RSSVote

Pro RSS

I'm going to push for a vote on abandoning the Atom syntax in favor of re-using RSS 1.0. About half of the big contributors to Atom have told me they would support such a move, but fear there is no consensus. I think the problem is that they don't know who each other are. This page will help the pro RSS 1.0 people know they are not alone. Please add your name here if you are in favor of an Atom that re-uses RSS 1.0 and abandons the current Atom syntax. This is not an Atom vote.

Pro-RSS 1.0 add your name here

• RandyCharlesMorin

• IanDavis

• DannyAyers

• JeremyGray

People who oppose this suggestion, sign here

• AsbjornUlsberg

• RobertSayre

Assumptions

1. Some Atom syntax elements not already present in RSS 1.0 would be added to RSS 1.0 as a module extension.

2. Atom IETF working group would take full control of RSS 1.0. Although I don't speak for the RSS working group, many of them have indicated they would favor such a move.

3. The IETF working group would be free to introduce a new version of RSS 1.0 (example RSS 1.1) in order to meet requirements of the Atom API. These changes would not invalidate existing RSS 1.0 feeds.

Comments

[AaronSw] I like RSS 1.0 but I don't really see the benefits to using it here.

[RandyCharlesMorin] I'd say the primary benefit is removing the Yet Another Syndication Format argument, which you all know is holding me off supporting Atom fully.

[DannyAyers] Although I strongly favour RSS 1.0 over RSS 2.0, I'm still a little ambivalent about this proposal. As the vanilla XML tribe will testify, the syntax of RSS 1.0 is less than optimal for syndication as a stovepipe application, i.e. if publishing/reading news is all you want. Coming the other way, a simple XML language like RSS 2.0 leaves a lot to be desired in terms of extensibility (btw, I don't think RSS 2.0 is an option for other reasons). As the RDF tribe will testify, the syntax constraints compared to RDF/XML means that most developers haven't taken advantage of the RDF model (which is the good bit!). It's certainly easier to consume RSS 1.0 with RDF tools (it just works), but that's only half the story. There isn't that much practical difference when it comes to production between a narrow subset of RDF/XML syntax and a custom XML syntax (assuming the latter is well-specified, and has an unambiguous RDF mapping, both of which we should be looking for in any Atom). So if Atom were to go the RSS 1.0 route, I'd urge that the syntax is opened up to full RDF/XML.

Regarding the assumptions:

1. ok, fine;

2. if the RSS 1.0 group are happy with this, it definitely gets my +1;

3. Ok, but then I fear we may lose some of the benefit of the work done so far. Rather than calling it RSS 1.1, I'd stick with Atom 1.0.

One final note - I recommend that anyone wondering whether or not to add their name to this list (re-)read the RDF Primer first.

[RandyCharlesMorin] +1 on sticking w/ Atom 1.0 as a name in place of the example name RSS 1.1

[IanDavis] I'll qualify my addition to this list by stating that my desire is to have a syntax-neutral atom API that can be used with RSS 1.0, RSS 2.0, XHTML, RDF or whatever else people want to use to publish their content.

[RandyCharlesMorin] This is exactly what I want to do. Maybe I should propose this instead. Thoughts?

[AsbjornUlsberg] Yes, do that. Proposing that the work we've done with Atom the last year is just a waste of time, is almost insulting. Instead, I would very much agree with you to make the API syntax independent, and let the format development go it's way unaffected.

[DaveWalker] As much as I can see some benefits from going down this route, I'm a "view-source" advocate at heart, so I can't really throw in my lot with it. I've said before that the thing that sold me on Atom was that I was able to put together a Blosxom template (which validated) for the first draft format with nothing more than a text-editor and a close read of a few other people's feeds. I'm not sure an RDF/XML-based syntax could ever keep that degree of simplicity.

[RandyCharlesMorin] The price of simplicity is loss of capability. If we want to use this format for anything beyond simple newsreading, it's going not going to be trivial, however it's done.

I understand Dave Walker's argument and agree w/ him. That's always been my fear of RSS 1.0; the complexity. But on the other hand, why can't we just merge RSS 1.0 and Atom, in this way preserve the RDFism and get some of the simplicity of Atom into RSS 1.0.

Will this move towards RSS 1.0 be supported by the RSS 2.0 crowd?

[RandyCharlesMorin] iM trying to convince the 2.0 crowd that this would be good for them. I've managed to get a reserved OK from u know who, but others are steadfast against.