A Web API lets you use a web site’s computers, data, algorithms, and functions to create your own web services. [Google](http://www.google.com/apis/), [Ebay](http://developer.ebay.com/common/api), [Amazon](http://www.amazon.com/gp/browse.html/102-1840640-1685746?%5Fencoding=UTF8&node=3435361), [Yahoo](http://developer.yahoo.net/), and many other web services have APIs.
__RSS is like an API for content__. RSS gives you access to a web site’s data just like an API gives you access to a web site’s computing power. Most important, RSS gives you access to _your_ data that you have locked up on a web site.
Every Web 1.0 company will have to decide what content they will open with RSS. For example, Amazon already makes _their_ content like their book catalog available through their API. But will Amazon open up user-contributed content through RSS?
Will you have access to
* _Your explicit content_ like your purchase history and reviews you’ve written?
* _Your [drive-by](http://paul.kedrosky.com/archives/001502.html) content_ like the books you have recently browsed on Amazon?
* _Other user’s content_ such as book reviews?
I believe I heard [Jon Udell](http://weblog.infoworld.com/udell/) say that the winners of the Web 2.0 orgy will be the sites that don’t lock up user-contributed content. Instead, the winners will create a compelling ecosystem for you to store your content and bring in your content from other sites via RSS. Food for thought.
Note: This is Part 3 of a continuing series called _RSS is the TCP/IP of Web 2.0_. You may also like Parts [1](http://www.nivi.com/blog/article/rss-is-the-tcpip-packet-of-web-20/) and [2](http://www.nivi.com/blog/article/rss-is-the-tcpip-of-web-20-ii-yubnub/).
Is not other user’s reviews already available by Amazon Web Service 4.0? If so, converting them into RSS, if necessary, is not a big issue.
I have not checked how Amazon made a deal with its customers about the re-distribution of those reviews. (just because lazy to read it all)
Is RSS an API, or more of a model of content? I would view an API as the interface into a system, the parameters (or feeds) I need to know about in order to access the content. RSS is, at its core, an abstraction of what a piece of web content is, which is why it works so well in so many cases.
Incidentally, we built our search platform as an RSS-modeled API; Blogdigger’s entire system and data store is accessible via RSS. The API part comes in the form of the various ways of getting at the data (by keyword, site, links, location, media, etc.).
Hyperlinks help us to see the relationships between communities but RSS/Feeds actually allow these communities to fully connect with one another.
“Fire Brown” Community Aggregation
For example, everyone is pissed off that FEMA Chief Michael Brown hasn’t been fired yet. Well using a RSS feed aggregator, you could get a bunch of sites to setup a new blank blog with just one post on it entitled FIRE BROWN with a description saying add your name in the comments if you want Brown fired. Then you collect all of the comment RSS feeds from those various sites and aggregate them into one large feed using a web RSS aggregator like FeedDigest. Then grab the javascript for the aggregated feed and put that up on another page for people to view. There you go, now you’ve got the collective people speaking.
Now obviously you can see all of the readers comments at once due to the way feeds are displayed but I just found out today that FeedDigest does archive their full feeds and those feeds are searchable as well! It’s too bad feeds are item limit because if they weren’t then everyone could see the entire feed data and even more so they could distribute the full listing of names on any site!
Sorry error above, I meant to say “obviously you can’t see all of the readers comments” in one feed.
Greg G. makes a good point in post #4. RSS does not employ the granularity that is commonplace in a well targeted API. This does not always have to be the case; RSS “overloaded” with RDF syntax or additional (and permissible) markup. It has been done before in RSS 1.0, and it is being done now with RSS 2.0 with Dublin Core Metadata.
RDX is an experimental example of this that, in theory, could provide site specific actions and/or information. Web forecasts have been excitedly predicting a future where websites have automated functionality. RDF is a proven way of integrating semantic technology into a site where other programs can understand it. This is, after all, one of the goals of using an API.