[foaf-dev] Vocabulary for internal structure of text documents

Mauro Bieg mauro.bieg at gmx.net
Sun Oct 3 15:59:30 CEST 2010


Thank you both for your quick reply!

@Dan: yes, the infoset spec is too low level.
>> Re foaf, open q is being clearer about patterns for describing how small instances of foaf:Document can fit together to make bigger ones.
That sounds interesting.. but what is open q?

@Simon: I'm thinking of building a semantic CMS where the only database is basically an RDF-triple store. The user would then enter all the data in a semantic way and could later (re-)combine it to form different views. For example you could create a new event:Event with a property event:time and this information would once be rendered as an iCalendar file and another time be included in the web page for that event. And the text for that page would be a  foaf:Document that has as a foaf:primaryTopic that event, etc. 
I hope by making the user interface slick enough, people will actually find it easier to enter their data semantically as opposed to the WYSIWYG approach. Obviously I'd like to do that also for the text itself, but there you run into this "is RDF really suited for text documents?"-thing. As you said, in XHTML or similar some things are much easier, for example inserting a <reference> tag in the middle of a paragraph that will be rendered as a footnote later. But it would be kind of a shame to have all this data in RDF and then when it comes to structured text break that and just use XML because we're all so used to it. But maybe it will turn out to be the better solution.
So yeah, that's what would be my use-case :) I have just started thinking about this idea and I'll see whether the approach is feasible and stuff.


Am 03.10.2010 um 13:53 schrieb Simon Reinhardt:

> Hi
> Mauro Bieg wrote:
>> Hello,
>> Does anyone know of a vocabulary or ontology that describes the internal structure of a text document with headings and subheadings? Basically like HTML or LaTeX but in RDF. Or do you think RDF isn't suited for that? I think it would be great to have that internal information of articles, books, etc. available as semantic data too, and not only the metadata. You could generate table of contents, reference a paragraph of an article or a figure directly by its URI, etc..
>> I'd be glad to hear your thoughts because I'm thinking of building such a vocabulary if it doesn't exist already. So why hasn't this been done yet or what are the potential problems?
> I think formats like XHTML are very well suited for this purpose already.
> What's your use-case? Do you need to aggregate such data over a collection of documents? Which might be in different formats? Find connections between documents? Infer new information?
> If not then the formats we have will do the job just fine. :-)
> Regards,
> Simon

More information about the foaf-dev mailing list