Re: ISSUE-238 (POWDER): Use of Powder to "label" CT proxies and Server Preferences for CT

Just a quick response to this being raised.

The debate about re-instating the HTTP Link Header has been a constant in
my life for the last 3 years. It is currently being discussed in the
Semantic Web Coordination Group and as a result I am now beginning to
gather evidence in support of an internet draft (now expired) written by
Mark Nottingham a couple of years ago [1].

_any_ proposed or current usage of HTTP Link - i.e. the equivalent of the
HTML Link element but sent as an HTTP Header - is very relevant and,
whether related to MWBP or not, if you have a use case, do please let me
know. I've spoken to Jo about the possible application within CT.

Also, I am happy to say that there will be new public working drafts of
the key POWDER docs by the end of next week. These will set out the 2
stage approach we're adopting. Operationally, POWDER will be a constrained
dialect of XML which can be processed using XML tools. However, it also
has an associated GRDDL Transform (that uses XSLT) that will render the
data in RDF/OWL for more sophisticated Semantic Web processing,
inferencing etc.

The POWDER WG is holding a stakeholder event at the GSMA in London on 18th
March [2] - MWBP members wishing to attend should please contact me or
Matt Womer. The basic message will be "you can improve your service and
deliver more personalised content - and therefore make more money - using
POWDER." Given the co-sponsorship by GSMA, you can bet that the discussion
will have a strong bias towards the mobile world.

Cheers

Phil.

[1] http://www.mnot.net/drafts/draft-nottingham-http-link-header-00.txt
[2] http://www.w3.org/2008/01/29-powder-outreach-agenda.html


>
> ISSUE-238 (POWDER): Use of Powder to "label" CT proxies and Server
> Preferences for CT
>
> http://www.w3.org/2005/MWI/BPWG/Group/track/issues/
>
> Raised by: Greg Aaron
> On product:
>
> ACTION-668
>
> Open an ISSUE on how POWDER might be used
> a) to label CT proxies (by use of a URI in the comment field of the Proxy:
> header)
> b) to label site preferences for content transformation (by use of
> standard POWDER mechanisms)
>
> Given that we can't extend HTTP, POWDER might be just the ticket to solve
> some of the questions of how the server and the CT proxy can get to know
> more about each other.

Received on Wednesday, 27 February 2008 17:26:55 UTC