Arc is a new dialect of Lisp for EXPERT PROGRAMMERS, especially suited for web applications (faster than ruby on rails). Arc is being designed and implemented mainly by Paul Graham. The following resources are currently available:
Name:
Anonymous2007-08-07 2:10 ID:/zA+tLDV
Too bad development on this language is slow as fuck
Name:
Anonymous2007-08-07 2:33 ID:/zA+tLDV
Also, Paul Graham seems like a funny guy.
Name:
Anonymous2007-08-07 4:01 ID:ydu2+jNC
Arc is a blast from the past and it doesn't even exist yet.
Touring-completeness will not be implemented in Arc.
Name:
Anonymous2007-08-07 12:46 ID:Nzk0XuVR
>>5
well hes actually made billions of $ with lisp and written several books so I think you can't really call him that.
Name:
Anonymous2007-08-07 14:48 ID:6kkIgvcW
>>9
paul graham wrote yahoo's store software in lisp. when he left, they rewrote everything in C++. therefore, lisp is a great language to know if and only if you are paul graham.
Name:
Anonymous2007-08-07 17:33 ID:ydu2+jNC
>>9
Did you get them impression that smug Lisp weenies weren't completely right? That's why they grate so.
Name:
Anonymous2007-08-07 17:44 ID:Nzk0XuVR
>>10 >>11
thats just further evidence that pg is right.
There are plenty of examples of successful software systems in language X, Y and Z. How many are there for Lisp?
See, this is the problem with anecdotal evidence. So PG wrote it in Lisp. So what? I'd be more interested in the ratio of successful to unsuccessful projects, not some outlier developed by a good marketer.
Name:
Anonymous2007-08-07 21:46 ID:StJXpHLV
Also, I remember reading in one of his essays that he hired people right before Yahoo bought them so they looked like a bigger outfit. So Yahoo probably thought they were buying a development company, but of course the real talent probably left sometime after they became millionares. Clever, maybe, but hardly a sustainable way to do software development.
Of course software development goes better (in any language) if you only have geniuses, but realistically most companies can't depend on having only that type of talent (or rely on development methods that assume that). So having 5 pretty good programmers (of course bad programmers are a mistake) might often be better that having 1 brilliant programer, even if it is much more expensive and the result is not quite as good.
For example, if that brilliant programmer quits you're fucked, you don't have anyone left who knows the codebase and good luck finding someone else as good who wants to work on your boring enterprise product. What if they write code that works well, but is not easy to understand or doesn't have good documentation? What if they use some unpopular language that is very hard to find talent in?
Software doesn't exist in a vacuum. Sure there are many ways that standard practices could improve, but I don't think that "everyone should be as smart as I am" is a real solution.
Name:
Anonymous2007-08-08 7:02 ID:1nnRXS04
its not "geniuses"
its "Great hackers"
Name:
Anonymous2007-08-08 11:08 ID:J6HPy4PE
PAUL GRAHAM is doing it for the EPIC LULZ. He got that LISP in an TERRORIST accident, when he BLEW UP A VAN. TWICE.
>>18
Because this is currently the premier ``expert programmer'' community on the planet.
Name:
Anonymous2007-08-08 17:42 ID:SBloOOTu
>>18
This is the only programming board where you can bitch all day long without even knowing how to program. A nice excuse not to program at all and instead hang out here all day ;)
Name:
Anonymous2007-08-08 19:15 ID:1nnRXS04
>>22
its only you who doesn't know shit about coding
>>15
just hire people and make them learn the language. Ericsson did that with erlang, you just didn't find people out there knowing how to program in erlang. ericsson taught them.
Name:
Anonymous2007-08-09 4:05 ID:L2pJvHZk
>>23
Actually, I do, I've even achieved satori (which means you never have to code again).