June 24, 2003
Comments on the RSS Controversy
There's a lingering open wound in the weblog technology community, and it is called RSS. Last week, Dave Winer and I had a long phone call discussing the issue, and I've been doing some reading and thinking about the controversy. My perspective on the RSS issue: I think it all comes down to naming. Dave has said (on
numerous occasions) that if the folks who created RSS 1.0 had given it another name, he wouldn't have had a problem with it. For example, here's Dave's perspective from
Scripting News on 8/31/2000:
The proposed syndication format takes a new direction. It should have a new name. Then there's no problem. Further, a new format is not bad or good. If there's a suitable new name, and if it gains traction, we'll support it, as we would any Web syndication format that has content support.
Mark Pilgrim has also written up a
great summary of the history of the fork.
Does the problem really boil down to this fundamental issue? Last week, I asked Dave on the phone, "So would this whole controvery be solved if RSS 1.0 was renamed to something else?"
His answer? "Yes, absolutely."
Now here's the
really good news. A number of folks who helped write the RSS 1.0 specification, like
Aaron Swartz and
Sam Ruby, have expressed willingness to drop RSS 1.0 future development, and to rename their
future weblog standards development. Sam's even put out a
call for new names. Here's my suggestion: Call the new work "MSS 1.0". MSS would stand for Metadata Site Summary. Make it clear that this is solving a different set of problems than RSS 2.0 solves - I think the wiki already goes a long way to describe the differences, both in scope and in philosophy. Let's let the confusion end, and bring some healing to the weblog technology world.
Then we can all move forward together. And get to what's really important - sophisticated interoperability and new features that users will love.
UPDATE:I've been told that Sam Ruby was not one of the original RSS 1.0 authors, sorry for the mischaracterization.
Posted by dsifry at June 24, 2003 3:42 PM
| TrackBack
| View blog reactions
Let me guess: Dave never visits Yahoo.com because he still believes that stanford.edu/~yahoo was the proper URL?
Worse yet, Chris: Dave thinks he is the one in the position to make the call regarding who gets to call what RSS. I wonder if Netscape or anyone else for that matter would take issue with that claim?
Worse yet, Chris: Dave thinks he is the one in the position to make the call regarding who gets to call what RSS. I wonder if Netscape or anyone else for that matter would take issue with that claim?
Netscape? What's Netscape? And hey, keep the URL's out of this!!!! ;o)
I can only speak for myself, but RSS 1.0 is an RDF vocabulary as well as syndication format so unless Pie (or whatever it gets called) is/has an RDF mapping that covers the same ground then it will still be needed.
Interestingly, the limits of RSS 2.0 have really been visible with Dave's recent 'funky' comments. That it is necessary to go outside of this (frozen) spec to progress suggests not only that RSS 2.0's days are already numbered, so Sam's initiative is rather timely.
(((btw - MT's giving a server error after the comment post)))
Just for myself, my preference would be something neutral that replaces both RSS 2.0 and RSS 1.0. It would nice if it were in RDF format, or at least translateable to RDF in order to work with other data efforts on the web. At this time, something, anything to stop this constant bickering. Can't make both parties happy? Then perhaps they'll both settle for losing.
But there is more to this issue then just the name -- and the only peace is if all sides are disarmed.
Just for myself, my preference would be something neutral that replaces both RSS 2.0 and RSS 1.0. It would nice if it were in RDF format, or at least translateable to RDF in order to work with other data efforts on the web. At this time, something, anything to stop this constant bickering. Can't make both parties happy? Then perhaps they'll both settle for losing.
But there is more to this issue then just the name -- and the only peace is if all sides are disarmed.
Dave S, kudos to you for this post.
I thought Ian Davis had a good idea with his Unified RSS, described at:
http://internetalchemy.org/2003/06/unifiedRSS.html
Ian suggested that both RSS 2.0 and RSS 1.0 be renamed to, respectively, RSS Content and RSS Metadata, and he made suggestions on how each format could be modified to point to the other and be used together.
I didn't know how to respond simply, but Shelley nailed it: there's more to RSS history than the name.
The fundamental issue is not that Dave would find resolution in RSS 1.0 changing its name but in how that became Dave's prerogative.
Since we're talking about standards and protocols, might an RFC (Request for Comments) be in order? This is how the Unix world was able to agree on how different systems could interoperate with regards to things like e-mail programs and FTP servers and such.
If there was a single document that said "An RSS feed *must* do this" or "An RSS feed *must not* do that", then people can stop worrying about the politics and just write compliant applications.
As for a name, who cares? It's the behaviour that matters.
An RFC would be a fine idea.
I'd take inspiration in writing it from the syslog RFC that got produced by the secure syslog working group. The first thing they did was document how existing systems actually produced syslog, not just what the originally somewhat ill-specified program was supposed to do, but the range of implemented experience.
If people are making mistakes in implementation of RSS, the RFC dictum "be liberal in what you accept" dictates an awareness of that.
Informational RFCs - or even documents in "internet draft" format, which don't need to be approved by anyone before being circulated - have a certain amount of dry implementation friendly rigor for the most part that the existing RSS documents lack.
Who cares what it's called? Who cares about water under the bridge? Let's the take the initiative and move on. PIE - MSS 1.0 - whatever. This is a major juncture in our nascent little industry. I'm sure all the aggregator vendors can support 1.0, 2.0 and PIE! Some are still spewing out .91 and .92!
