Hi Radek -- I think your comments are excellent. I haven't had time to incorporate them in the doc, but I will this afternoon, after conversation. I also think that Guy's picture of the elements of the request are helpful. I attach them here to separate them from the larger document. My take is that we should now take each of the "objects" in the request and work out the details. For example, what is needed in the NSA field for trust, and when is trust info filled in; or what is an end point - is an STP always the end point or can it be something less specific like a port; or what is a path description - what is required and what is optional, and is a connection object as opposed to a topological path object required? I suggest getting the issues listed as best we can before trying to work out the answers. John On Aug 31, 2010, at 5:39 AM, Radek Krzywania wrote:

Hi all, My comments are in the attachment.

Best regards Radek

________________________________________________________________________ Radoslaw Krzywania Network Research and Development Poznan Supercomputing and radek.krzywania@man.poznan.pl Networking Center +48 61 850 25 26 http://www.man.poznan.pl ________________________________________________________________________

...

-----Original Message----- From: nsi-wg-bounces@ogf.org [mailto:nsi-wg-bounces@ogf.org] On Behalf Of John Vollbrecht Sent: Monday, August 30, 2010 9:29 PM To: NSI WG Subject: [Nsi-wg] updated doc showing NSI message fields

Attached is a doc that attempts to document the changes in the fields Guy had is the doc he sent last week. I think the fields are pretty close to what we talked about -- correct me if I am wrong.

I did add some definition to what is in each of the fields. These need more discussion, but I put something down to try to focus the discussion. Again, please suggest changes, prior to the call makes for more progress on the call.

John

<RequestConnection fields_RK.doc>