So how interesting is the course Web of data on Coursera (see previous post)? Well, first let me put this question in a broader context. I’m interested in the decentralized web from a learning perspective. One of the most fascinating projects in that regard is Solid, empowering people to separate their data from the applications that use it. It allows people to look at the same data with different apps at the same time.

I’m interested in the technology behind this project, and it seems this technology is rather complex. Look at this roadmap on GitHub for developers wishing to prepare themselves for Solid, it seems daunting for people who are just starting out. I guess it’s far less challenging for experienced web developers since Solid builds on web development fundamentals (think HTML, CSS, JavaScript and the major frameworks), and also on linked data technologies such as the Resource Description Framwork (RDF) and SPARQL, an RDF query language. RDF and SPARQL are less familiar to most developers.

The above mentioned course gives a broad yet thorough overview of RDF. It starts with deceptively simple topics such as the difference between web and internet and the separation between presentation and content. The course prepares you for another vision on the web, not as on a collection of pages, but as a collection of data which can be linked.

We discussed fascinating tools such as the Discovery Hub, an exploratory search engine built on top of Wikipedia and more precisely on top of the data extracted by DBpedia. I searched for Linked Data on Discovery Hub and it returned me 227 topics and 10 kinds of topics. One of these topics is important for the Discovery Hub itself: DBpedia, a project aiming to extract structured content from the information created as part of the Wikipedia project. This is linked with Wikidata, which is more and more replacing the infoboxes in Wikipedia. DBpedia wants to use the data provided by Wikidata and also help the project to become the editing interface for DBpedia.

We also learned using curl, the command line tool and library for transferring data with URLs and we learned about the Web service OpenCalais, which automatically creates linked data from the textual content one submits.

The second week of the course was (even) more technical and delved into the composition rules for RDF. The context and abstract syntax of RDF 1.1 is undoubtedly of importance for all those who actually will build things, but I pretty soon decided that I stick on a more generalist level – I simply don’t have the time nor the inclination to become a developer.

For those who want to try things out, part of the whole syntax hell is automated. RDF Translator is an online conversion tool enabling to transform RDF statements from one RDF syntax to another, e.g. from the RDF/XML syntax to the N3/Turtle syntax (and yes, the course stimulates you to study all those syntaxes,). There exist also web services to visualize data using these technologies such as Visual RDF. This stuff is not exactly slick nor user friendly, but then again I guess the core audience is academic.

All this is very interesting and helps me to understand what linked data is about and what kind of progress is being made. As usual I put myself in the center of the learning experience, focusing more on certain aspects and neglecting others for now – which presumably means I won’t pass all the tests – but who cares?