[alicebot-archcomm] Program D and ArchComm

[Noel Bush]

On Tue, 2002-05-21 at 00:34, Anne Kootstra wrote:
> One thing that has been bugging me for quite some time now is: "What part
> does the ArchComm play in the development of Program D?" This may be a
> touchy topic but its something that sometimes confusing me when I get the
> feeling that the topic on the table has little to do with the AIML
> specification and more with the interpreter.

Here's my version:

Specifications can be designed by committee, but designing software by
committee seems to be an ugly an inadvisable endeavor.  For a while,
Program D was standing as the "official" implementation for the
Foundation, simply because as the most actively used and developed
public codebase it provided the most reasonable basis upon which to try
to stabilize a specification for AIML.  But we've come a long way, and I
think that Paul Rydell's and Kim Sullivan's recent work has proven that
we've reached a point where this point is no longer a reason to give
Program D any special status.  In fact, we are in the process of adding
Program E and Program P to the Foundation CVS repository at this very
moment.

Anybody who wants to thumb their nose at Java (not that this is what
Paul and Kim are doing by any means) now has a specification they can
follow to implement an AIML interpreter in whatever language looks like
fun.  :-)  Seriously, there is no reason anymore to regard Program D as
the authoritative AIML interpreter.  There might be other reasons to
regard it as a good choice for a lot of purposes, such as those listed
by Paul Rydell in a recent mail to alicebot-general.