Re: out-of-band transformation information

At 08:11 AM 10/26/2006 -0500, Dan Connolly wrote:

>We recently got this comment suggesting that GRDDL should
>provide a way to give out-of-band transformation information...
>
>specifying GRDDL transformation for document with no transformation
>attribute? Bob DuCharme (Wednesday, 25 October)
>http://lists.w3.org/Archives/Public/public-grddl-comments/2006OctDec/0010.html
>
>Brian McBride made a similar comment back in January...
>
>"I think there are at least two things missing:
>
>2) a way to describe a transformation on a (set of) pages without access
>to the pages themselves or their schema. "
>  -- Brian McBride, 27 Jan 2006
>  http://lists.w3.org/Archives/Public/www-archive/2006Jan/0049

I think that I mentioned this idea to Dan once too. He was very quick to 
point out
that allowing out-of-band nomination of GRDDL transforms breaks the GRDDL 
contract.

I am not at all confident that GRDDL will always be used such that the 
document author or
the namespace author are asserting a contract or promise to the recipient. 
In particular,
I am not confident that user of a namespace will know or care that a GRDDL 
transform
exists for that namespace or profile.

However, having said that, I think it important to trot out my position on 
GRDDL as a
precedent for the development of parallel technologies which utilize the 
GRDDL technique
but use different property names and make entirely different promises to 
the consumer.
It is my position that getting GRDDL to REC status will open the way for 
speedy development
of a new class of W3C spec, built in the likeness of GRDDL. Just think how 
easy it will be
to write a spec for out-of-band transformations once we have detailed the 
mechanism for
doing it in-band.

If we spend too much time debating whether GRDDL should do things that are 
not part
of its original mandate, then we will delay the ultimate release of GRDDL 
the REC.
I want to avoid such a delay.

Regards,

Murray

P.S. At one point I had written text for the introduction that expanded on 
the Gleaning metaphor.
Something about "harvesting RDF data from the field of XML documents. Dan 
didn't like it at the time.
But I come back to it today.

Gleaning, according to Wikipedia, "is the collection of leftover crops from 
farmers' fields after
they have been mechanically harvested or on fields where it is not 
economically profitable
to harvest. Often gleaning is practiced by humanitarian groups which 
distribute the gleaned food
to the poor and hungry."

Part of the subtext that I glean from this is that the Farmer owns the 
fields and
grants permission for groups, humanitarian or otherwise, to collect 
leftover crops.

By analogy, I would think that out-of-band transformations are more akin to 
pillaging
than gleaning. No permission is being sought. No commission paid to the farmer.
And the crops' provenance comes into question. After all, do I really want 
to eat
food whose provenance is in question? Or data?

So, in spite of Dan not wanting to use the agriculture metaphor in the 
spec, I think that
it still applies and supports his position.