Re: [dtn-interest] IANA registry for LTP (and the BP?)

[Stephen Farrell <[e-mail]>]

Peter Lovell wrote:
> Hi Stephen,
> 
> I think that having a registry is a good idea although I think that the
> ciphersuite details are not quite firm enough yet. 

I guess you mean for the BSP. We should probably add some IANA stuff
in the next rev of the BSP spec I guess, just to get the ball rolling.
Will you add that to your list? You can use whatever I draft for the
LTP stuff as a template.

The LTP ciphersuites are cooked sufficiently to go ahead with a
registry now.

> "Good plan - go for it"

Thanks,
S.

> 
> Cheers.....Peter
> 
> 
> 
> On Wed, May 7, 2008, Stephen Farrell <[e-mail]> wrote:
> 
>> Ok, so that got a stunning response - guess registries aren't
>> what get people here excited, which is fair enough:-)
>>
>> But the IANA folks've now asked as well, so we need some
>> kind of answer.
>>
>> Assuming no further input, I'm going to propose that we
>> ask IANA to set up LTP registries for the extension types and
>> ciphersuites, with the update rule for both being "specification
>> required" where the specification can be an RFC (of any type)
>> or any other SDO's publicly available specification (wording
>> for that might be tricky so we might get reduced to needing
>> an RFC).
>>
>> I'll ask them to do that tomorrow, so yell now if it's a problem.
>> We can address the BP registry issues later.
>>
>> Stephen.
>>
>> Stephen Farrell wrote:
>>> The comment we got about LTP extensions was:
>>>
>>>  > An IANA registry for LTP extension values seems appropriate. Since
>>>  > [LTPSPEC] does not establish one, it should probably happen here to
>>>  > assign the two values used here and have a place to register more.
>>>  >
>>>  > Since the extension space is small, I recommend expert review as the
>>>  > policy for new assignments.
>>>
>>> I think (anyone know for sure?) that the relevant guidelines here
>>> are BCP 26. [1]
>>>
>>> So the questions are whether or not to do this for LTP extensions
>>> (and maybe ciphersuites) and secondly for extensible bits of the
>>> BP (block types, ciphersuites, anything else?).
>>>
>>> And if we do want any IANA registries, then what rules should
>>> we adopt for updates?
>>>
>>> I guess there is the possibility that CCSDS's DTN group might
>>> want to extend LTP and/or the BP, so maybe it is a good idea
>>> for us to think about this now. If anyone knows of other groups
>>> that might want to create their own extensions or blocks that'd
>>> be useful input here too.
>>>
>>> As I said in the other mail, my plan for LTP would be to create
>>> a new draft-irtf-dtnrg-ltp-iana to document whatever it is we
>>> think is the right thing to do. (If anyone wants to take on
>>> doing that, let Kevin and I know.)
>>>
>>> Thoughts?
>>>
>>> Stephen.
>>>
>>> [1] http://tools.ietf.org/html/bcp26
>>>
>>>
>>> _______________________________________________
>>> dtn-interest mailing list
>>> [e-mail]
>>> http://maillists.intel-research.net/mailman/listinfo/dtn-interest
>>>
>> _______________________________________________
>> dtn-interest mailing list
>> [e-mail]
>> http://maillists.intel-research.net/mailman/listinfo/dtn-interest
> 
> 
>