> ** Investigate possible TCP header compression scheme
> 
> ** Explain strategy for compressing RPL headers ( I understand this could be 
> done in the ROLL group, but I have not seen a definite statement either way )

Joseph,

I appreciate your comments.

I do have to point out that the work item on header compression this WG was 
chartered with has a defined scope:

> 2. Produce "6LoWPAN Improved Header Compression" to describe
> mechanisms to allow enhancements to the 6LoWPAN headers. Specifically
> this document will describe compression of addresses that are not
> link-local. Additionally this document may include other enhancements
> or optimizations of the HC1 or HC2 6LoWPAN headers.

The one feature that was explicitly called out was compressing global 
addresses, and I haven't heard any complaints on the way HC-06 addresses that 
item.
There are also "other enhancements or optimizations" in HC-06, in the rather 
subordinate sense this was meant here.

Having been part of the five-year effort that produced RFC4996, I'm pretty sure 
TCP header compression would be on a different timeline, and on a different 
scale altogether.  
It is way beyond what we had in mind when we discussed and agreed the above 
charter item.
(Yes, a less complex and less efficient form of header compression than RFC4996 
can be found, but this is still likely to be a couple years of effort.)

I also believe that RPL might very well address its header space problems in 
its own protocol -- in any case, with RPL still being in flux, covering RPL 
would also be a different timeline.

We have to deliver on our commitments before we start new work.
We are already late meeting our milestone:

> Aug 2008              Submit Improved Header Compression document to IESG for 
> consideration as a proposed standard

Now re:

> ** Resolve the "forward compatability" issue ( ideally, while maintaining 
> backwards-comptability )

I think this comment is the key to resolving the above issues -- can we discuss 
now how we would embrace, e.g., TCP header compression (or RPL compression if 
header size issues are not addressed by RPL), later?

Gruesse, Carsten

_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan

Reply via email to