Hacker News Reader

Comment by booleandilemma | Hacker News Reader

What’s HN’s opinion of Stephen Wolfram?

Is he legit?

Is the tome A New Kind of Science worth reading?

He was a child genius, known as a bit too egotistical today, but he has made genuine contributions to the field. NKOS has a nice narrative at least, and might be worth reading as a pop sci work, but isn't very scientifically noteworthy in itself.

I met him at a talk about a year ago, I would agree he is somewhat egotistical, but he didn’t come off as arrogant - which was interesting.

I think the answers are, essentially:

> Is he legit?

Yes - but he's amazingly full of himself and incredibly self-serving.

> Is the tome A New Kind of Science worth reading?

Not really, no.

He is the creator of Mathematica:

https://en.wikipedia.org/wiki/Wolfram_Mathematica

I would consider that legitimacy.

Mathematica is literally the worst PL I've ever used, especially since you effectively have to use their built-in editor that doesn't even have multiple undo. Of course, it is the least-worst symbolic system out there currently. But, it's not like lisp wouldn't have developed similar capabilities if Mathematica didn't come around.

You mean like Macsyma, now Maxima, where he probably got some early ideas from?

Mathematica (SMP) started out as a Macsyma clone. Here is a anecdote of how Wolfram went about this with his usual humility:

https://groups.google.com/d/msg/comp.lang.lisp/BUxXH76CYdc/I...

That's a more direct way of making my point ;-) I don't recall seeing that Pitman post before... good stuff. Interesting to see that Wolfram always had his <cough> confidence.

It's had multiple undo in notebooks (quite more than "an editor") for a while, and you can run it from a command line or use an IDE.

https://reference.wolfram.com/language/ref/program/wolframsc...

I mean, I wouldn't make a website with it, but for computational/mathematical code I'd choose Mathematica and its editor over a Lisp language any day, even if it's just for the syntax (though it most certainly is not). Some of us actually find it easier to understand and write math when it looks like math and not just a wall of parentheses or ASCII art.

That's why basically all Lisp systems in that area (Macsyma, Maxima, Axiom, REDUCE, ...) provide non-lisp syntax programming.

> Mathematica is literally the worst PL I've ever used

Have you used Maple? I had to for a General Relativity course, and far preferred Mathematica and its syntax.

I thought A New Kind of Science was wonderful. It's a deep dive into cellular automata by a very creative, systematic and productive mind. He doesn't succeed in remaking physics with these models, but he does succeed in demonstrating that the natural world is full of vast computational potential that emerges everywhere from simple interactions. It's a vision of an awesomely fecund ordinary reality.

I loved the book as well. I found it exceptionally well written, researched. The content was eye-opening and the author was . This is the first I notice that there's so much discontent with it. I'm not sure where it's coming from.

Edit: actually, I'm understating things. I found not only his central ideas very noteworthy, but how he managed to portray the historical development of mainstream science as largely random (as opposed to the, now silly, idea I may have had that today's science is somehow "optimal") was great and a positive influence on my outlooks, I'd say.

I think a lot of the discontent is that (reportedly, I have not read it) he seems to present the material as 1. largely his original work, and 2. revolutionary, while the ideas may be interesting and well presented, neither of those things are actually true.

> Is the tome A New Kind of Science worth reading?

The best thing about ANKOS were the footnotes. The footnotes/appendices could have made a good popular book on the history and relationship of computation and physics, except for all of the gag-reflex-inducing Mathematica self-promotion. Here is an example from the first footnote I clicked on just now:

"But in the so-called lambda calculus of Alonzo Church from around 1930 what were instead used were pure functions... of just the kind now familiar from Mathematica."

https://wolframscience.com/nks/notes-11-12--lambda-calculus/

The whole book is full of things like that.

He’s like a Mozart who thinks he’s better than Bach.

Lol good analogy, although Mozart had a sense of humor going for him

More like "A New Kind of Science".

on NKOS: http://bactra.org/reviews/wolfram/

Thank you! That's a great review - long, well-informed, savage yet trying to be fair; totally accurate as far as I could tell, although the criticisms seemed far weaker in places. Lots of good links too. ... 168 other long reviews by Cosma Shalizi on that site, and a lot of other stuff. I hadn't heard the name before, thanks again.

NKOS is probably the (or at least one of the) defining works on cellular automata and the thought around/behind them, at least when it was written. It will likely be as interesting as that sounds to you.

Well, he started Wolfram Research. You should try their products if you haven't. If you have, just try to imagine how you'd make something like Wolfram Mathematica or Wolfram Alpha.

Mathematica is good software (I posted on it here recently).

A New Kind of Science is part of where Wolfram gets his bad rap...just read the Amazon reviews to get an idea.

I liked a ANKOS, especially the chapter about physics.

NKoS is definitely worth reading especially if you have "boolean" in your username. And it's mostly pictures anyway.

Reply on news.ycombinator.com