DeWitt Clinton has a great article about why Atom is a better format than RSS.
I’ve found the Atom 1.0 standard to meet the needs of nearly every single problem that I’ve thrown at it. Amazingly so, actually. I’ve been consistently impressed with how well the authors of the Atom syndication format anticipated the needs of the advanced content syndication community. There has yet to be a use-case that I’ve explored — and I work with some thorny ones — in which Atom has let me down.
That, and the Atom Syndication Format specification is the single best technical spec I’ve ever read. Seriously, give it a read just to see good spec writing in action. It’s concise, accurate, unambiguous, and contains the right amount of illustrative detail.
If Dave responds to this, what are the odds that he can restrain himself from using the FUD tactic of “It doesn’t matter if Atom is better, we all need to use RSS to stay consistent! One bad format is better than two great ones!”
I was happy to discover that the Safari Browser will automatically select Atom feeds, if given the choice between multiple feed formats. Hooray Apple!
Heh… at least Safari is good for something!
Raging hypocrisy so thick:
When it comes to RSS and Atom, Dave says 1 bad standard is better than 2 great ones. (Set aside for a moment the fact that there isn’t 1 RSS, there’s umpteen.) This is a convenient argument when you’re the current buy-in leader and want to argue no one else should even enter the race, as RSS was when he started making the argument.
When it comes to XML-RPC, Dave argues that XML-RPC should be supported alongside SOAP and REST, because XML-RPC is an easier-to-use standard that should be offered in addition to the harder-to-use ones. This is a convenient argument to make when you’re the 3rd-most-popular protocol.
boggle How does this guy’s brain not explode? Why doesn’t the “1 bad standard is better than 2 great ones” principle apply to coalescing around SOAP and sweeping XML-RPC under the rug?
In XML-RPC, he is screwed. He has to make a Goldilocks argument, and there’s no way he’s right. He can’t jump up and down on simple-simple-simple, REST captures that. He can’t jump up and down on complete-complete-complete, SOAP captures that. He has to say XML-RPC is “just right,” just simple enough, just complete enough, when real-life experience shows that by trying to capture both it attained neither. A man who has never, and will never, query XML with something like XPath, is not going to design XML that is easily queried in XPath. He handles his XML as some sort of hierarchical data structure, because that’s the 1980′s-era-datastructure that his programming environment tops out with, so in his mind, your data format is future-proof so long as the current value is always the first value in the response — not the semantically named value, regardless of position.
A quick search on scripting.com for “semantic” shows up only 8 mentions of the word in the last 3 years, and one of those was linking to an old story from 2 years previous.
And for XPath? 2 results ever. And they were linking to other people talking about XPath.
Correct me if I’m wrong, but surely an XML guru like Dave Winer always likes to portray himself, surely could have written a teeny tiny bit of info about these 2 massive XML topics? The fact he hasn’t just highlights the previous comments point: he’s stuck in the 1980s and worse than that, rabidly tries to get us to think in the same way. So much for innovation.
I’m sure he’s got a few mentions of “semantic” on his site, he carried an ax as if “The Semantic Web” was the enemy for quite some time. That correlates highly with the period when he thought namespaces were too complex and should not be used, and instead that people wishing to extend RSS 0.91 should do so in RSS 0.91′s namespace because, after all, other processors should just ignore tags they don’t recognize.
I honestly think changing his mind on this snuck up on him. He got really into having a validator and I think one day realized, a validator could be told to ignore tags in another namespace, because that truly was still valid, but a validator couldn’t be much of a validator if it was going to ignore illegal tags in the RSS namespace…