Klout is currently putting the finishing touches on a new version of its API. The company has started a private beta with selected partners, and ProgrammableWeb had the chance to catch an early glimpse of what is to come in this new Klout API.
Version 2 of Klout's API has been in the making since the beginning of the year, comprising of an entirely new backend and serving system. Its main purpose is to provide better support for the large volume of API calls, that Klout has seen in 2011. The new API will also be a preparation for new features and uses cases. "[The Klout team is] working on building new widgets and really pushing +K into the platform for developers and consumers everywhere. We continue to search for ways to help users without social ids get rewarded for their influence.", commented Tyler Singletary, Developer Evangelist at Klout.
Since beginning of April Klout has been inviting selected partners to join the private beta phase. Together with the new API, Klout also relaunched its Developer Portal. The new API is supposed to be in beta for a couple of months, to guarantee a smooth transitioning phase for existing API users. Tentatively v1 of the Klout API will be shut down at the end of 2012.
So what is new in Klout API version 2?
The most major change is that the API now requires two steps for fetching data: In the first API call one has to look up the Klout ID for a user, based on either twitter ID or twitter screen name. In the second call one can then use this Klout ID to fetch the Klout user profile, Klout score, topics or influencers/influencees.
"The intention was to allow the possibility to request data for users that have registered via networks other then twitter, and to refactor the API to be less Twitter-dependent.", explained Singletary. While this implies that new networks will be added to the API in the near future, Singletary did not want to comment on exact roll out dates for these networks. "It is more complicated than the technical problem: you have to respect a user's privacy and the terms of service for those networks.", Singletary said.
Other important changes to the Klout API are:
- To ensure the highest data availability and the lowest latency, bulk user calls are no longer allowed. To compensate for that the APi rate limits will be increased.
- XML requests and responses have been deprecated, although Klout is willing to reconsider this decision if absolutely required by its integration partners.
- The Network Score, Amplification Probability, True Reach, and Klout Style properties have been removed.
When will the new API be available for everyone?
Klout wants to open self-service registration for the new API around beginning of May.
To foster the adoption of the new API and to gather more feedback, Klout will host its first real hackathon at the weekend of May 12th-13th. "Together with a number of great sponsors we are hoping to offer a few more endpoints and surprises to the attendees.", Singletary commented.
"We are at just under 30B pieces of Klout data served a month now, with over 5,000 API partners. We have seen phenomenal growth in the last few months alone.", Singletary mentioned when asked about their API request volume.
We at ProgrammableWeb are exciting to see so much movement in the API space, and with Klout there is apparently a strong contestant for this year's edition of the API billionaires club.