[uf-discuss] Identity-related hCards?

[Scott Reynen]

On Feb 20, 2007, at 1:05 PM, Ryan King wrote:

> However, the simpler problem of two hcards representing the same  
> person (or organization) should be solved first, because it is a  
> simpler problem, with a simpler solution (which may not require  
> adding any properties to hCard).

The implication I've always taken from the "simpler problem first"  
principle is that this makes the more complex problems easier to  
solve.  But this is only true if the simpler problem actually helps  
solve the more complex problem.  Related hCards is also a simpler  
problem than book citations, but we don't half that discussion  
because solving the simpler problem has no descernable impact on the  
more complex problem.  I think what's confusing those of us  
interested in the more complex problem of authoritative hCards is how  
solving related hCards would actually help solve that problem.

And maybe that's because what you're describing is actually more  
specific that "related hCards" implies.  It seems here you're just  
talking about a single relationship: identity.  My brother's hCard is  
related to mine, but those aren't two hCards representing the same  
person.  If representing the same person is your primary use case, I  
don't think "related hCards" communicates this very clearly.   
Similarly, "canonical" didn't communicate the concept of  
"authoritativeness" very clearly, because it suggested much more than  
we needed.  I'm not sure what would be a better word to communicate  
identity relationships, but I think that would clear up a lot of  
confusion if the above is the full scope of the problem you're trying  
to solve.  If, however, you are trying to also come up with a way for  
me to indicate that my brother's hCard is related to mine, I don't  
see how that will move the authoritative hCard problem forward at  
all, and it seems to unnecessarily complicate the issue.  Identity is  
a simpler problem than relationships in general, right?  And it's  
also a clear building block toward authoritative hCards.  So what  
about focusing on that simple problem instead of a more general  
"related hCards"?

Peace,
Scott