Presence and IntComm WG

["Ben Chinowsky" <>]

*Action Items as of November 10*

[ACTION] Ben T. will recruit someone from Vocera to join the PIC calls and
explore partnership possibilities.
[ACTION] Candace will a) find out if there's any documentation of how location
information needs to be delivered to the PALS server, and b) look into what
would be involved in delivering location information from sources other than 
the
skiffs.
[ACTION] Jeremy will ask Deke if he still wants to work on the UA.
[ACTION] Jeremy will outline a high-level paper on the PIC WG's experiences 
and
lessons learned.
[ACTION] Ben T. will write a short document describing the motivation for the
paths-in-the-snow approach to PIC development.
[ACTION] By mid-November, Candace will produce an outline of the features that
Jamey has added to SER.
[ACTION] Jamey will send the group a note clarifying what code belongs to HP 
and
what's open source.
[ACTION] Jeremy will coordinate the production of a functional description of 
a
3G UA; all will send their thoughts on this to the PIC list.
[ACTION] Jamey will update the interface requirements document.
[ACTION] Jeremy will ask Paul Love to schedule a PIC plenary session at the 
Salt
Lake City Joint Techs.
[ACTION] All will send Jamey their thoughts on how HP should proceed with the
skiffs.

*Attendees*

Jeremy George (chair) - Yale
Deke Kassabian - Penn
Steve Blair - Penn
Dennis Baron - MIT
Ben Teitelbaum - Internet2
Joe Rork - Ford
Candace Holman - Harvard
Ben Chinowsky (scribe) - Internet2

*Discussion*

Jeremy opened the call with a review of PIC's recently-agreed-upon priorities:
continuing the trials (the next will probably be at the Spring MM),
versioning/packaging (which we want to do in time for the MM), UA development
(on hold for the next couple of weeks), publishing results from the trials so
far, and long-term trials (paths-in-the-snow engineering). Most of the
discussion was devoted to the last of these.

Jeremy suggested that the group begin by writing a comprehensive trial 
proposal,
then identify institutions to participate and shepherd them through the
necessary grant process. The group discussed some possible settings for
paths-in-the-snow engineering: classes, dormitories, and hospitals. The 
hospital
setting has the advantage of already having a clear need for presence
technologies. It has the disadvantage of liability issues for anything
experimental, despite the fact that PIC trials would exist side-by-side with,
not instead of, existing mission-critical systems. There was general agreement
that we don't want to restrict ourselves to just one type of setting for
paths-in-the-snow trials. Joe asked whether participants would need to have 
some
kind of affiliation with Internet2; this is an open question.

Steve noted that some hospitals in the Philadelphia area are doing a PIC-like
trial with the Vocera active badge (see http://www.vocera.com/). [ACTION] Ben 
T.
will recruit someone from Vocera to join the PIC calls and explore partnership
possibilities. Ben stressed that we'd want any such partnership to leave room
for continued experimentation on the server side. Steve suggested that PIC 
might
be able to help Vocera develop a more fine-grained way to control who can 
reach
you and when (right now there's just a "busy" option). Joe suggested we might 
be
able to help them add non-voice communication. Ben T. noted that the Vocera
device is hard to fully appreciate until you use it; in particular, its voice
recognition capabilities are impressive (e.g., say "Find me a doctor on the
third floor", and it will), and the sound quality is quite good.

The group discussed some specifics for the paths-in-the-snow trials:
- Candace said that we should include the network weather indicator.
- Ben T. noted the importance of "always best connected" (ABC) technology, in
which the device keeps track of the various link-layer possibilities and sets 
up
connectivity accordingly. Many prospective users assume we will do this.
- Jeremy asked if we should provide server boxes; there was general agreement
that we should not. Candace suggested we either use the Internet2 PALS server,
or have people install PALS on an existing local server. Ben T. suggested we
should maintain the core server code and build core releases, but have the 
trial
participants provide operational support.
- The skiffs are not the only possible way to report and aggregate location
information. There's a new DHCP recommendation for publishing location
information; this is part of Walt Magnusson's i911 work, and is being
incorporated into HP switches. Candace noted that Ta-Yi had called her 
attention
to wireless medical sensors that go on your finger. Ben T. noted that 
skiff-like
devices are commercially available, as are client-based things like GPS.
[ACTION] Candace will a) find out if there's any documentation of how location
information needs to be delivered to the PALS server, and b) look into what
would be involved in delivering location information from sources other than 
the
skiffs. Candace stressed the importance of keeping the option of manual 
location
reporting. There was general agreement that different means of location
reporting will likely be appropriate for different settings.

Joe observed that, given the range of possibilities, there is a danger that 
our
proposal could become too general. Jeremy suggested that we indeed start with 
a
general proposal, and later develop different versions for different
constituencies. Jeremy also suggested involving a sociologist to help us 
design
questionnaires on what people want from PIC. Candace reported that she'd had a
hard time explaining the motivation for the paths-in-the-snow trials to 
various
people she'd tried to interest at Harvard; she suggested that we need to 
provide
lists of options that people can pick from. Candace also suggested that we 
need
more marketing-like materials, such as slide presentations, emphasizing
functionality more than technology.

Ben T. suggested that we not get involved in doing grant proposals, instead
"going in with our good examples, our proposal template, our marketing
materials, our corporate partners, and our enthusiasm, and trying to get 
people
to take it from there." Ben also suggested that, for trials in the context of
EE or CS departments, we could encourage the participants to modify the code;
what they did with it would constitute their paths in the snow. Jeremy
suggested we pursue funding for PIC training workshops, which would include a
PIC cookbook as a takeaway. There is also a need to figure out what the 
overall
costs of doing a PIC trial would be.