It’s the battle of the real-time search APIs. This morning Collecta released an API in beta for developers who want to integrate real-time search results into their Web apps. To try to spark interest in the API, it has put up a ChallengePost offering a 15-inch Macbook Pro (retail value $1,699) to whoever comes up with the best mashup. Meanwhile, competitor OneRiot is trying to steal Collecta’s thunder with its own ChallengePost offering $1,500 to the developer that can come up with the best real-time visualization using its API, which it launched publicly back in July.
The battle for the most developer buzz and engagement is worth more than $1,500 to each company. In fact, OneRiot’s early lead in real-time search is precisely because of the growing adoption of its API among other Websites, browser add-ons, and apps.
Collecta’s API is a simple REST API that brings back 15 search results per query, with a limit of 5,000 queries per hour. It includes the most recent blog posts, news articles, comments, photos (from Flickr, TwitPic, and yFrog), and videos (from YouTube and Ustream). The API does not include Twitter results, although that is what dominates Collecta’s own real-time search engine. CEO Gerry Campbell says that since Twitter offers its own search API, developers are better off going directly to the source. But offering an API which offers all real-time results in one easy call would be preferable. OneRiot’s API, for instance, does include results that are at least influenced by Twitter, although it does not include Twitter results themselves. (These limitations could be due to the caps imposed by Twitter itself through its API).
The unique thing about Collecta, however, is that it creates a continual XMPP stream for search results on its site. While it is not currently offering an XMPP API, Campbell says that the next version of the API will, which means that developers will be able to create apps with a steady stream of real-time results that just flow into their apps without any refresh.
The real battle here is not for $1,500 or a Macbook. It is for which API will spur the most creative and useful apps. Let the battle begin.
good pr move for both companies.
Got to say interesting, but don’t these two search engines work only because of twitter?
I mean these two search engines are tide into Twitter so much it does not make sense why we don’t just use twitter to search.
I noticed that Bonush has today romoved twitter from there search engine and instead have there own index engine.
It looks like Techcrunch is even sending them information. I am I right or wrong?
http://bonush.c...hp?q=techcrunch
There blog even says that they are including Live Stock information in the next version, collecta does not do that, do they?
We do include Twitter results on the Collecta.com website (along with millions of other data sources), but we do not include Twitter results in our API. Anyone can just get the Twitter data from Twitter themselves.
Our reason to exist is that there is an entire world of Real-Time Data beyond Twitter, and we are working towards obtaining it all, and driving traffic back to the original data sources. We are the only real-time search engine streaming non-Twitter results from such a wide collection of sources, which sets us far ahead of anyone else in this space, especially in regards to our API for building apps upon.
Once others start copying what we now alone are doing, other factors (such as our architecture, interface, speed etc) will come in to play. But as for now, we are (quite surprised to still be) the only ones doing what we do.
I don’t understand this reasoning: why exclude Twitter specifically? You say you aggregate millions of other data sources, but anyone “can just get the Twitter data from Twitter themselves.”
That doesn’t seem like such a good paradigm.
I’m sure they would, but Twitter won’t let them. Search API limits are such that the Collectas and OneRiots of the world are unable to offer that service.