Re: Resource Token Exchange

On Sat, Mar 5, 2011 at 3:57 PM, Gregg Reynolds <dev@mobileink.com> wrote:

> Hi WG,
>
>
Hi again,

My apologies for not properly introducing myself and providing a little more
background info as to why I posted my notes on RDF.

I am Miram Abacha, wife of the late Sani Abacha.  No, not really.  My name
is Gregg Reynolds.  I work as a low-level paper pusher at
NORC<http://www.norc.org/homepage.htm>,
and do some programming on the side (although I spent the '90s working as a
programmer.)  I'm involved in two projects that use RDF for linguistics.
 One uses RDF to model Afro-Asiatic morphology; the other is a project to
translate  Sibawayhi's great Kitab, a grammar of Arabic (written approx. 790
CE.) and to encode traditional Arabic dictionaries, among other things.
 Lots of interesting issue come up in using RDF for linguistic modeling, but
I'll spare you the details, save one:  it works better /not/ to treat RDF as
a Knowledge Representation language, in order to avoid encoding ontological
commitments, which are always theory-laden.

I am not a troll, and you'll be relieved to know I won't be badgering you to
adopt my admittedly unorthodox understanding of RDF.  Also, please don't
take my criticisms of the current docs personally.  I was on the XSL version
1 WG so I know how hard it is and I appreciate your work.  On the other
hand, if you really don't like something in a public standards doc I think
it's better to just say so directly.  Maybe with a little rhetorical
flourish now and then, heh heh.

I considered writing a series of notes as feedback on the RDF standards
docs, but I fear that would inevitably end up as a disjointed collection of
notes badgering to you do things my way, rather than a clear explication of
alternative approaches.  So I thought it would be better to try to write
something up explaining and illustrating my ideas and let it stand or fall
on its own, in hopes that it might prove useful.  Ideally, I'd write an
alternative definition of RDF, but the chances of my finding time to do that
are approximately nil; hence the collection of notes.  Anybody who finds
something interesting can run with it, otherwise no harm done.  I do hope to
find time to explain the Category Theory stuff a little more, with
definitions and diagrams for those unfamiliar with CT, but that probably
won't happen for a few months if ever.

[Which reminds me, I'd like to find people might be interested in exploring
the ideas sketched in my note, but this list is obviously not the
appropriate forum.  If anybody can recommend an appropriate mailing list I
would appreciate it.]

What prompted me to submit notes now is Pat Hayes' note on tokens; I was
pretty excited to find that somebody else was thinking along those lines.
 Another thing is that, although standard RDF argot  had annoyed me for
years I could never quite put my finger on why.  But in February I exchanged
some messages with the SPARQL WG and suddenly the scales fell from my eyes,
and I knew exactly where the problem was - that's mostly the stuff about
meta-language, object language, abstract syntax, etc. in my notes.  And
since the RDF-WG is just getting started I figured now was as good a time as
any to foist my ideas on an unsuspecting world and take my lumps.

I don't really expect anything in particular from the WG.  My interest is
mainly in the challenge of technical writing; it's an under-appreciated art
and I like the challenge of mastering the wide variety of technical fields
implicated in RDF and translating it all into elegant technical writing.
 (Believe it or not it is much like translating 8th century Arabic.)  I also
happen to think that the obscurity and looseness of much of the language
usually used to explain RDF bears a large share of the blame for RDF's niche
status.  For example, the fact that after all these years we still don't
have a good definition of what a resource is (as evidenced by the ongoing
stream of blog entries etc. trying to explain the idea) is pretty good
evidence that something is amiss, in my view.  So if I can come up with
language (and concepts) that meet the criteria of clarity, simplicity,
conciseness, rigor, etc., maybe that will be useful for the revision of the
standards.  That's about it, really.

Cheers,

Gregg