[microformats-discuss] Proposing RelSource

[Ryan King]

On Jul 12, 2005, at 8:33 AM, Andy Skelton wrote:
>> If anything it bolsters the case for rel="cite" (as opposed to  
>> some other
>> value like rel="source").
>>
>
> Oops, that was the exact point I was trying to make. I *do* plan to
> adopt and promote rel="cite" but I don't see a need to go through the
> hooplah of defining it as a microformat. Or am I missing something?

Depends on how you define 'hooplah.'

A generalization of citeRel would be this:

The XHTML 2 fragment:

<cite rel="foo">

is equivalent to this is XHTML 1

<cite class="relfoo">

(or rel-foo, I dare not suggest caMel-case :))

Perhaps we should define this as a microformat before moving on,  
since it seems elemental.

(I'm ignoring the XHMLT 2.0 vs HTML 5 issue for now.)

> As for whether people /should/ indicate primary sources differently
> than non-primary sources, i.e. hat tips, consider the definition of
> the word /cite/:
> "to quote or refer to as a precedent or authority"
> (Merriam-Webster's Dictionary of Law, (c) 1996 Merriam-Webster, Inc.)
> With that in mind, it would not be incorrect to cite both the primary
> source (authority) and a hat tip or "via" (precedent) with the same
> markup.

Who cares whether people /should/, they already /do/. We just trying  
to put a little structure to what people already do.


> Robots can construct "citation chains" to discover a likely primary
> source if a standard markup is used.

Right, but its a real bitch to do so.

It was done here http://www.hpl.hp.com/research/idl/papers/blogs/.  
And I've seen Adar give a talk about it and the majority of their  
work was making inferences about the citation chain.

Now, we can't force anyone to put these structured citations in, but


> The position of any resource
> relative to Primary on a chain could be of great value to search
> engines. It was this idea that excited me to propose RelSource in the
> first place. I'm just as happy with RelCite.

It's all good.

-ryan