PROJ: Flare

From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Mon Jul 16 2001 - 12:38:53 MDT


Hello, all. This email is to announce that, although the Flare project
has not yet been started, there is now at least some preliminary material
intended to provide an idea of what Hello, all. This email is to announce
that, although the Flare project has not yet been started, there is now at
least some preliminary material intended to provide an idea of what Flare
is about. This material is intended for programmers only - it's informal
and quickly written, and is not supposed to be a complete introduction.
Its purpose is to let people decide whether they're interested in Flare,
and if so, how interested.

This material may be found at:

  http://singinst.org/flare/prog-overview.html
  http://singinst.org/flare/idioms.html
  http://singinst.org/flare/flarespeak.html

(For those of you who are wondering what "Flare" is, "Flare" is a proposal
for a new programming language that the Singularity Institute would like
to have on hand for AI, both because there are specific features required,
and because we need more powerful programming tools in general. In
dialects of LISP, both the program and the program data are represented as
lists. In Flare, the program, program data, and ideally the program
state, are all represented as well-formed XML. Flare would be run as an
open-source project.)

I'm posting this preliminary material now, instead of waiting until
something more complete is available, because at least one Singularitarian
programmer who lives here in Atlanta has expressed serious interest in
implementing Flare and acting as the leader of the open-source project.
We still don't have funding for Flare, but he's interested in volunteering
on a part-time basis and he has substantial hours open.
 
However, even though it's a very real possibility that Flare will be
"handed off" in the immediate future, I know that there are others out
there who have been waiting patiently for me to start writing things up.
No formal selection, or even partial commitment, has been made - if you
want to be the Larry Wall of Flare, or if you want to be in the running,
please contact me. You're seeing this new material at the same time as
the potential leader in question; nobody is being cut out of the loop.

I must admit that I view "being in Atlanta" as a significant factor.
Being able to physically meet will probably reduce the amount of writing I
need to do before handing off Flare. Still, if you have good
qualifications to head up an open-source programming language project and
you want the job, you'll be in the running - nobody has a head start and
no commitments have been made.

At this time, I'm also collecting some preliminary names for people who
think they might be interested in working on Flare during the initial
stages. (The potential project leader has worked with remote
collaborators before, and he says he thinks it would speed up the project,
even during the earliest stages.) Take a look at the referenced URLs and
see whether you think you'd be interested. Again, this does not represent
a commitment unless you want it to; we just want to make sure nobody who
wants in winds up being left out.
 
Current candidate languages for prototyping the interpreter are C++, Perl,
and Java, with C++ currently favored by the potential project leader. If
you have a strong feeling about language of initial implementation, and
especially if you feel that this is a make-or-break issue for whether you
can participate in the project - i.e., you don't know Perl or you aren't
willing to work in C++ - please let us know. Regardless of the choice for
prototyping, we will almost certainly want to drop into C++ eventually.

Again, I'd like to note for the record that we are *not* announcing that
the Flare project has been initiated, and that announcement may not arrive
for a while yet - if any problems develop with the current plans, you may
not be hearing that announcement for some time to come. We're just trying
to get a preliminary handle on what we can expect if things *do* move
quickly in the immediate future.

Sincerely,
Eliezer.

-- -- -- -- --
Eliezer S. Yudkowsky http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence



This archive was generated by hypermail 2b30 : Fri Oct 12 2001 - 14:39:48 MDT