Joshua Tauberer wrote: > Richard Cyganiak wrote: >> I noticed today that Tabulator actually can read N3. I didn't know >> this before, I thought it could only do RDF/XML. At this point it >> might be a good idea to encourage broader support for N3 at the client >> side. I blogged about this here: >> http://dowhatimean.net/2008/03/tabulator-does-n3 >> >> Do you all think that N3 is ready for prime time? Can it replace RDF/ >> XML? What stops you from publishing your data in N3? > > I publish data in N3, and for that matter maintain a C# parser for N3, > but I worry in general about whether the aspects of N3 beyond Turtle --- > formulas and paths, if not other things --- are going to be implemented > universally. In the former case, how to interpret them semantically and > represent them in RDF/XML, and in the latter case whether it will be > implemented everywhere.
I find myself resonating with Joshua. As much as I like RDF/N3 or RDF/Turtle over RDF/XML, when you take them for a spin you realize that when people say N3 they really mean Turtle (and they don't even know that N3 is more powerful than turtle). They just see a more friendly syntax and call that "n3"... most people don't know that CWM uses some of the N3 features to perform something "active" on top of the RDF model, while Turtle is completely passive and a subset for it. In fact, Sesame has a turtle parser but not a N3 one (but has a both a turtle and N3 writer, mostly because every Turtle model is a valid N3 model as well). The fact that N3 has more features than turtle and that it's not very evident what these features are and what tools support them is, IMO, hurting the adoption of this much prettier/readable/useful syntax a great deal. Today, if you really want interoperability between systems, RDF/XML is a a *much* safer bet, despite the fact that it's much less readable, much more verbose and much more fragile (you can't append XML files together and still get an XML file, while you can with other RDF syntaxes.. moreover, most xml parser can't read non-well-formed data, so a single unescaped " in an attribute field might break the entire data stream.. oh yeah, that happens.... a lot!). -- Stefano Mazzocchi Digital Libraries Research Group Research Scientist Massachusetts Institute of Technology E25-131, 77 Massachusetts Ave skype: stefanomazzocchi Cambridge, MA 02139-4307, USA email: stefanom at mit . edu ------------------------------------------------------------------- _______________________________________________ Linking-open-data mailing list [email protected] http://simile.mit.edu/mailman/listinfo/linking-open-data
