Every time RSS comes up for discussion, someone links to Mark Pilgrim's misleading article about the version history of RSS, making a comment like this:

RSS with its 9 [diveintomark.org] +1 [rss3.org] incompatible versions is hardly a standard for anything. It is a huge pain for a implementer to decide which versions to support.

There are only two significant versions of RSS: RSS 1.0 and RSS 2.0. These formats have one major technical difference that prevents their merger: RSS 1.0 makes use of RDF, a standard for data exchange, and RSS 2.0 does not, favoring a slightly simpler approach.

There's only one significant version of Atom: Atom 1.0, which recently became a proposed standard of the IETF.

The other seven versions of RSS identified by Pilgrim are older versions of either 1.0 or 2.0.

An implementor of syndication publishing software can support RSS 1.0, RSS 2.0, and Atom 1.0 and ignore their older versions. All three formats are stable, and code that produces RSS can easily be adapted to produce Atom 1.0. I added Atom support to an RSS-feed PHP script in a half-hour.

-- Rogers Cadenhead

Comments

Heh. I started to take up the same old dance, about how it's easy for a producer to just say whatever, because the whole burden of "do they mean to skip 11pm or midnight (and do they know it's GMT, or, like a certain very large producer of skipHours, are they brokenly using local time), and is that supposed to be escaped HTML or literal text" all fall on the consumer, but then I actually RTFA's parent.

Have the Slashdot editors all been kidnapped and replaced by monkeys? First "RSS 3.0" and now some random marketer making Microsoft's case for using "web feeds" rather than "RSS" to mean "RSS and Atom" by being confused into thinking that MSFT's support of "RSS" means "RSS and not Atom"? Has Slashdot's quality just suddenly gone into the toilet, or was it a long slow slide?


 

Can I vote for Phil as head of the RSSAASO (RSS And Atom Standards Organization)?


 

Someone on the Textpattern boards just linked to the "9 versions" silliness as one excuse for Textpattern only supporting RSS .92. Thanks for the timely debunking.


 

There was a time you could justify sticking to an RSS version prior to 1.0 or 2.0, but at this point both of them have been stable for years. Do you know if anyone has submitted a patch to add RSS 2.0 support to TextPattern?


 

The above comment about Textpattern isn't correct; Mark's piece was linked humorously in response to someone pointing out the "RSS 3.0" guy, not as an "excuse" of any sort.

The official response given when this was discussed in the support forum was threefold: first, Textpattern does RSS 0.92 as a "baseline" of sorts; they expect there's nothing on earth that can't handle it. Second, they're not happy with the RSS 2.0 spec's ambiguities and are wary of trying to implement. Third, for people who want features RSS 0.92 doesn't offer, Textpattern happily generates Atom 1.0 feeds.

But maybe now that the first officially stable release is out the door, there'll be more work on RSS 2.0 support (I get the feeling there were other priorities leading up to the release). And there is a patch circulating for RSS 2.0 which I know a couple of people have tried it out, but I don't know how much consideration it's gotten on the dev team. I'll get out my poking stick...


 

Just for the record, RSS 3.0 was "linked humorously" as well. And while I wouldn't disagree that RSS 2.0 has far surpassed 0.92, the fact is that 0.92 was the version that everyone used for a relatively long time and was in vogue when Dean initially added RSS support to Textpattern. After that, Dean had a lot of other things to work on besides waiting for the standard to stabilize. But I am sure the support for "RSS 2.01 rev 2" will be included soon.


 

Everything listed on www.intertwingly.net is feed format (even 1.1). Only one is a standard. Ta!


 

Add a Comment

These HTML tags are permitted: p, b, i, a, and blockquote. A comment may not include more than three links. Participants in this discussion should note the site's moderation policy.