I've been following what the 3 big Internet companies have been doing with RSS and I largely agree with Dave Winer's recent assessment:
"Yahoo is dashing in front, with Microsoft close behind. Why isn't Google in the race?"
Yahoo has been by far the most proactive company of the 3 this year. Microsoft has shown glimpses of what they're plotting - my favourite so far was the start.com prototype RSS Aggregator integrated with MSN Search.
But Google hasn't done anything with their Blogger unit and they seem reluctant to add RSS functionality to their product line - e.g. Google News and indeed the Google search engine ('subscribe to this search', anyone?) OK, there's a hint of an RSS Aggregator in a new feature being tested for Gmail - Web Clips, it's called. Maybe that's a sign of things to come.
Today I found out a bit more about Microsoft and RSS. Evan Williams blogged a dinner he and 7 other bloggers had with Jim Allchin, a major Microsoft honcho who is responsible for Windows, .NET and "new media technology". Robert Scoble set up the dinner. I found this bit from Evan to be especially interesting:
"One of Jim's repeated statements was that he wanted to bring "this stuff" to the
masses. I asked for clarification because, in a lot of Microsoft's talk, they speak of
RSS and blogging as the same thing. He agreed they weren't the same thing, and it seemed
to be RSS he was talking about implementing in a variety of ways throughout Windows
(e.g., built-in readers, automatic feed generation from a variety of lists...). While
Microsoft does have a blogging tool, that's MSN—not Jim's department."
(emphasis mine)
This is another hint that Microsoft will build in RSS functionality as part of the plumbing for their next OS (Longhorn). It was clear from the start.com prototypes too that Microsoft's strategy with RSS is to integrate integrate integrate. They want RSS to be baked into their products and particularly their OS, so you don't need to download web apps or subscribe to web services like Bloglines.
Yahoo on the other hand is more about utilizing RSS as a media and content enabler - giving My Yahoo users more content to choose from, letting Yahoo News users dive into their niches, enabling mobile users to access content via RSS on their phone, etc.
So what's Google up to with RSS? Or Atom, as the case may be. If I was to guess, I'd say they want to harness RSS. Most of their huge take of advertising revenue comes to them via webpages - their own pages, plus external webpages that use Google Adsense. So Google wants to ensure that revenue doesn't get siphoned off if they make content available via RSS.
It's interesting that the Gmail 'Web Clips' feature mentioned above currently has adverts alternating with the content (according to Ev). So Google is actively experimenting with ways to monetize RSS.
Once Google has gotten a handle on the revenue-generation options, then they'll commit to RSS and roll it out to the users.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/2444
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
I think there is a real potential for RSS to become nothing more than a slightly-more-structured XML. Right now, RSS means something: a "news item" or a "post" or an "update" that has _some_ sort of "reasonable" _meaning_ to _some_ human _somewhere_.
An example of my concern with the idea of RSS being "baked in" to everything and how this would lead to an RSS usability collapse would be, for example, if every file in a journaled file system had its entire changlog available as an RSS feed all the time - including files that were themselves RSS feeds.
RSS makes sense to users now - an RSS item is a "news" item. The moment systems both provide "system event logs" in RSS format, _by_default_ and there are system-wide RSS aggregators that aggregate all RSS feeds that the system provides _by_default_ then RSS will become something that normal people will be pushed away from because of the complexity of finding quality sources of content - not because they aren't out there, but because too many "baked in" content sources will provide too much technical details about information that too few people care about.
And then, if these "baked in" per-system content sources aren't automatically searchable by a local aggregator, then the geek crowd that actually wants such feeds (feeds based on system event logs, for example) instantly winds up not being happy either and so the promised simplicity that RSS might be able to provide - by aggregating content sources - fails again.
So, what do I expect? I expect, in typical fashion for the company that wound up baking the browser into the file system navigator, that they'll wind up having a variety of options including what sort of feeds one wants to have subscribed-to/searched/indexed initially - and that management of that set of options will, themselves, become dizzyingly complex, not only for the individual user, but also for the IT management staff that wants to control all such settings for all computers in that particular company.
It may sound like I'm not exactly happy about such an approach. Maybe I'm not. But then, I can't think of a better way. Because honestly, RSS is just an event stream, and event stream technology has been around for years....
That is what I think Google is waiting for. They're waiting until they think of a better way to embrace RSS. If you look at Google's adoption of XML, they followed a similar course. They generally adopted later than they could have, and open-standard technologies that were more mature than they could have.
I'm really afraid that with all the RSS feeds of wiki page updates and CVS log entries and whatnot, that any attempt to have a service that crawls feeds, aggregates them, and provides a uniform search result, is going to be pretty hefty duty, and of questionable result compared with a typical Google web search, at least for the time being. It was better back in the days when people hadn't come up with all these automated feed things.
Posted by: Andrew | April 18, 2005 9:45 AM
I agree with Andrew and almost made a similar post on here the other day. As RSS is continually massaged to allow more flexible data feeds it certainly begs to question whether the simplicity has been removed from the definition. RSS is simply XML and should not be thought of as the only available format.
