Funding/Supporting future Common Lisp infrastructure development
in [Lisp]
|
Prev: run-program and binary input
Next: Lisp and ncurses
From: Zach Beane on 18 Dec 2009 08:10 Elena <egarrulo(a)gmail.com> writes: > On 17 Dic, 17:32, avi <ant...(a)googlemail.com> wrote: >> What do you think whether such a bounty infrastructure would work for >> the Common Lisp community? > > No, it wouldn't work. Rich Hickey has built a community around > Clojure. No such community exists around any CL implementation. Clozure recently raised $20,000 for someone to hack on their IDE. http://ccl.clozure.com/blog/?p=28 Zach
From: Duane Rettig on 18 Dec 2009 12:51 On Dec 18, 8:39 am, Rainer Joswig <jos...(a)lisp.de> wrote: > Saying that Common Lisp does not have user communities > around implementations is completely wrong. Especially > since some of these communities keep on working since > two and a half decade and keep their favorite > Lisp branch alive. Agreed. [...] > The vendors of commercial Lisps are active, too. > Franz just had a meeting with japanese Lisp users: > http://jp.franz.com/base/seminar-2009-11-20.html Yes. I don't think the western world sees the impact that Lisp has had in the Japanese world. It's likely that it's a bit harder to do google research on this subject, because unless you know Japanese, your eyes glaze over when hitting a Japanese website. But they're there, in numbers. > Franz started before Common Lisp, the company > itself started in 1984. And even before that, there was a Franz Lisp (pre-Common Lisp) community, which was instigated in Berkeley and which involved those who would eventually start Franz Inc, as a business. At that time, there were so many 68020 machines with different operating systems on them, and Franz Lisp was so popular as being part of the BSD distribution, that requests kept coming in for a port to this or that operating system, and that made a good business case for putting together a company for that purpose. Duane
From: gary.schiltz on 18 Dec 2009 18:35 On Dec 18, 12:51 pm, Duane Rettig <du...(a)franz.com> wrote: > On Dec 18, 8:39 am, Rainer Joswig <jos...(a)lisp.de> wrote: > > Franz started before Common Lisp, the company > > itself started in 1984. > > And even before that, there was a Franz Lisp (pre-Common Lisp) > community, which was instigated in Berkeley and which involved those > who would eventually start Franz Inc, as a business. At that time, > there were so many 68020 machines with different operating systems on > them, and Franz Lisp was so popular as being part of the BSD > distribution, that requests kept coming in for a port to this or that > operating system, and that made a good business case for putting > together a company for that purpose. > > Duane Wow, thanks for the blast from the past about Franz Lisp. I wrote my first AI programs in Franz Lisp on a VAX 11/750 running BSD in 1983 (grad school AI classes at Kansas State University). I also used it to parse Sowa's Conceptual graphs and store them in a frame system, as part of my thesis work. I didn't have a reference manual, but managed to borrow a pre-publication copy of Wilensky's Lispcraft from my major professor. This was before CL and SLIME, so it was a matter of switching between 'vi' and 'lisp' on a VT100 dumb terminal with ^Z. And then writing the thesis in nroff and printing it on an impact- wheel printer... Amazing what a 32MB machine would do, especially when sharing it with 20 graduate students and a professor working on tweaking a Simula compiler. Man, are we spoiled these days :-)
From: verec on 18 Dec 2009 19:51 On 2009-12-17 17:32:56 +0000, avi <antvid(a)googlemail.com> said: [...] > Many of you will have noticed the call for funding Rich Hickey made > over at the Clojure group a few days ago. [...] > As a Lisp newbie, I would like such central incentive infrastructure > also to exist for the free software community of Common Lisp. This is just not going to happen. Rich Hickey has a story. And a vision. He doesn't disagree with himself as the single Clojure outside voice. CLers keep moaning, without vision but only envy and disdain, lust for a glorious past forever dead. Clojure challenges the status quo (state vs identity vs value) CL *is* the status quo. Clojure offers hope, and motivates towards new frontiers (working, scalable parallelism) shunning OO and mutable state, while CL keeps grinding the same old axes (aspects? closer to mop? cells?) People remember Lisp, the 7 primitives, code is data is code, the ying-yang "eval/apply of page 13", not the IBM 704 (or whatever that was, who cares?) that gave us the ugly car/cdr legacy. The IBM 704 is as relevant to McCarthy's Lisp as the JVM is relevant to Clojure today: not. Or not much. What is relevant is the new directions that Clojure sets for us, while, finally, pushing to the forefront the essence of Lisp (as above). Rich Hickey didn't invent persistent data structure anymore than McCarthy did invent Church lamdda calculus. McCarthy was the catalyst that made code as data and "pure" function, legitimising recursion. Hickey is the catalyst who bridges pure functional (well understood) and mutable state (much less well understood) in a compelling way. Clojure brings *questions* to the table. CL only has *answers*[1] People don't pay (donate) for what they already have. People pay for what they want but don't have yet. They pay for "dreams", for projections of themselves in the future, not for deeds commited in the past and long (and best) forgotten. CL is (still!) dead! Long live Clojure! (though it is probably better to be as dead as CL than as stillborn as Arc ... did he say willing to increase the number of his ennemies :-) -- JFB [1] My paraphrasing of a quote attributed to Picasso.
