On Thu, Jan 22, 2015 at 12:08:13PM -0800, Karen Coyle wrote:
> This is an absolutely first draft from the W3C Shapes group, so much
> could and will change, but it seems to meet the general expectations
> of that group. It would be interesting to compare it at this point
> with the DSP (which I haven't done yet, we just saw it this
> morning). If anyone has time to take a look, please report back on
> your impressions.
>
> http://spinrdf.org/ldomprimer.html
Thank you, Karen!
One first reaction: I really, REALLY dislike "Linked Data Object Model". The
potential for confusion is huge. The name seems to position the "model" as a
competitor to RDF in the Linked Data space.
That said, the tagline "an RDF-based modeling language that is compatible
with Linked Data principles and leverages some object-oriented concepts
to the Web" is very good.
I _really_ like the notion of "Data Shape", though "Shape" is only used in one
of the alternative names proposed:
LDOS (Linked Data Object Shapes)
LDOS (Linked Data Object Schema)
LDOS (Linked Data Object Specification)
LDML (Linked Data Modeling Language)
LDDL (Linked Data Definition Language)
LION (Linked Object Notation)
LOML (Linked Object Modeling Language)
LODL (Linked Object Definition Language)
LOMM (Linked Object Metamodel)
I slightly dislike seeming to limit the language to "Linked Data" (as opposed
to just "data"), but I strongly dislike the names with "Linked Object". What
is a "Linked Object" supposed to be?? "Linked Object Notation" is particularly
bad -- a language for notating linked objects? Or which of the following does
"Linked Object Modeling Language" mean:
-- (Linked Object) (Modeling) Language
A language for modeling linked objects
-- (Linked) (Object Modeling) Language
An object modeling language that is linked
If "linked" _must_ be part of the name, how about:
Linked Data Shape Language
Yes, it sort of collapses Linked Data Data Shape Language, which is what I
understand it to be in a conceptual sense, but it avoids the really bad,
ambiguous, misleading, or mysterious words Object, Modeling, Notation, and
Metamodel. Definition is not a bad word, as in: Linked Data Shape Definition
Language, though that makes the name longer.
What would really be lost if it were called simply
W3C Data Shape Language
??
This would reduce to a nice acronym, though "DSL", alas, is taken... [1]
That said, a three-letter acronym would be preferable to a four-letter
acronym. Or perhaps a two-letter acronym tacked onto "RDF":
RDF Data Shapes ("RDF-DS")
This name would make the basis in RDF clear without seeming to limit its
application to "linked" data and without introducing false-friend concepts
such as Linked Object.
Tom
[1] http://en.wikipedia.org/wiki/Domain-specific_language
--
Tom Baker <[log in to unmask]>
|