This week began with a welcome to Zeph Harben of TRUSTe.
The focus of the meeting was the document setting out current thinking on resource grouping
. Will scope statements always be written by machine or will they also be written by hand? This is pertinent to the decision on whether grouping statements should only ever be written in relatively terse Regular Expressions or whether a simpler string-based approach is also desirable. The end result is that we will use both — that is, strings when that's all that is required by still retain support for Regular Expressions.
The discussion then turned to scoping by property. Is this necessary? If so, how can it be done? If you have to fetch the resource to find out if it is a member of a group does that not defeat the object of DR
s? Although the use cases that motivate the group's charter are centred on things like trustmarks and labels for things like accessibility, child protection, eCommerce and mobileOK, POWDER should not be restricted to this and one can fairly quickly envisage a use case where scoping by property becomes useful. For example, the statement "all blue T-shirts in this store are made from bleach-free cotton" implies knowledge of a particular item before the bleach-free cotton claim can be relevant.
Other cases call for different mechanisms to come into play. If a content provider has an established data system that can return data about particular items then that can be used to define membership, or not, of a set. There is a good deal of work to do on this but the resolution was taken that we wish to support scoping by properties by various means
Finally the group noted the Rule Interchange Format Working Group has got various documents ready for review. Pantelis Nasikas will take part in a telecon with the RIF WG in the near future on behalf of POWDER.
There will be no meeting next week. Next meeting is 16th April.