If you ever doubted that personalities shape the path of technology, all you have to do is look at the checkered history of RSS. Even its acronym is the source of rancor: is it Rich Site Summary? RDF Site Summary? Really Simple Syndication? Any answer is bound to piss someone off.
When Steve Bryant pointed me at Rogers Cadenhead’s blog today, I suspected I was in for another RSS-related rumble. And I wasn’t dissapointed. Cadenhead got a letter from Winer’s attorney last week, threatening legal action over a variety of claims, such as infringement of copyright and a dispute over “third-party information”–other people’s publicly-shared OPML feeds.
Here’s Dave Winer’s side of the story.
Update: Here’s Burningbird’s summation of the whole timeline of this mess.
At its most basic, this is the story of a verbal agreement that went bad. But at another level, this is a fight over the future evolution (or not) of RSS, and something much subtler –if someone makes an API “public”, how public is it? And how open is a standard if someone can come along and slap you with a legal threat when you don’t use it in a way that they like? And just how far can you go using information provided openly by an agreement between two other parties–like an OPML feed, or phone directory listings?
The future of RSS is of more than passing personal interest to me. As the newly-minted director of IT strategy for Ziff’s enterprise group, I have to figure out how to make stuff work together, and right now RSS figures pretty heavily in that. But Atom could just as easily fill that role for my integration purposes, and if the whole RSS process is going to drop dead because of personality issues, I’m going to get as far away from it as I can as quickly as I can.
It doesn’t help any that the entire blogosphere is gathered around Winer and Cadenhead chanting, “Fight, fight, fight!” But then, we’ve seen this happen to people who’ve gotten on the wrong side of Dave before. And inevitably, it makes the higher questions involved subordinate to the battle of personalities. Which means the movement toward fixes in the holes in RSS is going to cease while Dave postures.
Before I go into full rant mode here, it’s time for full disclosure. I once worked with Dave Winer tangenially, as he was a columnist for XML Magazine, of which I was editorial director. I have spoken with Rogers Cadenhead all of twice or three times, more if you include comment threads on blogs and email exchanges (but not much). None of the above qualifies me to be much of anything other than an informed passing observer of either of these guys at the moment. But I do know that Rogers is one of the nicest guys in tech that I’ve ever interacted with, and Dave is…well, Dave.
If you take everything at face value here, the center of the dispute is OPML content. OPML is an outliner format, widely used as a format for exchanging sets of RSS feed subscriptions. We use it on Ziff to package our feeds for easy uploading onto RSS readers, and it’s used by a variety of blogrolling and RSS-based services for similar tasks.
The OPML Factory site that is the main point of this dispute is a port to the LAMP platform of the functionality that Winer had originally run at feeds.opml.org. The site didn’t scale well, so Winer turned to Cadenhead to port it to LAMP. Cadenhead had stepped in to save the bacon for Weblogs.com users who found their Manila blogs dropped when Winer could no longer support them, and had also helped Winer port the Weblogs.com “ping” service to LAMP before Winer sold the service to Verisign.
The two never arrived at a formal contract for the OPML project — the verbal agreement included parntership, says Cadenhead, and the written contracts offered none of that, making his contribution purely work-for-hire. So he refused to sign them, and now Dave wants to take his ball back (plus the $5,000 he paid Cadenhead up front for the work he’s done so far) and go home.
Cadenhead says the third-party data is “open” by agreement–it’s subscribers’ OPML files, which they voluntarily posted to Dave’s defunct OPML service in 2004. But Dave says that the data is not “open” — that people trusted him with the data, and he’s claiming copyright on it.
The dispute is not over OPML itself. OPML is claimed on opml.org as “a trademark of Scripting News Inc.”. The specification for OPML, however, states: “No claim of ownership is made by UserLand to the format it describes. Any party may, for commercial or non-commercial purposes, implement this protocol without royalty or license fee to UserLand. The limited permissions granted herein are perpetual and will not be revoked by UserLand or its successors or assigns.”
The OPML data from subscribers is another matter. Winer’s interest in it is probably fuelled by the indications it provides of what subscribers’ “attention” is trained upon. My former colleague Steve Gillmor, and my boss Mike Vizard, are both big fans of the potential of the “attention economy,” and I’m sure that Winer sees some potential “monetization” of information about what RSS feeds people are subscribing to. And I can see Winer being reluctant to dilute his interest in that cash flow by sharing it with someone who merely stepped up on more than one occasion to save his ass with generosity and programming skill.
But the dispute does have other root causes. In a post I made about RSS advertising recently, I noted that Dave and Rogers were having something of a disagreement about the future of RSS. Rogers wants it to keep evolving, as he thought Dave had intended when he set up the RSS Advisory Board and licensed RSS under Creative Commons, donating the standard to Harvard’s Berkman Center. Dave doesn’t want anybody touching the damned spec, because he says it’s done.
There is no other single person I can point to in the world of web technology who has shaped the development of web standards so much with his or her personal enmity as Dave Winer. If you’re involved in blog or syndication technology, odds are you’ve at least virtually bumped into Winer in some way, and you’ve formed very strong opinions about him. The Atom spec came about mostly because Dave claimed RSS as his intellectual property, and a large set of web developers just couldn’t bring themselves to work with him. Some, like Rogers ,who intially had mostly good things to say about Dave’s knowledge and contributions, have been driven off later by his orneriness.
Scoble weighs in on the competing lynch mobs, or at least on the group that’s been gathering around Rogers’ blog. Mike Arrington, Winer’s former lawyer who is now better known for TechCrunch, takes a driveby potshot at Rogers without any sort of explanation. Rogers responds to Arrington’s comment here.
It’s almost impossible to have a civil conversation over any topic through duelling weblogs. In the end, nobody’s mind is going to be changed about Winer one way or the other. All this argument will do is cast a bigger shadow over further development of the RSS standard for new applications, and we’ll be stuck with whatever proprietary extensions of RSS get hoisted on us by Apple, Microsoft, Yahoo, et al.
And so, I’ll be left with having to deal with a different framework for every point-solution integration point for my syndicated content, podcasts, etc.. All because of a clash of personalities.