Excerpts from Tony Li on Fri, Apr 17, 2009 09:03:09PM -0700:
> More specifically, if we can converge on what properties we would
> like  identifiers to have, it might help to constrain the solution
> space.
>
>> First, there are a different uses for identifiers and each has
>> different requirements.  Before we can talk about properties for
>> "identifiers" in general we need to separate out which particular
>> uses we are talking about.  For example, are you including
>> discovery starting from FQDNs?  Kerberos identifiers for auth/auth?  
>
> I'm trying to stay independent of any particular use case and look
> at  the abstract properties that we'd like to have in a locator/ID
> split.

But identifiers need different properties for different purposes. 

> To help simplify the discussion (I hope ;-), I'll certainly stipulate  
> that every solution is certainly going to have a number of identifiers.  
> I think we're most interested in the identifiers that correspond to  
> locators.

Let's be clear what we're working on then.  It's properties of
identifiers for the things that will initially be contacted in order
to set up communication flows (probably aka stack IDs).  Yes?

After that we can look at properties of identifiers used for session
control.  They may be the same identifiers, as in HIP or ILNP, but
they need not be.  OK?
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to