[Sis-csi] RE: networking details

[Jane K. Marquart]

Not having e-mail since Thursday, I was able to read this thread 
back-to-back.
35 e-mails and 25 minutes later, I'm not sure what the original question 
was, what the answer was, or where we are right now.
My background, for those of you who don't know me, is Flight 
software.  I've worked many many years/missions, side-by-side with Project 
Mgrs, system engineers, hardware engineers, ops folks, etc., sometimes 
24/7, building and launching spacecraft.  I've had to convince PMs and 
engineers why "this is the best way" (including an onboard IP network for 
GPM).  I do not have the depth of expertise on network and IP protocols, 
etc., that most of you do.  Nor do I have any religious preference for IP 
or CCSDS.  Given that, if we want to produce something that may actually 
"fly" or "be accepted", my feeling for this WG is that we should get back 
to the types of discussions we had in Colorado Springs.  That is, what are 
the issues within the operational Cis-lunar scenarios that need to be 
resolved?  What are the protocol profiles?  What are the solutions and are 
they realistic and palatable to PMs or systems engineers, or today's ops 
personnel?  Base it on technical merit, which may have different solutions 
for different scenarios.  Do prototyping that supports the technical 
merit.  And START SIMPLY, because when it comes down to it, no PM that I 
know of will take the risk of implementing something that poses high risk 
or high cost to their mission.  Their hides will be in the canner.  Yes we 
need to be aware of a migration path, and future needs, but let's not get 
stuck in a long-term debate when we haven't even gotten out of the batter's 
box.  As an international WG, we should concentrate on analyzing the 
proposed (Cx) and existing (IP, IETF, CCSDS) networking protocols and 
figure out how the issues may be resolved where more than a single 
mission/program can benefit.  Cx is looking out for Cx, ESA concentrates on 
ESA, other NASA missions may look for single solutions.  Its 
understandable.  The hardest part of standards work is finding agreement 
across different interests, but isn't that our goal?  Finding solutions for 
CisLunar that might support the majority?

Just my 2 cents.....

Jane
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ccsds.org/pipermail/sis-csi/attachments/20070212/fe9ee8d8/attachment.html