[uf-discuss] hCard SOURCE

[Scott Reynen]

On Feb 23, 2007, at 9:51 AM, Brian Suda wrote:

> On 2/23/07, Scott Reynen <scott at randomchaos.com> wrote:
>> > Previously, you objected:
>> >> SOURCE is already used by X2V to indicate the URL at which the
>> >> current hCard is available. I don't think we'd be able to override
>> >> that at this point.
>> >
>> > I respect the work done by Brian Suda and others on X2V, but this
>> > argument doesn't pass the smell test.
>>
>> I agree, obviously.
>
> --- i think our definitions of SOURCE are different. As i am
> interpreting it; you are seeing SOURCE as, the canonical place to get
> the data, the fountain from whence it all came forth....

No, that's not what I mean.  We stopped talking about "canonical" a  
long time ago.  We're just looking for something *better* than the  
current hCard, and SOURCE seems like a good way to find it, provided  
the SOURCE contains an hCard for the same person, which we can  
indicate with UID.

> i am seeing SOURCE as where this hCard/vCard come from so you can
> trace it back down.  Like a CITE, from where did you extract this data
> from?

Right.

> THAT is why it is not controled by the user, but instead gleaned
> from where the hCard was extracted.

Huh?  As a user, I use <cite> all the time.  Why can't I tell you  
where I got some hCard information from in the same way?

> If someone showed me a vCard with incorrect data, i would say, where
> did you get that (from what source?). If you override that with a
> user-defined URL for SOURCE then you could never trace issues down.

I'm just talking about me saying "I got that from A" and then A  
saying "and I got that from B."  What A says doesn't change what I  
said.  It just changes where you probably want to look.  You're  
welcome to stop at A if you don't trust or don't care about A's  
pointer to B, just as X2V does now.  That doesn't break the chain of  
pointers at all.  Anyone else loading A can still evaluate A's  
pointer to B themselves later if they choose to do so.  No one is  
suggesting all implementations must follow the entire chain of SOURCE  
pointers.

> A
> user-defined SOURCE URL could point somewhere that is NOT where the
> data came from.

Here's a real difference of opinion.  You want to restrict SOURCE to  
something machines identify, but not people, because you find  
machines more trustworthy.  I think SOURCE should be available for  
people to use as well.  Where does the vCard spec say SOURCE should  
be determined by machines only?  If it's not in there (and I don't  
see it), making that assertion is changing the spec, which we should  
avoid wherever possible.

More generally, trust is no more related to this problem than any  
other microformat proposal, and the issue doesn't belong in this  
discussion.  If you'd like to discuss the trustworthiness of data  
(e.g. my ability to lie about my SORUCE as well as my UID, FN, TEL  
and, well, everything else), please start a separate thread.  The  
issue gets raised as an objection to almost every proposal on this  
list, so I'm sure there's plenty of interest.  But let's try to keep  
this discussion on-topic.

> Again from the vCard RFC:
>>  If the SOURCE type is present, then its value provides information
>>  how to find the source for the vCard.
>
> I think that small definition is the hang-up on how we are
> intrepreting SOURCE. When the definition use the the term to define
> itselves no one wins.

I'd say it's not really using the term to define itself, because  
there's a difference between SOURCE and source (a subtle  
capitalization difference I believe is standard throughout RFCs).   
The former is defined in vCard.  The latter is defined in any English  
dictionary.  For example:

"a place, person, or thing from which something comes or can be  
obtained"

So:

>>  If the SOURCE type is present, then its value provides information
>>  how to find the place from which the information comes or can be  
>> obtained for the vCard.

That's a pretty simple definition, and I think we should stick to  
it.  When I want to point to a place where the information in a given  
hCard comes from, I should use SOURCE.  When you want to find where  
the information comes from, you should follow that SOURCE.  Whether  
or not you want to replace the first hCard with the SOURCE, after  
using UID to determine they represent the same person, is up to you  
and the needs of your implementation.  I see no need for X2V to  
change at all to support Joe's proposal.

I think defining SOURCE more narrowly than vCard does (i.e. machines  
only), making Joe's proposal impossible, without a compelling  
argument for doing so, contradicts the microformat principles.

> I read that NOT as canonical source, but the SOURCE of the current
> data. The place to go to track down where this came from.

So do I, and I don't see Joe or anyone else using the word  
"canonical," or implying anything like that, so I'm not sure where  
you're getting that impression.  I think we're all agreed SOURCE is  
not canonical.  The only disagreement I see is whether or not we  
should allow people to use the property, which strikes me as an odd  
thing to oppose.

Peace,
Scott