There Was a Time before Mathematica (2013)

[scottlocklin]

Reduce, Axiom and Macsyma or the inferior "Maxima" anyway are still around and open source. Reduce and Axiom seem to have lost all open source momentum, which is a bloody shame as they were extremely interesting and powerful tools. I used Reduce in my own research, and remember finding it more powerful than contemporary Maple or Mathematica on certain classes of integrals.

I guess there are a number of others: https://en.wikipedia.org/wiki/List_of_open-source_software_f...

One missing from this list is Yacas, which was very impressive as it was basically a one man project written in the same way Wolfram wrote his thing: https://en.wikipedia.org/wiki/Yacas

Three that should not be forgotten:

1) Maple predated Mathematica, and was, as I recall easier to use. I think it survives as a standalone and as a Matlab package. It would be an interesting HBS case study why Maple doesn't have Wolfram's success or brand recognition. Wolfram is a weird guy (as others point out), but somehow was more successful than the humble Canadians.

2) Mathcad: in the early 90s, a lot of people used this instead of Mathematica or Maple because it was cheaper and it had a nice jupyter notebook like UI that didn't require you to learn a whole language

3) Derive: this one I really miss (it also predated Mathematica). It had a UI which was vastly better than Mathcad or any of Mathematica's later attempts at efficiency. You could crank out integrals or perturbation series on miniscule hardware; I ran it on my 186 HP100LX. And I believe it was internally based on a lisp engine, which gave lie to Wolfram's engineering choice that C/C++ was needed to make a tight CAS. This lives on somehow in TI calculators, but I'd actually consider buying a tablet or "smart phone" if there was a version of it for modern hardware.

FWIIW Wolfram is again worth a HBS study, as they're really the only first generation of "AI" company that survived and thrived. Computer algebra systems were considered AI back in the day, and the more general expert system shells were considered a viable path to strong AI (hint; it didn't work out) in the same way that "Deep Learning" is now (hint: it probably won't either).

[lispm]

Derive was based on muLISP

http://www.softwarepreservation.org/projects/LISP/other#muLI...

[enriquto]

notice that today you can still "apt install maxima" and it is a perfectly fine computer algebra system (the continuation of the Macsyma mentioned in the article)

[lelf]

And it works in the terminal (there’s GUI too)

    (%i48) integrate(2^x,x);
                                           x
                                          2
    (%o48)                              ------
                                        log(2)

[4thaccount]

Neat to see the history. I've heard there is more to the matter (think I've read a blog post before) on his issues with Macsyma and Lisp in general.

Anyone who runs Mathematica knows it is a pretty big system now. Lots of C and Java I believe.

[xvilka]

In my opinion mathematical software should be free, as a core of any science. Imagine Wolfram Software tomorrow dissolved and there are tons and tons of mathematical code for it. What we are supposed to do? And it is not a general purpose software, it requires a tremendous experience in the domain to be able to rebuild required piece. Just wasting of time.

[mmind]

Do you know if anyone was ever able to decrypt the SMP source code mentioned as a challenge in the article?

[philpem]

I was just wondering about that as I read it... though the source code wasn't for SMP, it was for the program he'd used to encrypt the SMP source.

Apparently it's "a version of crypt(1) with some parameters changed"... the question of course is which parameters... and of course, what the key is...

With the algorithm and key unknown, it'd be a pretty hard problem to solve.

EDIT: Looks like there are three crypt(1) variants.

- An exact implementation of the M-209 from V6 UNIX: https://minnie.tuhs.org/cgi-bin/utree.pl?file=V6/usr/source/...

- A single-rotor Enigma-style machine from V7 UNIX: https://minnie.tuhs.org/cgi-bin/utree.pl?file=V7/usr/src/cmd...

- A slight variation on the Enigma implementation: https://sourceforge.net/p/schillix-on/schillix-on/ci/default...

The M209 seems to be alphanumeric-only, and as the "source code" is binary, that'd rule it out.

Chances are, it's a butchered version of the Enigma algorithm.

Practically you'd need some known plaintext to recover the keystream. Even so, that wouldn't translate to the SMP source code unless you could find a weakness in the keystream generator.

The incoming password is hashed-and-salted by crypt(3) - so however long the password is, the "real" key will be 13 printable ASCII characters long (and the first two will be the salt).

[lqet]

Is it just me, or is almost every text written by Stephen Wolfram plagued by excessive, almost neurotic self-praise?

> Perhaps the first faint glimmering of an orientation toward something like Mathematica came when I was about 6 years old

> Looking back at its documentation, SMP was quite an impressive system, especially given that I was only 20 years old when I started designing it.

> My physics papers started containing all sorts of amazing formulas.

> I was 20 years old, and I’d just gotten my PhD in physics.

> I wrote lots of code for SMP myself (about 1000 lines/day). I did the design. And I wrote most of the documentation. I’d never managed a large project before. But somehow that part never seemed very difficult.

> [...]

I get that he is almost certainly a highly intelligent and gifted individual who has achieved a lot in his life, but humbleness seems to be a quality he is seriously lacking. Even if he tries to be humble, it just comes over as coquettish.

[Phemist]

This has already gone meta-meta, in that there is a thread in every comment section on a post by Stephen Wolfram that complains about his self-obsession, and also one that complains about the complaints.

Basically, there is an HN gentlemen's agreement to collectively ignore Wolfram's self-praise, so that discussions about the worthwhile nuggets of insight in his blog posts are not drowned out.

[chongli]

It's likely that he is a narcissist. If you find his writing intolerable, then don't read it. I don't.

[api]

Oh yeah, it's a well known personality tic of Wolfram to the point that it's a joke. He'd be full blown "iamverysmart" if he were not actually smart.

If he had a sense of humor about it and started knowingly self-parodying that would be awesome.

[ken]

I think you're reading too much into this.

For every one that refers to an age, I just see facts. There are many child prodigies, especially in math and physics, and I don't see it as a point of pride (or shame). He's not at all the same person he was 50 years ago, and he probably doesn't even remember who he's looking back at. The alternative, that he hasn't changed a bit in 50 years, would be even more damning of himself.

As for the other, I don't see how observing that your physics papers contained "amazing formulas" would be praise. Everyone's physics papers do. The field is full of amazing formulas! Perhaps I'm missing something, but he didn't even claim he'd done anything, here, except write them down, and observe that they could be modeled by a computer.

Reading any accomplished person write about their accomplishments will sound similarly hyperbolic, when you step back and look at the facts. It's funny to me how some people write about the crazy things they've done, and HN loves it (e.g., Feynman), and others do essentially the same, and HN always finds it necessary to comment about the apparent narcissism.

It's enough to discourage me from ever blogging. That's not a coin I'd ever want to flip.

[wavefunction]

That's his deal, unflagging hyperbolic self-promotion.

[safgasCVS]

I can see how it comes across like that but that wasn’t my reading. It seemed matter-of-factly. I think I would much prefer dealing with someone who’s openly arrogant versus someone virtue signaling their false modesty

Not saying I’d like to have him around for dinner though

[yardie]

You must be new to Stephen Wolfram.I figured this out from other people's experience around him and reading "A New Kind of Science." He's brilliant, he lets everyone know he's brilliant. All you can do is go along with it.

[petters]

I think you are right. I could not get past the introduction to A New Kind of Science.

[FeepingCreature]

There was a time after Mathematica too, and for most people it was very similar to the time before Mathematica.

[fifnir]

I Mathematica heavily using in any field?