I think the community that drifts from CL towards Clojure are those that deeply appreciate the JVM. Personally, I would have preferred Richs efforts to have taken place in SBCL, or one of the existing CL implementations. It definitely would have improved the state of Lisp VMs / Lisp libraries. Paul Graham also could have gone there – but I suppose his changes are purely syntactical anyway.

Its funny because five years ago I had a lecturer who was addicted to C#, and when we spoke about Lisp he called it a Toy LanguageTM. His next question, which pushed his previous statement into the land of questionability, was: “But does it support interfaces?”.

LISP IS A MYTH! Everything I have heard about Lisp is just a myth and no Lisp Development Environment (LDE?) exists out there; no stable enough implementation; no real success story; it is just PG and copies of that! After 50 years if really Lisp has anything “pragmatic” to offer, I (for one) should not programming in C# (Or Java, PHP, …) by now.

(
And I know: every cool idea is a Lisp idea and all people out there are thieves! It is just there is no stable Lisp implementation of the idea and Lisp was so busy with being elegant that could not waste time with being pragmatic.
)

Note: Maybe it never happens to me to develop in a dynamic language. But this harsh rant on Lisp (and it’s community) is an extremely obsessive and positive one. So to Lisp community: Clojure is not a toy language; it is a usable Lisp and that’s you do not like about it.

From Paul Grahams Arc tutorial. So he’s abandoned seriousness as well then?

As such I suspect that your notion of seriousness will be misunderstood (certainly one can write business oriented, “serious” code in clojure or jruby or whatnot). More importantly it’s not even common lisp specific– it has to do with the priorities of the “free market” instant gratification globalized society that we live in.

What i mean is, before the 1.0 relase all things should be tried, after that caution should prevail.

To be considered serious by *you*, maybe–but that is an incredibly narrow-minded definition of “serious”.

Self-hosting in and of itself has no direcly practical goals, but it ensures that language is expressive enough to hold its own definition, and that language’s definition doesn’t implicitly depend on other language’s semantics (e.g. Clojure depends heavily on Java (the language, not the VM) interfaces behaviour). Dependence on single language together with being implementation defined leads to unclear behaviour, where parts of semantics, esp. language core semantics, come from wholly different language.

@Luc Préfontaine

Congrats on putting Clojure in production! Seriously. Please, read my explanation above: I didn’t intend to dismiss Clojure, and I think other examples of languages I called “toy” (Perl, Tcl, TeX) should tell that clear enough. It doesn’t mean it’s unsuitable for production, or anything; it only means that convenience of single, Java-based implementation has its price. Simple trade-off, if Clojure (or any other tool) solves your problem, good for you. I just think Clojure has potential to actually become a hundred-year language, for which it would have to grow out of implementation-defined functional programming layer on top of Java.

@Bill Robertson

I used CL in my job: I hacked Embeddable CL (http://ecls.sf.net/) to be used in embedded Linux boxes, and later I developed and maintained UnCommon Web based (http://common-lisp.net/project/ucw/) middleware web-app, contributing to UCW itself. I also completed a Google Summer of Code project in CL: http://common-lisp.net/project/cl-openid/. At the moment, I don’t use Lisp professionally (only hobbyist/university stuff, mostly Web: Hunchentoot/Postmodern, with a vision of completing my own toolbox-framework on top of those), but I plan to go back to using Lisp (CL or Clojure, whichever is more handy for the problem at hand) as soon as possible.

@Tom Faulhaber

To get things straight first, you seem to be mistaking me with Maciej Cegłowski of idlewords.com, who wrote the “Controlled tango into terrain” post. I have the same name, but that’s all to it. The post, however, seems interesting (didn’t finish reading it yet), thanks for the link. :)

I think you’re right, words I used may seem pejorative. Too much Usenet flame wars, I think, I got used to phrase myself in a direct way, which may be misread. The “toy” word was just first that I had on my mind, with no offense intended. And for “parasite” thing, the intended meaning was not pejorative: being a parasite is just an evolutionary strategy, and since we’re talking about languages/libraries/platforms ‘ecosystem’, it seemed natural way to say what I meant: a strategy as good as any other, having its merits and downsides, and having a potential to evolve into a commensalism, mutualism, or to grow out of strict dependence (which is what I hope will happen, Clojure is too good a language to stay hard-wired to a single host).

A side note on parasite thing: when I was taking a university course in molecular evolution, professor said in one of the lectures that in these days cats can be considered parasites feeding off humans. Not literally eating people, of course, but cats held as pets provide no actual value besides being cute, yet we feed them, and keep them warm and healthy. Cats co-evolved with humans and some of them successfully employ this kind of strategy, and it doesn’t seem to me there’s anything wrong with it. I personally find this view fascinating.