  Battle Over Blocks... Newer »


This thread has been archived and is closed to new comments