[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [rohc] RE: [robhc] Meeting at IETF47
Khiem,
(sorry for the late reply, I've been out of reach for a couple of days.)
I agree that we do have the option of developing multiple compression schemes.
However, there are substantial benefits associated with reduding the options
available to implementors, in terms of easing interoperability. Therefore I would
prefer a somewhat unified solution if such a solution is possible.
I suspect that in the end this may boil down to weighing the goodness of possible
bandwith efficiency benefits against the badness of having several solutions which
do not fit into a unified framework.
I hope we can find a unified framework that combines the best parts of the various
proposals.
Micke.
At 06:52 PM 2/23/00 -0600, khiem.le@nokia.com wrote:
>Mikael,
>
>The latest charter states that "ROHC may develop multiple compression
>schemes, for example, some that
>are particularly suited to specific link layer technologies." I gather it
>means we are not constrained to a "one size fits all" approach. That
>addresses my concern, because I feel that to meet the stringent performance
>requirements and harsh environment of cellular links, we need to be able to
>consider deeply optimized solutions, tailored to the technology (error
>distribution, feedback channel or not, etc.). Not to mention future
>technologies, which may differ widely from existing ones. Of course, we
>would like to achieve maximum commonality among the solutions, but at this
>stage, it is premature to prejudge about the outcome and result.
>
>Khiem
>
>-----Original Message-----
>From: EXT Mikael Degermark [mailto:micke@CS.Arizona.EDU]
>Sent: Tuesday, February 22, 2000 6:53 PM
>To: khiem.le@nokia.com
>Cc: lars-erik.jonsson@ericsson.com; Basavaraj.Patil@nokia.com;
>robhc@cdt.luth.se; micke@optima.CS.Arizona.EDU
>Subject: RE: [robhc] Meeting at IETF47
>
>
>Khiem,
>
>It is important to have a good charter. One aspect of the charter is that it
>outlines
>what the rest of the IETF expects from us. So it is not entriely up to the
>WG
>(or the WG chair) what goes into the charter, that is something that the
>IESG
>ultimately decides. I don't have a problem with the current charter, if you
>do
>after reading this letter, please let me know.
>
>When it comes to unidirectionality, there was a pretty strong sentiment at
>the
>BOF that it would be preferable to have a scheme that can compress over
>unidirectional
>links. A number of important people expressed this opinion, and that is why
>that
>sentence is in the charter. I hope and also believe that it is possible to
>find a
>solution that works well over unidirectional links but possibly can provide
>improved
>performance when there is a feedback channel (i.e., link is bidirectional).
>
>I would think that it is in the requirements document that it is most
>important to
>have input from the cellular industry. I will publish a first draft of the
>requirements
>document in a couple of days, and do invite comments on that document.
>Please
>contribute. The basis of that document will be the 3GPP requirements.
>
>If you feel that other standardization bodies than 3GPP should participate,
>please
>ask appropriate people in those bodies to contact me (or Carsten Bormann,
>who the IESG has asked to be co-chair). They are welcome.
>
>Mikael Degermark
>
>>Hi,
>>
>>It is crucial to have the proper charter for this WG, if we want the output
>>of the work to be used or useful, especially to the cellular community.
>From
>>what I see and hear, I am not sure the various people and companies in the
>>cellular community have had enough opportunities to see this charter and
>>comment on it. I think we should have provisions to allow for adjustments
>to
>>the charter once we have in place the proper communications channels and
>>connections with organizations such as 3GPP, etc.
>
>>
>>In the meanwhile, I have two comments on this charter:
>>- "Ideally, it should be possible to compress over unidirectional links".
>>This statement should be qualified, as it may imply that a solution that
>>works over unidirectional links is ideal. That is not necessarily true,
>>because, when a system is bidirectional, the optimal solution may very well
>>be one that takes advantage of the bidirectional property. The WG should
>>provide a solution for unidirectional links, but that does not necessarily
>>mean that other solutions, applicable to the case of bidirectional links,
>>should be precluded.
>>
>>- "Creating more thorough requirements documents will be the first task of
>>the WG". We should add something like "Input from the cellular community
>and
>>standards organizations will be sought".
>>
>>Regards
>>
>>Khiem
>>
>>
>>-----Original Message-----
>>From: EXT Lars-Erik Jonsson [mailto:lars-erik.jonsson@ericsson.com]
>>Sent: Sunday, February 20, 2000 8:04 AM
>>To: Patil Basavaraj (NTC/Dallas); robhc@cdt.luth.se
>>Cc: micke@cs.arizona.edu
>>Subject: Re: [robhc] Meeting at IETF47
>>
>>
>>The IESG has reviewed the charter and we are awaiting the official
>>announcement.
>>I think it is Mikael Degermarks responsibility as chair to ensure that we
>>get a
>>time slot. but I'll check that with him immediately.
>>
>>/L-E
>>
>>
>>
>>
>>>
>>>What is the current status for ROBHC?
>>>Has the IESG approved the charter and the formation of this
>>>effort as a WG? Is an official announcement all that is awaited?
>>>I am confused by what you mean when you say that 'the group is
>>>"formed"'.
>>>
>>>You mention that the group will (and need to) meet. So far the
>>>agenda for IETF47 (draft) does not indicate a timeslot for ROBHC.
>>>
>>>-Basavaraj
>>>
>>>>
>>>>Yes, the group will and need to meet in Adelaide (otherwise we could not
>>>>possible have a protocol as PROPOSED in December). However, the WG has
>not
>>>been
>>>>officially announced yet but we will make sure that the group is "formed"
>>>and a
>>>>meeting slot allocated
>>>>
>>>>Regards!!
>>>>/Lars-Erik
>>>
>>>
>>>
>>>
>>>>
>>>>Hi,
>>>>
>>>>Is the ROBHC WG/BOF planning on meeting at IETF47? If
>>>>so has there been a slot allocated for it?
>>>>
>>>>-Basavaraj
>>>>---
>>>>Mailing list for Robust Header Compression WG
>>>>Archive: http://www.cdt.luth.se/robhc/
>>>---
>>>Mailing list for Robust Header Compression WG
>>>Archive: http://www.cdt.luth.se/robhc/
>>
>>---
>>Mailing list for Robust Header Compression WG
>>Archive: http://www.cdt.luth.se/robhc/
>>---
>>Mailing list for Robust Header Compression WG
>>Archive: http://www.cdt.luth.se/robhc/
>
>>
>---
>Mailing list for Robust Header Compression WG
>Archive: http://www.cdt.luth.se/rohc/
>