Digg floats API, phishing mashups to come
- — 23 April, 2007 08:30
Digg is just the latest in a raft of Web 2.0 darlings to open their data doors to developers by way of APIs. The Digg API -- announced Thursday at a party to celebrate Digg's 1 millionth registered user -- will allow developers to expose Digg story, comment, and user data to mashups of their own making.
So entrenched is the trend of exposing APIs to the developer community that John Musser -- whose ProgrammableWeb catalogs dissect the growing mashup phenomenon -- declared APIs a must-have checklist item during a presentation on APIs Tuesday at the Web 2.0 Expo.
"It's going to be almost like a decade ago. Do you have a Web site? Check," Musser said. "That's how it will be with APIs."
The logistics of leveraging APIs in an effort to transform a Web offering into a bona fide platform remain in their infancy, however, with myriad complex issues quickly arising -- not the least of which is a tried-and-true means for monetizing the model.
The biggest issue surrounding APIs is identity," Musser said. "What's the standard? Is it OpenID ? I don't know. The whole area is vague. Most of the major API vendors have their own authentication APIs. Each is similar, but in the end, they're all different."
The result of this ambiguity will be a new form of malware.
"In 2007, we'll see phishing mashups," Musser said.
The rampant availability of personal data across the Web is another problem with wide-reaching implications, Musser added, demonstrating a mashup that mapped addresses of library patrons who had checked out copies of George Orwell's 1984.
Not unrelated are concerns surrounding copyright law -- that is to say, what data can be mashed up, and how.
Musser pointed to Flickr as a leader in solving this question for APIs providers. "Flickr's service has been successful in many ways because it's been built on a creative common license, and that trickles down to the API," he said. "The developer can pull out the photo and specify what level of creative common license they want. So Flickr is building the [intellectual-property] issue into the API."
As for making an API campaign successful, Musser offered advice aimed at increasing developer adoption.
"I'm amazed at how many APIs have limited documentation, no code samples," Musser said. "The fact of the matter is, adoption can be facilitated by giving people tools they can develop with. Give them libraries; give them examples."
Musser also pointed out the rising importance of transparency into server availability.
"Salesforce's trust.salesforce.com shows availability of all their API servers, which they had to give to the mashup community facing uptime issues," Musser said. "We'll be seeing more and more of that coming this year -- visibility into servers."
Musser also gave an overview of the API tool ecosystem, which he sees as a hot market going forward.
"The tool space is going to explode, both for developers and nondevelopers," Musser said. Of particular note were data mashup tools such as Yahoo Pipes, RSSBus, and Grazr; scraping tools for making structured data from unstructured data, such as Kapow and Dapper; and visual development tools, including JackBe, Teqlo, Bungee, and IBM's QEDWiki.
And in a merger of Web 2.0 philosophies, social networking built around API tools will figure prominently for those organizations serious about making good on the model, Musser said.
"You use this tool and then the components you build live in this community ecosystem, where they get popular, get rated, and you share them," Musser said, pointing to API mashup communities developed by Dapper, Kapow, and Yahoo. "We're going to see a lot more of that vendor-supported, open community around these tools in the coming months."