You (and the web application you’re writing) probably know whether you mean Starbucks the company (as in this Wikipedia article), or one of the thousands Starbucks coffee shops around the globe, many of which uberblic fetches location data about from Foursquare. And so if you know what you mean, then tell our Search API so that we can help you find the right uberblic entity.

Your Search API query may contain attribute/value pairs that correspond to the metadata (and data) about an entity in uberblic, like source (which is the metadata property for an entity’s data source). So when we query for “source:[enwikipedia] Starbucks“, we get back only entities extracted from Wikipedia, and the company entity ranks first.

To make sure the results are really only companies, we add a type restriction: “source:[enwikipedia] type:[uo:Company] Starbucks“. (Note: the “uo:” in “uo:Company” is the prefix for the uberblic ontology URI, i.e. http://uberblic.org/ontology/).

Let’s look at the Starbucks coffee shops. We find those by searching for “source:[foursquare] Starbucks“, and as this query will return thousands of results, I’ll restrict it to find only shops in Berlin: ”source:[foursquare] Starbucks city:[Berlin]“. However, this will again find all Starbucks in any city named Berlin, so I may want to use the ID (the last part of the uberblic URI) of my Berlin in Germany: “source:[foursquare] Starbucks city:[ab940ab4-cefe-4278-9e9d-404ab4a3a246]“.

Or you’re looking for that film with Bill Murray and Scarlett Johansson… what’s its name? “starring:[Bill Murray] starring:[Scarlett Johansson]“. Copy that into the search box at http://platform.uberblic.org and press the spacebar, and our autocomplete-enabled search box will already present the result to you…

That’s it for a first look at the uberblic Search API. Please let us know what you think.

With this new version we introduce some exciting new features that will make it easier for application developers to use the integrated web data, as well as to make sure their repositories are kept consistent and up-to-date. Uberblic.org provides a reliable middleware layer and proxy for developers to build production-ready applications over curated web data.

But first of all: the data. We have integrated more and more data sources and we provide data dumps of our repository. Our data sources now include Wikipedia (in English and German), Musicbrainz, Geonames, Foursquare, MovieDB, LastFM, Freebase, several BBC datasets, Drugbank, and Diseasome. We fetch updates from these sources as fast as they let us, for example Wikipedia updates in real-time, Musicbrainz every hour, and many other datasets several times a day. And from Wikipedia we extract high-quality article abstracts in real-time too.

Then: the new APIs. These are very important for us. Having all this consolidated and reconciled data, we can and do provide developer APIs on top. The new Update-Feed API enables people who mirror our repository or parts of it to fetch updates. The Consolidation Feed API helps to keep data links consistent. The Search API and the Lookup API help to find resources. And of course there is Linked Data access and a SPARQL endpoint.

Why version 1.0? Because this is just the beginning. We have more very exciting new APIs, such as our Doppelganger API, waiting to be published, a new platform dashboard, and several applications showcasing the power of integrated web data. So over the next weeks and months we will write on this blog about all the different aspects of the platform, giving further overviews of specific features, and keeping you updated with what’s happening. We’d like to keep in touch with our users as much as possible, and will post to this blog at least once a week. So stay tuned.

But for now, start using the new APIs and data sets, and please let us know what you think of them by joining our mailing list or sending us an email to feedback at uberblic.com