I'd Rather Fight Than Switch

Anil Dash, responding to Don Park's take on the Atom 1.0 launch:

Don, you raise a lot of really good points, but surely you can't argue that changes like clarifying the RSS 2.0 spec, transferring ownership to a neutral party, and embracing namespaces weren't at least partially motivated by the existence of Atom? Even if you don't like the feed format (and I'll gladly concede the API is much more interesting), Atom's the best thing that's ever happened to RSS, no?

No. If you compare today's RSS 2.0 specification to the one in July 2003, back when the Atom project was launched, you'll find only minor edits.

The Atom Syndication Format's an intriguing, well-specified protocol, but in the time it took the world's most democratic spec-drafting team to finish Atom 1.0, RSS 2.0 has grown at an astonishing rate. One of the reasons is that it was left alone: Dave Winer froze RSS before passing the spec to the Berkman Center, then helped RSS Advisory Board members fight the urge to thaw it.

And when I say Advisory Board members, I mean me. I still want to take an icepick to that thing.

As a syndication and weblog API dork, I like Atom, but I don't understand why it took longer to create Atom 1.0 than it took to invent XML 1.0 (approximate count from first announcement to recommendation: 450 days for XML, 725 for Atom). This is a syndication format, not a space shuttle. I knew they were in trouble when the project became mired in a three-month-long bikeshed discussion over what to name the format.

Still, as someone who knows the pain of getting anywhere near a specification, I congratulate the developers who emerged alive from the two-year struggle to create and standardize Atom. And I for one welcome our new syndication overlords.

Rss · 2005/07/15 · 3 COMMENTS · Link

Comments

Rogers, you ignorant slut...

nah, I can't be bothered with this. :)

#1 · Robert Sayre · 2005/07/15

You're chronology may be more accurate, but it was pretty much the same pressures that led to Atom that led to the neutral party & namespaces in RSS 2.0.

It took 'simple' RSS about three years to get those - why did *that* take so long? How long did it take from the suggestion that RSS 2.0's content escaping was a bad idea to the acknowledgement that silent data loss could happen? (A problem still frozen into the spec, no matter how many informative examples are provided).

...and what Robert said ;-)

#2 · Danny · 2005/07/16

Congrats on the release. Atom solves the content escaping problem, but the complexity of the solution was a hard sell for RSS 2.0.

The keep-it-simple approach has done so well, and the spec is now so widely implemented, that there's little external pressure to change it.

#3 · Rogers Cadenhead · 2005/07/16