W3C home > Mailing lists > Public > www-p3p-public-comments@w3.org > April 2001


From: Lorrie Cranor <lorrie@research.att.com>
Date: Sun, 29 Apr 2001 16:10:34 -0400
Message-ID: <01ad01c0d0e8$7a2662e0$3a06cf87@research.att.com>
To: "Giles Hogben" <giles@ontv.com>
Cc: <www-p3p-public-comments@w3.org>
Correct, services cannot overide the category
of base data *elements* -- but they can define
new schemas that override the category of *structures*.
user.name.given is an element, and it can't be
overriden. But it inherits the structure personname.given.
You could define a new schema that contains a new
element, say customer.name.given and define it
to be a different category.


----- Original Message -----
From: "Giles Hogben" <giles@ontv.com>
To: <lorrie@research.att.com>
Cc: <www-p3p-public-comments@w3.org>
Sent: Sunday, April 29, 2001 12:58 PM

Thanks for the weekend reply!
But what about this sentence in the Appel Spec?

"Please note again that user agents MUST NOT allow services to override the
category of fixed category base elements, such as for example
"user.name.first" or "user.home.postal.city". "


This is indeed one of the more confusing areas of the spec.
I believe that if you create a new DATA-STRUCT that overrides
the category of the structure it inherits, then you should ignore
any categories in the original structure. Thus, in the vehicle example,
vehicle.built.where.city, vehicle.built.where.postalcode, etc. all
have the category "preference", despite the fact that they are
defined differently in the postal structure in the base data schema.
So, you need to have your APPEL parser recognize that the
category is being overriden in this case, and act accordingly.



I have a question on what appears to be an anomaly in the P3P spec:

According to the example given in the APPEL spec for category only data
element expansion,
it's implied that only structref references with matching category
attributes or no category attributes can match a structref element.
The example shows the onlne category which chooses only the uri elements
from the online category. Any subsequent relational matches via the
structref attribute are then only made to nodes which have  an <online/>
grandchild (i.e. structures also in the online category) and this appears to
make sense.


In the P3P spec, you give the example of a custom schema involving
motorcycles and cars which gives vehicle.built.where a structref of
and a category of PREFERENCE.
But none of the base schema postal structures are in the PREFERENCE
category, which means that a parser running according to the rules implied
in the APPEL spec would ignore any of the nodes with postal references:

Am I confused about something or what is the explanation for this?


Giles Hogben
Received on Sunday, 29 April 2001 16:14:07 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:43:00 UTC