Re: [dtn-interest] RG "last call" on draft-irtf-dtnrg-cbhe

["Eddy, Wesley M. (GRC-RCN0)[Verizon]" <[e-mail]>]

>-----Original Message-----
>From: [e-mail] 
>[mailto:[e-mail]] On 
>Behalf Of [e-mail]
>Sent: Tuesday, March 10, 2009 11:48 PM
>
>> #0, I guess I'm left wondering if this is the only spec 
>needed to implement a
>> CL for space DTNs? If so, cool. If not, would it be better 
>if it were?
>
>Stephen,
>
>weren't you in Minneapolis, where two very different 
>approaches to space DTNs
>and implementations, using two different convergence layers - 
>based on LTP
>(over CFDP over CCSDS) and based on Saratoga (over UDP/IP over HDLC) -
>were presented, based on actual in-space testing results?
>
>http://tools.ietf.org/html/draft-wood-dtnrg-saratoga
>
>The presentations there are a clear indication that more than 
>the cbhe spec
>is needed to implement a convergence layer for space.
>
>Perhaps you could look at finalising the meeting minutes 
>(draft not archived) or
>finalising the proceedings of the Minneapolis meeting (draft very hard
>to find) to refresh your memory.
>


I don't think this RG should be worrying about "space DTNs" and
trying to tailor for them ... there are other groups actively
doing that, and the discussion should be taken to them, in my
opinion.

I think the root problem here is just that "space DTNs" aren't
some homogenous class of DTNs.  The requirements for unmanned
near-Earth versus deep space are different, and when you factor
in the envisioned lunar scenarios with more ad-hoc connectivity
and richer options for infrastructure, the requirements and
assumptions are MUCH different.  It's like saying you're doing
something for "wireless links" without specifying if you mean
low-rate/low-power PAN, free-space optical, cellular, etc;
unproductively unspecific.

CBHE may be useful in the broader set of DTNs that contain links
with either constrained bandwidth or very short contact times,
which includes underwater/submarine (at least) in addition to
deep space.  Since it may be applicable in multiple domains, it
makes sense to me to publish it from DTNRG, but it will only
greatly confuse matters if we say we're doing it for "space DTNs";
instead focus on the properties of links/networks that it's an
optimization for.

---------------------------
Wes Eddy
Network & Systems Architect
Verizon FNS / NASA GRC
Office: [phone]
---------------------------