W3C home > Mailing lists > Public > public-grddl-wg@w3.org > February 2007

Re: Some spec comments (defn transformation property etc.)

From: Dan Connolly <connolly@w3.org>
Date: Mon, 19 Feb 2007 16:03:59 -0600
Message-Id: <5e471e015d6c9179b8555491ff4f128f@w3.org>
Cc: "GRDDL Working Group" <public-grddl-wg@w3.org>
To: "McBride, Brian" <brian.mcbride@hp.com>


Revision 1.223  2007/02/19 22:02:06  connolly
- Specify how relative URI references work while
   explaining the titleauthor.html example

- move defn transformation property before the first rule where it's 

- note transformation properties need not be total


On Feb 19, 2007, at 11:33 AM, McBride, Brian wrote:
> [...]
> [[
> Note that this serialization of the graph contains a relative URI
> reference.  The base URI for interpretting relative URI references in a
> serialization of a graph produced by a GRDDL transformation is the URI
> of the source document.
> ]]


> Reading the rule jars a bit because the definition of TP comes after 
> the
> rule - better to move it before.


> This raises two questions in my mind.
> 1.  Should we say something about what happens when the input document
> is not in the domain of the transformation - e.g. the code goes belly 
> up
> when executing.

Yes. I noted that in the section on Transformation Algorithms. Hmm...
the result is a little choppy, perhaps.

>   This might affect the section on conformance labels.

I don't think so.

> 2. Is a transform necessarily functional on the input document alone?
> For example, a transform that includes a triple containing the time at
> which the transform was run is not a GRDDL transform?

Indeed, by design, it is not. See also earlier discussion:

if a transformation is not well-defined, all bets are off, right?

> Hey - I got to the end of section 2.


Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Monday, 19 February 2007 22:04:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:39:10 UTC