It would be feasible to suggest the tools available to publish and consume XML data will continue to be simplified, allowing publishers and subscribers more flexibility in the format of the data they send to each other. There is no sense in trying to put a square peg in a round hole.
In terms of Google, I think they can't just sit on the fence until the technology matures. That's a fine way to become irrelevant.
Posted by: Brady Joslin | April 18, 2005 9:54 AM
Re: "implementing [RSS] in a variety of ways throughout Windows"
One of the things we talked about at dinner with Jim Allchin was the idea of extending the Longhorn system so that "virtual directories" (created by doing searches of file metadata using XQuery) could be represented to users as Atom/RSS feed documents. This would allow you to load the feeds up in an aggregator, send them to friends via email messages, or save them as documents (i.e. saved search results). Jim seemed to like the idea... In any case, whether or not it is actually implemented by Microsoft, it is clear that one could easily build such a system as an add-on to Longhorn given the APIs that will be available.
bob wyman
Posted by: Bob Wyman | April 18, 2005 12:17 PM
Andrew and Brady, interesting thoughts... if I can try to sum up your argument in a sentence: you're saying that baking RSS into the OS may make RSS consumption more complex and so introduce usability issues? That's a valid concern, however I'm an optimist who believes that new kinds of aggregators are being build as we speak that will solve that complexity.
Bob Wyman's PubSub is one of those new breed of aggregators, I believe (certainly one of the most clever aggregation services currently available). Bob's comment above I think gives a clue as to how the concerns raised by Andrew and Brady will be addressed...
As for Google, I agree they're waiting to pounce. The thing with them is, they've proven before they can wait a bit and then release an amazing product that is streets ahead of the competition - e.g. Gmail. So I'm pretty sure they're cooking up something special around RSS...
Posted by: Richard MacManus | April 18, 2005 12:52 PM
Richard: yes, exactly re: Google.
The thing is, Gmail is a service/product. RSS is a technology.
The discussion of the adoption of IMAP instead-of/in-addition-to POP was one that was generally a non-issue for most users, as well as it should be. The service/product of mail was already present. Google just provided a better interface/metaphor for mail via Gmail and did a pretty good job handling one of the major problems with mail (spam) at the same time.
RSS is a technology. While for many of us, the service is clear: end-user aggregation, content filtering, searching, etc...; I think that we're still losing out on a viable pitch to "the average joe" why they should care about RSS. And the fact of the matter is, they shouldn't. It is just a technology. Just like IMAP. They shouldn't care about it - they should care about what it does for them. What is the best way we could pitch it? "It is this service that figures out what news you care about and tells you the news that _you_ care about so you don't have to waste your time reading other stuff." But the thing is, that pitch already happened (I fail to recall the name of the company, if it is still around), and didn't wow people like lots of folks were expecting then.
The moment you have to explain to a user the difference between a push and a pull technology, you probably lose them. The moment you have to explain to them the difference between subscribing to an RSS feed and subscribing to an e-mail mailing list, you lose them. The moment you you have to explain to them why they would want to follow a blog and leave a comment there as opposed to participating in an online forum, you lose them. The moment the technology becomes the issue and not the service, the technology loses some of its adoption potential. The service that RSS technology can provide is still too conceptually difficult to explain to most people, IMHO.
Gmail revolutionized the way we view mail through very simple concepts: "search, don't sort", "conversations", "we will learn what spam is from you". I expect all Google needs to do is to have the right set of new simple concepts, and they'll have an RSS/Atom based killer app and not just a bloglines ripoff.
In fact, I wouldn't be surprised if some of the same technology being developed for spam filtering is going to be repurposed for per-person content-searching learning/customization. In fact, can you imagine viewing all the feeds you read and all the discussion about them through a Gmail-style interface?
Posted by: Andrew | April 18, 2005 2:32 PM
I guess I worry less about the "spam" issue that's fundamental to Andrew's comment. Not spam in the sense of true garbage, spam in the sense of "too much stuff". At this moment, there's a vast multitude of blogging going on. It's all just web pages. A lot of it is unsearchable for precisely the reasons discussed here (although certainly there are movements afoot to solve this through tagging, and folks like Feedster working on search solutions).
But we all happily find the things that are important to us over time because we have trusted third parties. Not aggregators as much as filters. We rely on our community to find good stuff for us. We all know this works, we all have our own lists.
So fundamentally the problem with RSS (especially with enclosure based content) isn't one so much of "there's too much" as one of emergence. As it emerges, the filters will emerge too.
Posted by: Dave Nadig | April 20, 2005 6:47 AM
Dave, Andrew - great thoughts, thanks for sharing them.
Andrew, re your comment: "can you imagine viewing all the feeds you read and all the discussion about them through a Gmail-style interface?"
Now there's an idea!! It wouldn't surprise me at all if Google did something along those lines.
Posted by: Richard MacManus | April 20, 2005 12:59 PM