Recovered from the Wayback Machine.
A note to people writing articles about RSS-based news aggregators. UserLand wrote and deployed the first one, in the spring of 1999. It was called My.UserLand.Com and was quite popular.
The concept of news aggregators, as well as RSS, had roots that extended back beyond Dave and My.Userland.Com. The concept of using XML to provide news feeds had implementations as channels with both Netscape and Microsoft (and other specialized companies that didn’t survive the dot-com implosion).
Don’t believe me? Then read an article I wrote and managed to salvage through the Wayback Machine from the now defunct Netscape Enterprise Developer magazine, January, 1998. In it, I showed how XML provided for IE channels, known as CDF could actually be picked up and used in other “aggregators” — except they were called “applications”. Jon Udell also writes about this, and references the use of RDF for describing channels in this Webbuilder article.
To provide more background material, Dan Brickley did a very nice overview of the history of RSS at the Yahoo-based RSS Development discussion group.
So, technically, Dave not invent the concept of using XML for aggregation. Nor did he write the first implementation of a “news aggregator”. And the examples I just cited are what is known in legal circles as apriori art, which means that Dave should use caution when he throws around “patent” with the implication that he’s the inventor of RSS or aggregation.
What Dave did do was help provide an implementation that gained popularity for the idea, especially when Netscape dropped out of the picture. For this, the RSS folks do owe Dave a debt of gratitude. However, at this point in time, it’s time for the concepts to slip out of one person’s hands and into the public domain where it belongs.
The RSS 0.9x family of RSS has been and will always be under the dominion of Userland. Debt of gratitude or not, I would rather put my money on a specification that isn’t owned by any one person or any one company.
The RSS 1.0 specification has two advantages. First, it’s based on RDF, which means much of the existing work and APIs and technologies that can be used with other RDF applications can be used with RSS. Secondly, it’s an effort that’s based on a team effort, with no one person ‘owning’ the effort at any one time. In fact, the RSS development team just voted to allow several new members on the team due to their outstanding contributions to the specification.
Dave asked us last week about what the RSS in RSS 1.0 means. He then printed up a page of our efforts, and then…nothing. Why did he do this? The only reason I can see — the only one — was to look for something with which to discredit the RSS 1.0 effort. And since those few of us who took the bait didn’t give him a weapon he could use against us, he somehow latched on to an email sent to him from an RSS user with little markup experience who believes that somehow RSS 0.9x is simpler than RSS 1.0.
News for those who think RSS 1.0 is too ‘complicated’ to work with: The RSS functionality I built into this weblog page uses straight XML processing to parse the RSS 1.0 page, and incorporate the contents. It was built using PHP, and took me about, oh, 20 minutes to write.
Piece of cake.
I’m not going to repeat my reasons for supporting an RDF base with RSS 1.0. I am going to ask my own question, instead:
Dave, don’t you think it’s about time you stopped fostering the split between the two specifications and work with a team of folks — a team — in making sure that the RSS 1.0 specification maintains its simplicity, even with the use of RDF?
In addition, Dave, you might as well know right now that I wrote about aggregators in my RDF book and didn’t include Radio. Why? Radio’s only one application, it’s not the first, and aggregation is only one part of its functionality.
Regardless, don’t you think it’s a bit bald to write the statement “Sloppy habits that come from working in a corrupt industry” when referencing both a publication and a reporter who can’t then defend themselves because you won’t even privide a link or a name?
Second Update: I received an email from a friend that perhaps Dave didn’t link to the article because he was trying not to be directly confrontational, and was trying to be nice.
Any professional author and publication can take the heat — I know I have for my articles more than once. Dave withholding the link for these reasons just doesn’t wash, though I will acknowledge that his intentions could have been ‘good’.
However, by criticizing the article, saying it was ‘wrong’ for not including Radio, and then not providing a link to this article, Dave’s preventing us from reading it and judging for ourselves. Now all we’re left with is rumor and inuendo.
Update: In the comments associated with this posting, b!x pointed out a thread on RSS and the ethics of republication at Blogroots.com.
In my January 1998 article, mentioned above, I covered the problems associated with providing a data feed using XML on the web and how the material could be republished in ways not intended.
There’s also more of a story with this article — I proved this concept and associated problem out with Wired’s CDF file at the time, republishing their data at my web site as an example of the problems associated with published XML feeds. Wired discovered this — and the awful page I plastered their data into, leading them to start protecting their XML feed at that point.
Now, almost five years later, we’re starting to question the ethics of republishing RSS feeds.
My darling webloggers, what did you think people were doing with all that data — printing it out on pretty paper and using it to paper the baby’s room?