ILC 2014
I counted about sixty people in the hall - really not very
many for an ILC where one might typically hope for twice that. I
didn't see anyone from the CL vendors, hardly anyone from any of the
other CL implementations, few people I knew from previous outings,
indeed really not many of the "usual suspects" at all. But note that some of
this works two ways: I could equally be upbeat and say there wasn't
much overlap between Montreal and the recent ELS in Paris: this
conference had netted a different audience, which is fine.
My only fault with the program was that there wasn't enough of it.
For an ILC there really should be enough material to fill four days.
At two and a half, this gig felt like one of the European local
meetings, translocated.
As for the presentations: the one that left me totally gaping in
amazement was the lightning talk from a 14 year old (sorry kid, I
didn't make a note of your name) who'd implemented his own lisp
dialect along with embedded image processing and all manner of other
bells and whistles. Of the named speakers, I should mention: Dave
Cooper's hands-on Gendl tutorial; Christian Queinnec's Small (but
Massive) Open Online Course; François-René Rideau's CL
scripting; Dave Penkler's amalgam of LISP 1.5 and APL\360; and Robert
Strandh's SICL implementation notes. All that time we spent building
bigger and better caches is now just history.
My own (co-)talk about
work on an NLP project went OK (I think). Judging by the questions,
the audience found Michael Young's social science side much more
interesting than my comments about the lisp, and I'm not going to
disagree with them.
So for two and a half days, some part of the lisp community had
come face-to-face. We presented and listened to the talks, we drank
the coffee and ate the food, we explored Montreal from below. And when
it was all over we all went away again, back to our separate corners.
I think the next "lisp community" meet will be European Lisp Symposium
2015, in London next April / May. I mention this "community" idea,
because it came up in the panel on the last day, and it's an
interesting thought to mull over: is there one? (or more than one?)
should there be one? and if you had it in front of you what would
you do with it?