Re: Schema.org in RDF ...

On Sun, 2011-06-12 at 05:51 -0400, Alan Ruttenberg wrote:
> >> David, as you know, it is trivial to distinguish in representation the
> >> difference between an information object and a person.
> >
> > Correct.  And that distinction is important to some apps and not to
> > others.
> 
> I am glad we agree. We also agree that what is important is not
> germane to the technical question of whether ambiguity is necessary, I
> hope.

I don't know what you mean by that.  It is trivial to say that
foaf:Person owl:disjointWith foaf:Document or some such thing, which is
what I thought you meant in your comment above.  And it is trivial for
*some* people to arrange for their URIs to return 303 redirects.  But we
have seen from experience that it is *not* trivial for others to do so,
and it certainly has not been trivial to convince the less pedantic
practitioners of the need to do so.  And applications certainly are
germane to the issue of ambiguity, because applications drive the
modeling choices that we make in defining our resources.

> 
> >> I don't understand why you keep repeating this misinformation.
> >
> > Huh???  That's a rather rude accusation.  What misinformation do you
> > mean?  Please be more specific.
> 
> A comment is posted that raises a specific form of ambiguity. Your
> response is that ambiguity is inevitable. Yet we agree that in this
> case(and in other specific cases in which you have responded with this
> flavor of comment)  it is trivial to avoid. So your response is at
> best misleading or non-sequitor.

You are missing the point.  Sure, any *specific* ambiguity can be
avoided.  For *any* particular ambiguity it is possible to define our
classes in a way to avoid that *particular* kind of ambiguity.  But if
you focus only on fixing each particular ambiguity you'll miss the
forest for the trees.

It is *fallacious* to think that the ambiguity problem in general can be
solved by getting people to "fix" their data to be unambiguous, because:
(a) different applications have different needs; (b) the number of
potential ambiguities is *endless*; and (c) there is a cost involved in
unnecessary disambiguation.  In fact, in the httpRange-14 case that
schema.org raises, that cost is so significant that many intelligent
people have chosen *not* to disambiguate between the web page and the
thing that it describes.  The reality is that we *must* learn to deal
with ambiguity, and this is merely one example of it.

I do not think it is at all misleading to point out the larger issue
that underlies this specific case.  On the contrary, I think it would be
misleading to ignore it.

> 
> I'm sorry if my response came off as rude, however I am concerned that
> there be clarity in these conversations as the outcomes may turn out
> to be important.

Apology gratefully accepted, and I agree clarity in these conversations
is important -- but also difficult to achieve, as ambiguities keep
sneaking in.  :)


-- 
David Booth, Ph.D.
http://dbooth.org/

Opinions expressed herein are those of the author and do not necessarily
reflect those of his employer.

Received on Monday, 13 June 2011 06:51:20 UTC