W3C home > Mailing lists > Public > www-archive@w3.org > July 2001

DanBri and DanC on anonymous resources RDFCore issue

From: Dan Brickley <danbri@w3.org>
Date: Fri, 20 Jul 2001 11:23:44 -0400 (EDT)
To: <www-archive@w3.org>
Message-ID: <Pine.LNX.4.30.0107201122500.30552-100000@tux.w3.org>

From private IRC, with permission.

<DanCon> how much longer are you in BOS? I think a model theory for RDF
would take about 2 hours for us to work out. Maybe if I explained it to
your satisfaction, you could explain it to the rest of the WG (and the
community)?
<danbri> i'm here another week, then a week west coast for sw/rdf stuff
<danbri> I don't think your 'textbook' crack is entirely fair, though I'm
perfectly happy admitting I'm still catching up on the KR reading list
<danbri> If you think it'd take 2 hours, maybe we could try sometime next
week. Any time after tues would do for me.
<DanCon> fair: maybe not. What frustrates me is: you understand FOL
perfectly well, but you go on about how "people understand 'there exists'
in different ways". Yes, people are confused. That doesn't change what FOL
is.
<DanCon> but your comment causes WG members to doubt that the KIF produced
by n-triples2kif.pl is anything less than 100% agreed as to its meaning in
the math, logic, and CS communities.
<danbri> FOL is a beatiful piece of machinery, and has power precisely
because it is diconnected from the grubby real world. The different
understandings are not w.r.t. to the internal workings of FOL, but about
how they map into the world.
<DanCon> yes, and those mappings are irrelevant to the issue of
rdfms-anon-resources
<danbri> Not if we want RDF/XML documents to mean something in the T-and-S
social/legal sense, for it to be clear what claims someone is making about
the world when they sign a piece of RDF data.
<DanCon> fuck fuck fuck! I just can't get you to stop muddling everything
together. yes, anon-resources *is* orthogonal to all those t-and-s issues.
<DanCon> my 2 hour estimate was wrong.
<DanCon> model theory is one thing. standard interpretations (bindings to
the real work) is another.
<DanCon> I apologize for my lack of patience.
<DanCon> This is a non-trivial subject.
* danbri stops sitting on his hands
<DanCon> It's just that I *know* you understand the technical
difficulties.
<DanCon> rather: I know you understand the technical solution.
<danbri> Yes, it's non trivial. And I do find it easier to see the
connections between the problem parts than the solution.
<DanCon> I just can't get you to separate the solution to this issue from
all the other issues in your head.
<danbri> So, are we both happy reading anonymous resources as 'there
exists'?
<DanCon> I am happy reading it that way (in the technical sense of 'there
exists' as in FOL/KIF)
<DanCon> I have expressed my happines reading it that way in running code:
n-triples2kif.pl
<danbri> And are you happy when I go on to say that this consensus doesn't
alone entirely fix the social/legal meaning of RDF/XML fragments that use
the anonymous resource construct?
<DanCon> absolutely. A solution to anon-resources does not solve all the
world's problems regarding social binding of RDF documents.
<danbri> Ok, I'm getting a sense for how we manage to talk past one
another. I've been hearing you as saying 'anons = FOL there exists, end of
story'.
<danbri> I should rephrase my position to sound less FUDish:
<danbri> I should say: "we can reduce the 'what do I mean when I use the
RDF anonymous resources construct' problem to the better understood 'what
do I mean when I use the FOL existential quantifier'"
<DanCon> QUITE!
<DanCon> BINGO!
<danbri> And you'll allow my mutterings about Quine and unicorns to live
in that 2nd problem space, which was all I wanted :)
<DanCon> any objections? hearing non, so ordered. yes. agreed.
<danbri> how'd you feel about copying this log (swearin' and all) into
<danbri> #rdfig for archival.
<DanCon> hard question.
<DanCon> er... ok, I'm OK to have it copied to public space
<danbri> Ok, copying from 'how much longer' to 'I'm OK'.
<DanCon> I was going to suggest starting with "so are we both happy..."
<DanCon> but very well.
Received on Friday, 20 July 2001 11:23:44 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:42:01 UTC