KIF would be a highly practical lingua franca
Lojban would work fine too
I agree that using English to interface btw modules of an AGI system
seems suboptimal...
I am glad that the different components of my brain don't need to
communicate using English ;-_)
Ben
Jey Kottalam wrote:
On 3/20/07, David Clark <[EMAIL PROTECTED]> wrote:
My proposal certainly allows that. If sockets and some form of standard
English is used to communicate between the different systems, then any
language should work. If you want to directly use major chunks of
code that
others have written within a whole set of your own code, then you
will have
to have compatible interfaces and work from a common language
(whether that
is mine or not.)
Why do you choose English as the lingua-franca amongst modules? Even
if you want to use natural language, English is a particularly messy
and internally-inconsistent natural language. How about lojban? Or why
use natural language at all, as opposed to statements in first-order
logic, or semantic nets, or some other machine representation?
-Jey Kottalam
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?list_id=303
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?list_id=303