I just wanna make sure that the blogger's FACE or an ICON of the publishing entity - is included in the spec. We need to SEE who's RSS feed is in our aggregators.
And congrats to Joi, Reid and Mark on teh Soicaltext deal!@@^&%^#&%^$&^%
Why not USF(Universal Syndicatable Format)?
very funny, chris, but you know it was akebono.stanford.edu/~yahoo
It's all about the name. I like XSS because specs with X in them are sexier. ;-)
-Russ
Russ, thinking about Simple Extensible Xml again?
Big news for anyone who hasn't noticed yet - folks from Blogger and Six Apart (Movable Type) are now supporting Sam's initiative.
Let's see if Dave stands by his words, or better still adds his support.
Dave, thanks for the message of support. It's appreciated. The key point still is the confusion over the naming that is hurting RSS and has been for too long. These guys want to make the issue about me, that's just wrong. RSS would still have the problem if I were gone. RSS is a good thing, and it doesn't deserve the shabby treatment it's gotten. These aren't personal issues, because RSS isn't a person, it's a concept, and an important one. Thanks again, now you can see where the problem is.
I'm sorry, if it's not about Dave, then how come when namespaces (ala RSS 2.0) were first proposed in 2000 wasn't the conversation "ok, I don't like namespaces, and I abhor RDF, can we compromise and use only namespaces?"
No, *Dave Winer*, a principle contributor to RSS 0.91, "restated" the Netscape RSS 0.91 Specification, put a *Userland* copyright on it restricting subsequent specification by anyone but Userland, "declared" what the roadmap for development was, and stated "any thing else is a fork". Suddenly, two years later, another Userland copyrighted specification, RSS 2.0, has namespaces and is received just as well as the original proposal for namespaces was *the first time around*.
No, the issue is not about a name. It's about how Dave Winer and Userland presumed the prerogative of saying what RSS was and should have been.
Correction:
No, the issue was not about a name, it's not about RDF, the Semantic Web, or even namespaces themselves. It's about how Dave Winer and Userland presumed the prerogative of saying what RSS was and should have been.
Ken, read the copyright notice on the UserLand spec again. It was taken straight from the IETF copyright, and it in no way prohibits anyone from writing their own spec. Where did you ever get the idea that it did. Y
Now about me having the right to say what RSS is and should have been, check out the First Amendment. So much of what you and the others complain about is covered adequately in that document.
And Ken, one more thing, the issue is the name and nothing else.
Dave says the issue is the name, because "if everyone else" had used a different name for even the slightest additional element in the RSS specification, much less the addition of namespaces or even RDF, then there wouldn't ever had been or now be an issue.
The original namespaces proposal, Rael Dornfest's "RSSx Straw Man" predating the RSS 1.0 proposal and not breaking compatibility with RSS 0.91, is virtually identical to the RSS 2.0 spec two years later. Had *that* spec been written and called RSS, it too would have had us in the same boat today.
That clearly indicates to me that the issue is not about the name but who has was and is allowed to use it on a spec.
Otherwise, if the IETF-like copyright allowed it and the First Amendment provided the right in this case, we wouldn't be having this discussion.
no, the issue is the name -- and the namer.
no, the issue is the name -- and the namer.
Is this conversation giving anyone else a boner?
Why don't we just go back to calling it Meta Content Format (MCF) and be done with all this forking around?
I am reminded of what's been holding back Linux, this same inner warring of factions that theoretically support the same ideal.
I thought we were all on the same side on this.
If it takes changing a name to move on, so be it. Mozilla dealt with the Firebird fiasco beautifully; let's do the same here and be done with this, and leave it behind to the dusty archives of prehistory.
Maybe a concrete example would help.
Let's say a well-respected non-partisan developer did some thorough research on the current usage of RSS 2.0 across all tools and identified errata that would clarify the usage of RSS (nothing new, no changes in the field, no breakage, just making the lives of new developers easier).
How do they author and publish a normative document that says, "this is RSS 2.0, 2nd edition"?
The featherbrain discussions on RSS has no merit.Self-professed intellectuals sometimes get tied up in their own intelletualism. If the program is correct, then the program designation is correct--even if it's called XYZ Blip. What a waste of time and talent.
Cal Gabriel
Sam Ruby and Mark Pilgrim are doing something very much like that with the RSS validator.
Good example. In that case, how would they go about making it normative? That is to say, the "current" spec that everyone would change their links to from previous versions.
Gee Whiz Bang: Excellent! thanks for adding a data point that it's not about a name, but who believes they can say it's a "frozen spec".
Has the "RSS for Weblogs" profile that Userland and Six Apart mentioned been scrapped? Is it still in progress? RSS has a lot of users.
RSS is cool and fashionable, maybe even hip.
The key point still is the confusion over the naming that is hurting RSS and has been for too long.
I'm founder and CEO of Technorati. I was the cofounder and CTO of Sputnik and Linuxcare, founding board member of Linux International, and board member of AOTC. I've been around the block a few
times. Some might call me a serial entrepreneur. A more personal but less-updated site has more information about me - pictures, speeches, software, etc. You can contact me at david-blog@sifry.com.
