SIOC-based microblogging
As John Breslin already detailed last week, we (Tuukka Hastrup, Uldis Bojars, John and myself) recently work on SMOB, a semantic microblogging architecture, that will be presented at next SFSW workshop, co-located with ESWC in Tenerife.
While there have been a lot of buzz recently regarding ways to provide open alternatives to Twitter, then fun thing is that we worked on this in last January while I visited DERI. Our main goal was to show how Semantic Web technologies could provide an open-platform for such way of publishing content, mainly using FOAF and SIOC. Moreover, one of our aim was also to demonstrate how such technologies can provide users a way to control, share and remix their data as they want, not depending on a third-party service, a goal also shared by the dataportability project. In that way, SMOB-published data really belongs to the user that wrote it. Indeed, while SMOB servers (which display a faceted view of agregated posts as you can see on the demo server) store data in their local triple-store, this information (i.e. each update) is hosted on the client side and available in RDF.
At the moment, the complete updates dataset is public, and can be browsed with any RDF browser as the picture below shows (with Tabulator) but we plan to introduce more advanced authentication and privacy issues, in which OpenID could have a role to play.
Users can parse it as any RDF data, mash-it up with other information, eg their FOAF profile (SMOB allows to re-use existing FOAF profile as the foaf:maker of each update), or any other RDF data. And most important, if a SMOB server closes, they still own their data.
Moreover, since SMOB content is SIOC-based, it becomes part of the SIOC-o-sphere, and could be merged with your other social media contributions (from any SIOC application) and discovered thanks to recent APIs and WordPress plug-in introduced by Sindice (automatic PTSW / Sindice pings will be soon in the code repository). This is one more advantage of getting a common semantics to model your data wherever they come from.
Finally, we just introduced in SMOB a way to publish not only to a dedicated server, but also to Twitter. So that you’ll get a real-time, self-hosted and long-life archive of your twits in RDF. Isn’t that cool ?
Comments
-
Current highlights
J’ai enfin pu tester Smob (client) et c’est vraiment super ! Félicitations à toute l’équipe.
Plein de bonnes idées, comme l’utilisation des hashtags reliés au linked Data…
Petite question : dans la mesure où vous utilisez ARC, prévoyez-vous d’indexer les fichiers RDF générés dans le triple store de ARC ? Cela permettrait de récupérer plus facilement les données.
Et, sinon, une petite suggestion : que dirais-tu de monter un smob server du Web sémantique français ?
J’ai installé le smob server et j’ai eu la réponse à ma première question. Je confirme, c’est super !!
Merci pour les compliments
Pour ARC, on se pose la question d’avoir un client léger comme on ce moment, ou qui nécessite une base de données pour avoir un entrepôt local également.
Mais les données sont maintenant disponibles en RDFa sur le client, et ils sont bien entendu indexés côté serveur (avec un endpoint accessible)
Pour un smob ws-fr, bonne idée, on peut voir ça avec Eric pour le serveur, et lancer un thread sur la liste pour promouvoir son utilisation.
Bravo !
Longue vie à SMOB. Rendez ça un peu plus sexy (user-friendly) et go-to-market. A mon avis la plus-value du web semantic dans le micro blogging est énorme !
[…] dall’introduzione che ne fa Alexander Passant, uno dei realizzatori ( a very good job Alexander!! ): our main goal was to show how Semantic Web […]