Re: issue-dbooth-9a: GRDDL should be usable in a messaging pipeline

Hi David,

I thought I would respond to the second half of your comment concerning 
time.

Note this is not a response on behalf of the WG, but I hope that if I 
can explain what the spec says, and you find that to be, if not ideal, 
at least minimally satisfactory, that the WG would be happy.

The key paragraph seems to be this one:

Booth, David (HP Software - Boston) wrote:
> For example, suppose an information resource, ir,  produces a different
> representation each time it is queried -- the current weather in Oaxaca,
> for example -- and I have two XSLT scripts that I use to glean RDF from
> them: one extracts the temperature (getTemperature.xsl) and the other
> extracts the humidity (getHumidity.xsl).  The final RDF should be the
> combined result of applying getTemperature.xsl and getHumidity.xsl to
> the *same* representation.  But the spec does not define merged GRDDL
> results for a particular representation, it only defines them for an
> information resource as a whole, which could have a jumble of
> temperatures and humidities from different days.
> 

On my reading, you are correct to say that:
[[
If F and G are GRDDL results of IR, then the merge [RDF-MT] of F and G 
is also a GRDDL result of IR.
]]
permits the merging of GRDDL results from different representations of 
IR over time, and that fora resource that has time-variant 
representations, particularly those that vary relatively quickly, this 
is likely to be the source of confusion.

However, I point out that a GRDDL aware agent is only explicitly 
required to merge GRDDL results for a single representation:
[[
given a URI I of an information resource IR, and an XPath node N for a 
representation of IR, a GRDDL-aware agent should:

    [...]
    3. Merge those GRDDL results.
]]

This provides a clearer RDF graph corresponding to a single 
representation of a resource.

Thus, the step of merging GRDDL results from different representations 
of the same resource, is licensed, but not required, in the 
specification. Thus an application, or a GRDDL aware agent in a user 
configurable option, may perform this step, and presumably will be more 
inclined to do so in those cases where it is useful, rather than those 
in which it is confusing.

This reply also suggests that the more general point you make that the 
spec would have been better if written in terms of representations 
rather than resources, is perhaps not as different a design as you may 
have thought. The normatively required behaviour in the specification, 
is that of a GRDDL aware agent, and this *is* defined in terms of a 
single representation, does this ameliorate some of the problems that 
concern you?


Jeremy

wistfully hoping that this comment is adequately addressed :)



-- 
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

Received on Tuesday, 12 June 2007 17:26:30 UTC