Re: [Geopriv] Short (5 page) ID to "update" RFC 3825

From: James M. Polk ^lt;>
Date: Fri Apr 24 2009 - 13:51:23 EDT

At 02:14 AM 4/24/2009, wrote:
> > Read my risk analysis email [1] on the trade-off between versioning
> > and not. My conclusion based on that analysis is that the addition
> > of a version field is not a good idea. Versioning forces rejection
> > of an option rather than allowing for the chance of a minor
> > misinterpretation. I invite further discussion on the issue.
>Notwithstanding that the datum most commonly in use (WGS84) is already
>covered, IMHO the same arguments would appear to prevent us from _ever_
>adding new datums to RFC 3825. e.g. "we can't add a new datum because the
>old client base doesn't know how to interpret it".

Ray -- this is true for every IETF protocol that uses anything that
isn't in the core spec. That's why there is an IANA registry, for
implementers to see what's available (and to prevent collisions ;-)


Geopriv mailing list
Received on Fri, 24 Apr 2009 12:51:23 -0500

This archive was generated by hypermail 2.1.8 : Fri Apr 24 2009 - 13:52:36 EDT