Is the Sussman...

...coming up with a new programming language?
http://groups.csail.mit.edu/mac/users/gjs/propagators/
http://www.infoq.com/presentations/We-Really-Dont-Know-How-To-Compute
"I'm at the Strange Loop conference and Gerald Sussman said that all languages are obsolete but called Haskell "the most advanced obsolete language he knows". That's kind of a compliment I guess." - deech

>>40
The Mercury users archive had a maximum of 46 messages only in October. The rest months of this year have about 15 messages on average. The developer mailing list is much more popular with a maximum of 72 messages in September. That's seems pretty good for a not-so-well known language.
The last message in Clean is from May 7.

I am happy to help people that lack the ability to read mailing lists.

>>28
You and this harper dude are both ignorant though. I would have been more inclined to agree with you if I wasn't used to sbcl catching all of my type errors at compile time without needing a single type declaration. But whatever.

Mercury is a functional programming language. We are talking about propagator-oriented programming languages. Get that shit out of here.

check'em

>>42
That's because the Common Lisp type system can't express anything more abstract than an array of integers. Even the generic functions are runtime-dispatched, pfft.

Penetration-oriented

>>45
You are missing the point. I'm talking about what sbcl does, not what the common lisp standard demands. It's detecting these type errors by collecting constraints based upon how you use arguments to functions (in fact sbcl will trace type constraints as you step into if statements). You can build these constraints for all functions in a module and generate a signature for each one. Then check that the generated constraints are satisfied when a function from another module is called. For generating efficient code, you have the choice of using a single function that handles generic types, or multiple compilations of the same function that handle different type cases. These would be compiled upon demand when called from other modules. In order for this to scale, an intermediate representation of the function in it's generic form would be made available to the other modules. C++ solves this problem badly with templates.

_{When Harper was saying >>47 was not feasible, it sounded like he was talking about ocaml's type inference and its module system. Ocaml is shit. Need I say more?}

haskell is too high level, makes me feel restricted and closed off from the machine.

>>49
That's not necessarily a bad thing.

>>50
YES IT IS YOU FREEDOM HATING COMMIE
IF DOING ARITHMETIC IS TOO HARD FOR YOU THEN MAYBE YOU SHOULDN'T BE AROUND COMPUTERS HUH NIGGER

>>10
Depends on the particular type of abstraction. Abstraction that makes code (the binary, not the source) smaller also tends to make it faster, due to cache effects. The problem is when it's overused.

overuse my anus

>>1

You act like your system is NASA with a hundred users. I act like mine is a C64
in a kid's basement. I don't need protection. DOS had no protection. It's a
kid in his basement. Don't make it a fucken monstrocity with 20 fucken gear
semitrailer because you want to force the poor kid to do file permissions
because your an arrogant annal retentive nigger who sadistically likes to preach
elaborite cult man-made laws because you are a fucken pervert gay faggot. Gay
faggots want to make everyone dance around with penis dangling, for no reason
but it is the right thing for them to do. I have a gay brother, LOL.

So Sussman basically reinvented pipes. And he also failed to provide a single compelling example of propagators. Mental masturbation indeed.

>>55
Shut the fuck up, Cudder.

>>1
HAHA, LOOKS LIKE OL SUSSMAN'S FINALLY GONE SENILE

N TO THINK, YEARS AGO HE WAS TALKIN SHIT ABOUT HOW I WAS GOIN INSANE N ALL DAT BULLSHIT

WAT A FUCKIN MATH BOI RETOID

>>57
Don't worry, you've gone senile years ago.

>>58
No, that's not true for lambda-sama.

pipe my anus.

SEMEN PROPAGATOR

The Pleasure Of Being Cummed Inside

I propose that the Intentional Misuse of Literally (IMOL) be a new meme.

prison cupcakes

literally anus

literally dubs

This literally meme isn't funny; it's retarded. I understand it's a sarcastic take on how the retards don't know what "literally" means, but with its use we won't be able to know who's joking and who's retarded.
Please, stop!

>>67
Great, they won't stop now. Thank you for putting so much thought in your post.

>>67
Sure, it's literally bad, but is it literally Abelson bad?

>>61
Semen treats everyone to pastry with shit.

>>67

but with its use we won't be able to know who's joking and who's retarded

http://www.calgaryautism.com/characteristics.htm

Difficulty understanding jokes, figures of speech or sarcasm.

Maybe you should get checked.

>>67
Don't worry, we'll still know you're retarded.

>>67
Aren't you the guy who made the post that started this in the first place? You're literally the only one who cares. You're also literally wrong on the topic.

>>73
No.

>>74
Umena Noa?

>>75
Noa.

check 'em

>>1

Maybe his students. Sussman is too old to come with anything.