Re: Blank Nodes Re: Toward easier RDF: a proposal

> Have you tried to express your use case in SPARQL?

Yes, and that would be possible in principle but with a very
verbose and not very pretty query and a heck of a lot of 
post-processing (or maybe some extension to SPARQL that
enables randomisation of results in specific ways, but I
can’t think right now what that would look like).

I don’t want to side-track the discussion with the details
of the use case, but it’s very nice to be able to write,

([ a Operator ] [ a Promoter ] [ a CodingSequence ] [ a Terminator ])

and have that mean exactly what it says. It’s a partially
specified gene fragment. It’s actually hard to think of a more
succinct way of writing it. Expecting people to jump through
hoops to write something different from what they mean
increases friction unnecessarily.

Also, RDF lists increase friction any way you cut it, but
that’s for an email with a different Subject header.


Received on Friday, 23 November 2018 01:10:21 UTC