> I forgot to ask: what would be a good programming interface to
> use? I mean for the visual module to communicate with the AGI
> module?


A protocol specification would probably be the best choice, followed by
a C language lib implementation of said protocol.  It is relatively
simple to bind C libs into just about any environment that you care to
e.g. into Python, C#, and similar.  Very portable.


> I can code in Visual C++ or C#, but I want to make the frontend
> the easiest to integrate with other programs (mainly for Windows).
> I think Linux and others will be handled separately.


Hmmm...  With one notable exception that I can think of, I believe most
AGI folks are using a Linux (or mostly compatible) environment to run
their systems on and I would recommend something that plays well with
the standard Unix tool chain.

AGI is a "big system" problem, and Linux is many years ahead of Windows
in its ability to scale well to very large shared memory and clustered
systems.  Linux also has a fairly mature tool chain that is already
well-adapted to massive systems.  That it is increasingly the de facto
operating system for supercomputing, both shared memory (a la Cray) and
clustered (a la Beowulf), there is quite a bit of momentum that probably
shouldn't be ignored.

I actually do most of my code hacking and testing on MacOS X, but for
"real" work the systems are Linux because it emerges as the clear choice
once scalability and performance becomes an issue, mostly because it is
the primary (and usually best) choice on larger hardware.  And because
MacOS X uses a Unix tool chain, the environment tends to be very
portable.  I think most serious AGI projects would have a hard time
using anything that requires a Windows tool chain, because most only use
a Unix tool chain for a lot of good reasons.  If you want the code to be
usable among a large number of projects, this is a serious consideration.

Which is why I suggested C and bindings...


j. andrew rogers


-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]

Reply via email to