I’ve often found myself countering the argument that the Internet is making us dumber by citing Google as a second brain: the search engine lets us remember stuff that we didn’t actually know, which is sort of true as connectivity becomes ubiquitous. But what if there was a service designed specifically to retrieve information that we have already come across but didn’t explicitly log for future use, perhaps because at the time it didn’t seem relevant or useful.
At a higher level, that’s the concept behind Sentimnt, a personal and social search engine that at its simplest aims to answer the question: “Where did I read that?”. It does this by connecting to a user’s Twitter, Facebook, Google Reader, Gmail and Delicious accounts, indexing all of the updates, articles, and resources that are referenced so that they can be recalled later.
The site is in private beta but TechCrunch Europe has invites (see below).
“Every time we wanted to find that article, story, status or shared link we were sure we’d read somewhere, we had a problem: Googling it just wasn’t working”, says Sentimnt co-founder and CEO Khash Sajadi. “So we thought, what if there was a search engine that was mine and was searching my stuff not the whole universe.”
That idea in itself is pretty neat, but where things get a little more interesting is how the initial batch of private beta testers have been using Sentimnt. Aside from recalling specific articles, some users have been using the search engine as a buzz tracker to keep abreast of various topics. Cleverer still perhaps, another use for Sentimnt is as a bridge between what Sajadi describes as “suggestion supply and demand”:
“Our friends are suggesting places to go and things to do but not at the time we want to go somewhere or eat something”, he says. Using Sentimnt, however, it’s possible for a user to search their Twitter and Facebook streams to find those suggestions at a time when they are relevant.
Since the service is currently in private beta, it’s still very much a work in progress although I really like the concept. Setting up an account is very quick and straight forward and the UI work that’s already been laid is solid. Its biggest limitation is that for new users, Twitter’s indexing doesn’t go back that far historically (a Twitter API restriction) but that isn’t an issue going forward as Sentimnt will index future tweets and any links/web pages that they point to. Relevancy should hopefully improve too as Sentimnt tweaks its search algorithms but it’s a good start.
Overall, Sentimnt has a lot of potential, something that I haven’t felt about a new search product for some time. Talking of which, Google already offers its own social search features (although they are pretty half-hearted) and the company could obviously copy the concept behind Sentimnt. That’s something that the bootstrapped London startup would do best not to focus on and instead keep iterating what is already a promising product.
We have a limited number of beta invites – you can sign up here on a first come first served basis.