VoIP WG Meeting at Fall Internet2 Meeting in Atlanta...

[Harpal Chohan <>]

Greetings VoIP WG members!

I have received 12 responses to my earlier e-mail
soliciting "initial" participants in the Internet2
VoIP test-bed. Those, along with a handful that
had previously committed, lead me to peg the initial
size of the test-bed to span 15 to 20 member campuses. 

This is a good starting point. As you may recall from
my previous message, the membership of this list is
over 70, but not all are active players at this time.
I propose that as the active participants proceed to
meet, work, and put the test-bed together, we keep
the list at large informed and involved in the 
design issues of the framework we put together. 

Now on to business...

At the Minneapolis NLANR BOF in May, there was a
general consensus that H.323 could not be avoided
if we are to put this thing together. I realize there
are pockets where the sentiment leans towards a 
pure SIP implementation "locally", but if we are
to tie all our diverse installations into a functioning
test-bed, H.323 seems to be the common glue that is available 
at this time - mainly due to widespread implementation.

This is *not* to say that other flavors of local implementation
are to be excluded. One of the primary goals of the test-bed is to 
tie in different local implementations so we can all inter-operate;
through gateways - both for media and signaling. To this end,
we'll also have the opportunity to bring inter-operability testing
out of participants' local labs, and give it a real-life deployment
test on the test-bed.

To get started, the immediate issues that "initial" participants 
need to hash out appear to be:

1. Take stock of what the "initial" participant group has:- IP Phone
   installations of what flavor, technology in use in local
   installations, gateway abilities, type of trunk access to local PBXs,
   and hardware in use.

2. Agree on a least common denominator for the test-bed framework.
   E.g., H.323 for the inter-domain glue, gateways for local variations.
   Policy on gateway use, placement and hierarchy of gatekeepers, etc.

3. Numbering plan:- I list this separately because it looks like this
   is going to be a bear. There are those with IP phone installations 
   that do not have real E.164 numbers, and then there are those that do.
   Also, when tying "regular" campus phone users in through
   VoIP gateways, the "regular" phone user may have no knowledge
   of VoIP or special dialing requirements. It's clear that we
   can't exclude the E.164 numbering hierarchy, but also have to
   accommodate the purely IP population that may not have that kind 
   of addressing. Hammering out a coherent numbering plan appears to
   be a prerequisite to getting any kind of inter-domain test-bed
   off the ground. 

We have an opportunity to attack these issues at the Fall Internet2
member meeting in Atlanta. There is a VoIP WG meeting scheduled for
1:30 - 5:00 pm on Wednesday, November 1st. I've reserved the 3.5 hr
block of time in case we need it, which I suspect we will, to
to hash out items 2 and 3 above.

This will be a closed Working Group meeting. The room assigned to
us can accommodate 30 people - enough to accommodate the
"initial" test-bed participants and interested contributing
members. Please e-mail me your intention to attend so that I may
draft up an invitation list in case that becomes necessary.

Also, please feel free to comment on the issues discussed above,
so that any glaring oversights may be accommodated before the
meeting agenda is finalized.

Cheers...

-h
---
Harpal Chohan
The Ohio State University

---------------------------------------------------------------wg-voip-+
For list utilities, archives, unsubscription, etc. please visit the
ListProc web interface at 

    http://mail.internet2.edu/

---------------------------------------------------------------wg-voip--