I wrote about this a while back (http://www.pchristensen.com/blog/articles/lisp-the-golden-age-isnt-coming-back-lets-welcome-a-bright-future/ http://www.pchrist

I wrote about this a while back (http://www.pchristensen.com/blog/articles/lisp-the-golden-age-isnt-coming-back-lets-welcome-a-bright-future/ http://www.pchristensen.com/blog/articles/lisp-the-golden-ag...) Basically, there are two Lisp eras - the big AI era and the new pg/Open Source era. The recent renewal of interest in programming languages, as well as publicity by writers like pg, Yegge, Raganwald, etc, have led more people to become interested in Lisp, but it's on an informal, person-by-person basis. Some of these people (like me) have chosen to stick with it and new Lisp communities have formed by gradual accretion. But this process has started later and been more distributed than other languages (PHP, Python, Ruby, Perl). In researching things and trying to figure out Lisp setup problems, I've noticed that there are fewer technical "How do I ..." questions now than a couple years ago but more usage questions. The technical problems are being solved and now more stuff "just works" - not completely but much more than in the recent past. I've noticed this for CLisp, SBCL, ASDF, and other tools. There's more documentation, more tutorials, etc. Now there are a couple young web frameworks (UCW, Weblocks), generally accepted libraries for common tasks, etc. Lisp is growing in popularity and is maybe 1-2 years away from being "tasteable" - ie you can try it out without a big hassle and a research project. I think that will be an inflection point, after which growth will come much faster.