Character recognition (was Re: Return...)
Alan Kay
Alan.Kay at disney.com
Sat Nov 4 00:23:08 UTC 2000
Duane --
At 1:46 PM -0800 11/3/00, Duane Maxwell wrote:
>Alan Kay writes:
>> BTW, Nathanial Shaerli spent the summer with us and did an
>>absolutely fantastic character recognizer that he also beautifully
>>integrated within morphic, and which includes a terrific UI for
>>understanding what the recognizer understands and for making changes,
>>new integrations, etc.
>> We plan to put this out when the current spate of big internal
>>changes to Squeak settles down in a month or so, but he might be
>>induced to send out his changeset earlier for those who are curious.
>
>Sorry to "poop in the punchbowl" again, but I've recently been reviewing
>the rulings in the Xerox v. Palm Computing case wherein Xerox attempted to
>assert its patent rights in single-stroke character recognition systems.
Do you have a reference to this patent (I wasn't aware of any patent).
>At first blush, and perhaps you can fill in more information about this for
>me, it would appear that the current Squeak recognizer would violate that
>patent.
Not possible. The current one was derived very directly from the
original GRAIL recognizer at RAND done by Gabe Groner in 1965 (and
written up in a RAND report in 1966). Sponsored by ARPA and free to
all (as it should be). It was the original and truly great single
stroke recognizer. I have a great movie of it from the late sixties.
Nathanial started with the current Squeak recognizer, but quickly
took a much different and more flexible approach to the matching
process.
The PARC Smalltalk recognizer has a line that runs pretty directly
from the original Ken Ledeen recognizer at Harvard (a student of
Ivan's, he did this recognizer in a very clever way because the
PDP-1 at Harvard was a much weaker computer than the RAND 360/40).
This recognizer algorithm is given in full in an appendix to
Newman & Sproull (edition 1). John Shoch adapted this twice for
Smalltalk-72. I adapted it for Smalltalk-76. Vickie Parrish did a
later version in LRG. Etc.
It is likely that any Xerox patents for a recognizer came later than
this, since we had the only one when I was there that I'm aware of,
and ours was not patented while I was there (nor should it have been).
>Palm won the case only because of some very technical, but
>apparently significant enough, differences and those are in turn covered by
>_their_ patent for Grafitti.
Patents are granted willy nilly these days. The patent office long
ago wimped out on trying to vet patents and have left matters up to
the courts (a terrible situation that gives rise to an infinity of
useless and usually meaningless disputes).
This is why there are a number of companies in Silicon Valley
that do nothing but demo prior art on old machines.
>
>On a related topic, we have a Squeak implementation of Ken Perlin's
>QuikWriting system, but are unable to use or distribute it because of its
>patent, currently held by NYU. It's quite elegant, though with a little
>bit steeper learning curve than glyph-based systems.
Well, maybe I should try to talk to Ken about all of us being able to
use it. (I'm personally not a big fan of QuikWriting, but it would be
fun to have in Squeak.)
Cheers,
Alan
More information about the Squeak-dev
mailing list
|