CU-SeeMe for sign language transmission?

Dennis J. Streveler (
Mon, 29 Apr 1996 11:14:59 -0700

At 10:12 AM 4/29/96 -0500, you wrote:
>This is my first posting to this list. I am a statewide Director of
>Programs for the Deaf in a state education agency, and the CU-Seeme
>technology holds immense promise for this population, because deaf
>persons communicate through a visual medium (i.e. sign language),
>and because internet connectivity is spreading very rapidly across
>our school systems at the student, instructional and administrative
>In this context, the potential ramifications of CU-SeeMe technology
>for the deaf are clearly apparent.
>One of my responsibilites is to serve as a technology resource person
>for our division, and to facilitate statewide initiatives using
>communications technology. We are currently doing this via Web pages
>and email mailing lists, and it is my desire to stay absolutely
>up to date on the CU-SeeMe technology. That being said, I would like
>to present a couple of questions to the list:
>1. I keep hearing contradicting opinions about what kind of
>connection is required to implement CU-SeeMe effectively. Some say
>a regular 28.8b connection works fine. Others say that won't work,
>you really need ISDN. Still others say ISDN is marginal, that you
>need video cable connection. Others say it all depends on how many
>windows (i.e., users) are running in your CU-SeeMe session.
>Since school systems across the country (including mine!) are on
>extremely tight budgets, connectivity costs are extremely relevant.
>That's why I'm asking this question. There is a *lot* of difference,
>costwise, between a 28.8 and an ISDN connection, especially when
>you multiply it times a couple hundred school districts, many of
>which already have 28.8 connections up and running. And I won't
>even get into the costs of statewide video cable!

Actually, you might be surprised. For example, PacBell in California is
GIVING ISDN connections to each and every school in the state! So bug your
phone company, they might very well like to sponsor such an undertaking, at
least as a pilot project.

To answer your question, though, I would recommend that in order to get the
sign language nuances at sufficient frame rate (frames-per-second) that ISDN
would be the minimum bandwidth you could use, and then it might require that
your participants sign slowly. Obviously connecting one-on-one requires less
bandwidth than through a "reflector" (which you would need for a "group"
discussion. Are you mostly interested in one-on-one conversations?

>This is a critically important issue in this case, because it is
>important for the hand movements of deaf sign language to be
>acceptably smooth and recognizable, within cost restraints, if
>that's possible at all.
>I would sincerely appreciate any and all definitive information
>or other feedback on this topic. Many thanks!
>Ron Morgan

My field is telemedicine, and that is why I would be very interested in the
results of your experiments. I hope you will stay an active member of this
mailing list.


Dennis J. Streveler, Ph.D.
Systems Consultant
"Future Technologies in Medicine"
"International Software Development
"Human-Computer Interface Design
for Casual Users"
CIS: 71036,1645
415 239-1441
415 469-9476 fax
127 Lake Merced Hill
San Francisco CA 94132 U.S.A.

My job? To send the appropriate electrons hurtling around the globe.