Tim Bray in a post asks aggregator developers to be picky about their support for Atom feeds. He specifically calls on the developers of a couple of ‘popular aggregators’ to enforce correctness when parsing Atom feeds. And it looks like those developers plan on doing just that. He doesn’t mention Bloglines, even though I’m pretty sure we have more users than the other aggregators mentioned. No big deal on that; we haven’t announced any user numbers yet. This whole debate is so clear to me, I hesitate to even label it a debate. For the record, Bloglines will always try to parse as much as possible from a feed, regardless of whether it’s well formed or not. That’s what we already do right now with our support of all the major versions of RSS and Atom. In fact, the parser is so liberal, you can mix the different formats in a feed, and it gets things right anyways (I’m not recommending anyone do this, however). It’s all about the users. I never want a user to have the experience where they cannot subscribe to a feed. If that makes our lives as developers more difficult, well, suck it up. We’re developers; we do the hard work so users won’t have to. Speaking personally as someone who has written a web browser, I know that non-conformance to standards can lead to more development effort. But it’s the right thing to do. So, if other aggregator developers want to try to enforce strict compliance with a spec, more power to them. But let me repeat our mantra: It’s all about the users. If a user wants to subscribe to a feed, we’re going to do everything we can to support that feed.