[11210] in www-talk@info.cern.ch
Re: please vote on this draft now
daemon@ATHENA.MIT.EDU (James Salsman)
Fri Jul 17 14:00:14 1998
Date: Fri, 17 Jul 1998 13:57:03 -0400 (EDT)
Old-X-Envelope-From: www-talk-request@www10.w3.org Fri Jul 17 13:57:02 1998
From: James Salsman <james@veritas.com>
To: masinter@parc.xerox.com
Cc: www-talk@w3.org
Old-X-Envelope-To: www-talk
Resent-From: Gerald Oskoboiny <gerald@w3.org>
Resent-To: www-talk@w3.org
Larry,
Thank you for your message:
> # ...approval.
> I can't imagine what might have given you this impression.
You sent a message to Ed Tecot and I around November last
year wherein you suggested business-card scanning as another
application of the extension to your INPUT TYPE=FILE RFC
using the DEVICE=name_of_input_device attribute proposal
which is the centerpiece of the draft. Your suggestion was
incorporated, attributed, and the draft re-distributed, and
I haven't heard back from you until today.
> (1) is there a problem of sufficient scope to require a new spec?
The problems, examples of which are in the draft, are vast,
pressing, and do not require much more than a simple extension to
your own spec, as described in the draft. Most of the problems
concern language education.
> (2) does your proposal addresses the problem?
For asynchronous speech input, and many other forms of device
input via form submission, yes. Syncronous inputs are addressed
elsewhere (e.g., RTP) but there are no other asynchronous speech
form submission proposals to my knowledge.
> (3) is your proposal is technically complete, implementable
> as specified?
Yes: security concerns are treated in full and I have a Linux
Mozilla implementation of the core behavior. I'm open to any
improvements to the advanced details, but many people have already
given it their best effort. You seem unlikely to have suggested
the use of other devices without some faith in the suitability
of the draft, the base of which is yours anyway. Without your
promulgation of RFC 1867 your implementors would have probably been
much less coherent, and speech input-enhanced forms have been on
the HTML WG activity page goals statement for more than half a year.
> (4) is it downwardly interoperable?
Yes it is. <INPUT TYPE=AUDIO ...> is to be read as <INPUT
TYPE=FILE DEVICE=MICROPHONE ...> for instance.
I've read the process documents and I obviously need your
guidance because I have no idea what to do now. I'd like to get
some officials such as yourself lobbying for widespread
implementation so the speech server folks (e.g., Speech Machines;
http://www.cybertranscriber.com/ ) can start providing solutions
to the problems (1) which the draft is meant to address.
Ref: ftp://ftp.bovik.org/draft-salsman-www-device-upload-02.txt
Sincere regards,
James Salsman
