Re: Comment on "Meaning and the Semantic Web"

On Jun 2, 2004, at 9:13 AM, John Black wrote:
[snip]
> But my real point is the symmetry between correct publishing behavior
> and correct interpreting behavior.
>
> So lets change the scene a little:
[snipped silly caricature that not only makes Peter and me out to be 
Machiavellian morons, but doesn't even vaguely hook up to anything 
we've said]

You might take a little time to edit your posts before sending them. 
Restating an example you just posted not a full day before, only this 
time incorporating your interlocutors in a fairly derogatory way and 
yet not otherwise augmenting or elaborating the example is just a 
waste.

> The point is that there are use cases where it would be critical
> that an interpreting agent be required to discover and report
> the actual meaning of a set of published documents.  And this
> might be aided by giving URI authors facilities to specify what
> that meaning is.

We author documents, not URIs.

I want my documents to be largely under my control. I prepared for some 
leeway in interpretation (e.g., looking at a document purely as well 
formed XML rather than as the particular PSVI I intended), but I don't 
think that every use of a URI in document *content* should give that 
URIs owner licence to insert whatever into my document. I'm not adverse 
to that entirely, obviously, since this pretty much is what owl:imports 
gives me.

It's like the difference between an <a href=.. and an <img src... You 
don't always want transclusion.

None of this has a WHIT to do with some intermediary inserting or 
altering content "for my own good". Nothing. Nothing at all. Not even a 
little. Let go, ok?

There is an interesting question of what aggregators and other 
republishers might reasonable want or be expected to do. But that is a 
separate issue.

Cheers,
Bijan Parsia.