🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.electronics
Article #20011

Watson Hackers?

#20011
From: Rick Bensene
Date: Sun, 13 Nov 1988 17:37
53 lines
3260 bytes
Greetings,

I was wondering if anyone out there has taken to trying to figure out just
how their 'Watson' works.  In case you might be wondering just what a Watson
is, it is a board for IBM-PC which provides a TI 32020 DSP as a coprocessor
to the PC.  There is also a codec to do the digitizing, and all the
necessary interfaces to connect up to the phone line.  The watson
program sets the hardware up to act as a really fancy answering machine,
allowing remote control and 'voice mail' facilities through the use
of "Touch Tone (TM)" signalling.

While Watson works fantastic as an answering service, I'd like to augment
it in such a way that I can use it also as an interface to my home
control system so that I can call in and do things like remotely
change lighting, heating, and security system parameters via Touch Tone
commands.

Natural Microsystems, Watson's manufacturer, provides a package called "VBX"
for Watson that is essentially a library of subroutines which you can link
to to provide full user control for the system (I think that the Watson
software as supplied with the basic system is simply an application
linked to these libraries). However,  they want over $500 for the library
and they require Lattice C to make use of it, which is another couple
hundred bucks. While I can see the value of such an expense for a
company that wants to use the features of Watson for telemarketing
and data retrieval systems, I can't justify that kind of expense for
my homecontrol application.

I'm wondering if anyone out there has done any work on reverse
engineering the Watson code to attempt to find out more about how it
works.  I did spring for Natural Microsystem's $50 book describing
Watson's hardware from a programmers point of view (which bits in
which registers do what), but it is useless to me at this point
because really all it describes is the interfaces for the phone
line control stuff, and the interface to the Signal Processor CPU
on the Watson board, which is apparently loaded up with all kinds of
fancy DSP code when the Watson package is started up.

Hopefully someone out there it netland has some information or has
poked around at the innards of Watson's code enough to get an idea what
is going on.  If you have, I'd appreciate hearing from you.

Many thanks,

Rick Bensene
---
Rick Bensene        Tektronix, Inc.  PO Box 3500 C1-970    Vancouver, WA  98668 
..tektronix!tekigm2!rickb (work)         Voice: (206) 253-5489 (10A-5P Pacific)
..tektronix!tessi!bucket!rickb (home)   BBS:   (503) 254-0458 300/1200/2400
-- 
Rick Bensene        Tektronix, Inc.  PO Box 3500 C1-970    Vancouver, WA  98668 
..tektronix!tekigm2!rickb (work)         Voice: (206) 253-5489 (10A-5P Pacific)
..tektronix!teksce!bucket!rickb (home)   BBS:   (503) 254-0458 300/1200/2400
Technical Details
Message-ID: <1138@bucket.UUCP>
Path: arc1-ca.pugleaf.net!utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!husc6!mailrus!cornell!uw-beaver!tektronix!percival!bucket!rickb

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!husc6!mailrus!cornell!uw-beaver!tektronix!percival!bucket!rickb
From: rickb@bucket.UUCP (Rick Bensene)
Newsgroups: sci.electronics
Subject: Watson Hackers?
Keywords: Watson Speech Digitizing Answering_Machine
Message-ID: <1138@bucket.UUCP>
Date: 13 Nov 88 17:37:49 GMT
Organization: Rick's Home-Grown UNIX; Portland, OR.
Lines: 53