opencog-dev team mailing list archive

["Trent Waddington" <trent.waddington@xxxxxxxxx>]

On Tue, Oct 28, 2008 at 3:54 PM, Ben Goertzel <ben@xxxxxxxxxxxx> wrote:
> I don't understand the distinction you're drawing...
> Joel's AA module was newly coded not ported ... and the forward chainer Joel
> built combined ported code with new code ... and both of these **actually do
> stuff**.  If you're curious, Joel can give you instructions as to how to run
> his Hopfield Net emulator (using AA) and his forward chainer.  I have seen
> them operate.  They produce meaningful output.

That'd be nice, yes.

> Also, the PetBrain code will also do stuff once fully ported into OpenCog:
> it will allow OpenCog to control virtual dogs in a virtual world.  It
> doesn't just need to compile, it needs to actually function, just like
> Linas's code...

Obviously I'm misinformed.  Both Linas and I have asked why the PLN
code is being developed on a separate branch and been told that it is
because it will break stuff in the core.  Except by Joel, who has told
us that he just doesn't want to give anyone the expectation that it is
"ready".  So the whole conversation has centered around people
breaking stuff.

Trent