Re: draft-gregorio-07.html

[Steve Jenson <sjenson@xxxxxxxxxx>]

On Thu Aug 07 11:07:06 PDT 2003, Daniel Berlinger <daniel@xxxxxxxxxxxxxx> wrote:
> > I went back and re-read and there is precious little
> > to quote on the wiki. I also realized some feedback
> > was via private email, which I will not quote.
>
> That was exactly the back channel sort of thing I thought you were trying to avoid, and which I feared. I specifically avoided saying anything about my thoughts in the hopes that it would not be the case.
>
> > So.
> >
> > Let me change what I stated earlier about 'slight preference' to be:
> >
> > 1. There is no consensus on RSD vs Introspection.
> > 2. I made a judgement call.
> > 3. If a consensus does emerge I will change the RFC to reflect that consenus.
> >
> > I would really like to see a technical discussion
> > erupt from here.
>
> Yeah, me to. Except why should anyone bother? You've made a judgment call, the unnamed folks have what they want, and the tools guys pick up the cost.
>
> I now have to push uphill against an unknown group of folks rather simply ask why a new format is required, when there is a supported format used by this community?
>
> Want to flatten the field again Joe? Pull the Introspection section from the RFC, with a "To Be Decided" and then let's see where things stand.
>
> Yuck.


My understanding of RSD (having read the spec and having implemented
it for Blogger) and my understanding of the proposed Atom
introspection leads me to believe that the two intersect in one major
way: the entry point for the api. RSD specifies a single entry point,
and Atom can specify multiple entry points. Also, I had thought (maybe
I was dreaming) that the Atom introspection would be expanded to give
details about the extensions those entry-points support.

Given that RSD can be extended via XML Namespaces (thank you,
Daniel!), it seems to make sense to try and move the proposed Atom API
Introspection contents into RSD as a module.


-steve