Yes, I agree that failure to model Atom 1.0 feeds accurately is a bug, though as Paul Hoffman pointed out off-list, there is at least one scenario where we will fail to do this (at least for this release), which is signed entries.
So, a more accurate way to state it, would be that, given the choice to normalize to RSS 2.0, we will do everything we can to ensure that the information publisher's puts in the feed is available to developers using the API. For what it's worth, there will be updates to this API over time, so it is possible that we will have an API that makes the content available in Atom format, as well (or in the native format, though that's harder because it requires that we persist the original XML). This feedback is all extremely useful in helping us figure out what the right thing to do is. Thanks, Sean -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, February 24, 2006 4:02 PM To: Sean Lyndersay Cc: James M Snell; Atom Syntax Subject: Re: [rss-public] Microsoft Feeds API Enclosure Test On Feb 24, 2006, at 3:05 PM, Sean Lyndersay wrote: > I'm sure that many people -- on this list in particular -- think that > the right thing to do is to normalize to Atom 1.0, instead. > Yep, that's certainly one way to think about it. But then I'd be > having this same discussion with Dave and with folks on rss- public. > :) In short, I'd rather avoid the issue altogether and provide some > value to the developers who are using the platform -- which means > preventing them from having to learn several different formats to get > common data, while allowing them to get access to extensions. Given that Atom 1.0, in practice, is a clean superset of RSS 2.0 (there are things in RSS 2.0 that aren't in Atom, but they are typically not used; see http://www.tbray.org/atom/RSS-and-Atom), the only one that's been through a formal standardization process, and the only that's guaranteed not to change (see IETF rules) this feels a little weird. But it's your API. I will say, though, that there are already a lot of Atom 1.0 feeds and there are going to be more, so it does seem like a basic requirement that your interface be able to model Atom 1.0 accurately without data loss. I assume you agree that a failure to do this would be a bug. -Tim