From: Jochen Schmidt on 20 Dec 2009 07:09
On 19 Dez., 01:51, verec <ve...(a)mac.com> wrote: > On 2009-12-17 17:32:56 +0000, avi <ant...(a)googlemail.com> said: > > [...] > > > Many of you will have noticed the call for funding Rich Hickey made > > over at the Clojure group a few days ago. > [...] > > As a Lisp newbie, I would like such central incentive infrastructure > > also to exist for the free software community of Common Lisp. > > This is just not going to happen. > > Rich Hickey has a story. And a vision. He doesn't > disagree with himself as the single Clojure outside > voice. He has visions? I hold it with Helmut Schmidt (Chancelor of Germany 1974-1982) "Wer Visionen hat, sollte zum Arzt gehen." (translation by me) "(anyone) who has visions should go to a doctor". ;-) > CLers keep moaning, without vision but only envy and > disdain, lust for a glorious past forever dead. Huh? Not my impression. > Clojure challenges the status quo (state vs identity > vs value) CL *is* the status quo. > Clojure offers hope, and motivates towards new frontiers > (working, scalable parallelism) shunning OO and mutable > state, while CL keeps grinding the same old axes (aspects? > closer to mop? cells?) Clojures pure, immutable functional programming, persistent datastructures, STM, abstraction of state and alternative approaches to concurrent programming is not really new. Haskell had it for years now. What Haskell really lacks is a solid implementation running on the JVM (or even CLR). This is what Clojure offers for Haskellers and others now to some degree. Thats cool! But I don't see why Clojure and Common Lisp has to be mutual exclusive. Sometimes I get the feeling that all those trolls moaning around on c.l.l (and creating nothing) are now over at clojure. I don't care were they are - they do not make a change to anything. In some years there will be a new place were they go, Clojure will be called dead by them but will really be a solid tool for those who actually do something. > People remember Lisp, the 7 primitives, code is data is code, > the ying-yang "eval/apply of page 13", not the IBM 704 (or whatever > that was, who cares?) that gave us the ugly car/cdr legacy. Seriously: Was that really a problem for you in the past? You can implement your own datastructures so easily in CL. I've implemented Clojures persistent vectors in CL in less than 1 hour - it isn't really rocket science, you know. There even are ready made libraries for things like this (fset for example). > The IBM 704 is as relevant to McCarthy's Lisp as the JVM is > relevant to Clojure today: not. Or not much. Do you really want to argue that the functions CAR and CDR are something like coding IBM 704 assembler? Its an abstraction, nothing else. If that *would* be the only thing Rich Hickey did with Clojure it wouldn't be interesting at all. > What is relevant is the new directions that Clojure sets for > us, while, finally, pushing to the forefront the essence of > Lisp (as above). Who is "us"? > Rich Hickey didn't invent persistent data structure anymore > than McCarthy did invent Church lamdda calculus. So what? > McCarthy was the catalyst that made code as data and "pure" > function, legitimising recursion. > > Hickey is the catalyst who bridges pure functional (well > understood) and mutable state (much less well understood) > in a compelling way. I disagree - Haskell actually did this before, and it is not surprising that Clojure got alot of mindset from Haskell(ers). Clojure is a pragmatic tool. It *has* good ideas and it is on its way to be a solid tool. I don't think that Haskell or Common Lisp or any other language has to "die" for this to happen. > Clojure brings *questions* to the table. CL only has *answers*[1] didn't get it (sorry). > People don't pay (donate) for what they already have. > People pay for what they want but don't have yet. > They pay for "dreams", for projections of themselves in > the future, not for deeds commited in the past and long > (and best) forgotten. Well, people pay for Clozure CL funding and its thriving really well. I personally support Lispworks because they actually do a really cool job delivering a rich Lisp IDE and a solid compiler and runtime. They also recognized concurrent programming as an important topic and they offer solutions and the tools to build your own, if one is still not happy. In the most practical sense: Clojure is no viable alternative to Lispworks for me - absolutely not. > CL is (still!) dead! Long live Clojure! > (though it is probably better to be as dead as CL than > as stillborn as Arc ... did he say willing to increase the > number of his ennemies :-) The Clojure of today is the dead clojure of tomorrow. Its dying every day ;-) Its changing so quickly that its sometimes difficult to see what language it actually is. I actually think it has a lot of cool ideas and we will see how much of them survive long enough to get taken over to other languages (including CL). CL isn't standing still. I'm really thrilled by the efforts of so many creators in the Common Lisp community. I'm in there since 10 years now - and I can tell you it was a very different world for CL in 1999.(Sadly, like Haskell today - and I really wish Haskell to thrive too!). ciao, Jochen |