Drupal 7 semantic by default, when will Alfresco follow?

My university friend Stéphane Corlosquet has spent the last few months adding a very exciting feature to Drupal for its imminent next release: Drupal 7 will expose your website's structural information as RDFa, by default!

This has huge implications. Drupal being one of the most popular CMS, it handles a significant proportion of the Web's information. So Drupal 7 will effectively make the Semantic Web much bigger. Furthermore, a website's manager will now be able to define the website's ontology based on existing ontologies, which means each Drupal website will now be both a consumer as well as a producer of semantic information.

For website owners, an immediate result is that their website will be better understood by Google. But the Semantic Web is much more than an SEO trick. It is a way to make information more useful, more exact, and to make it understandable by computers.

So how long before Alfresco follows? The open-source Enterprise Content Management System should make sure the content it manages is understandable by both humans and machines. A company's way to structure information should not be kept in the Data Dictionary, it should be exposed as RDF Schema. The metadata, including aspects, should be accessible via RDF in addition to the usual REST/Java/JavaScript/JCR/CMIS APIs.

Semantic Web for everybody

Suppose you are about to launch a great new Semantic Web application, but you don't have much hardware. The bottleneck of most Semantic Web applications is the triplestore, which can be seen as the equivalent of the SQL database in normal web applications. Companies can afford the hardware to host a big one, but volunteers (such as mashup and open source people) can not.

That's where free RDF data hosting comes in. Yes, you can get a triplestore for free. Couple that with a traditional LAMP stack (there is a lot of free LAMP hosting) and you can deploy a Semantic Web application for ¥0 ! A company called Talis is now offering free RDF data hosting, complete with remote SPARQL querying, provided the data is public domain. I hope other companies such as OpenLink will follow. I received an account from Talis' programme manager

♫

Leigh Dodds and tried it right away, loading and querying data, with success. They also provide an API that seems quite interesting, but I did not try it yet.

This will be extremely useful to open-source/open-data groups who are run by volunteers and want to enter the Semantic Web scene. In particular, I expect this will lead to an explosion in the number and quality of mashup websites on the Web.