[CC] Minutes of 1st Phone Conf. on Consent Choices 2003/05/15 (Version 1)

Hi Folks,

here are the minutes of today's call. Please send any 
corrections/questions/suggestions to me.

Regards,
  matthias

------8<-------

Participants:
  Lorrie Cranor
  Matthias Schunter

Agenda (see earlier mail):
1. Goals/usecases
2. Syntax

SUMMARY
- we agreed that grouping consent is useful (but unclear whether for P3P 
1.1 or P3P 2.0)
- MTS will draft a syntax that we will float to the implementors to check 
whether they like it

GOALS
- the participant agreed that some bundling of consent is useful in practice
- we were not sure whether we should aim at P3P 1.1 or P3P 2.0
- a feature that is not implemented by any agent is not very useful
- as a consequence, we decided that we will elaborate a draft proposal and 
float this
   to the implementors. If agent implementors like the feature, we will aim 
for 1.1 otherwise 2.0

Syntax
- the initial proposal from MTS was to add a 'consent-group' ID as an 
attribute inside the elements
   that are opt-in or opt-out. The reason was to identify a consent group 
for each opt-in/opt-out
- the initial proposal was not backward compatible
- in the long run, the consent should be to complete statements
- as a consequence, we agreed that the consent group will be attached as an 
extension to <statements/>
- If a statement is in a consent-group, all its elements must either be 
opt-in or all must be opt-out
   * in order to be backward compatible with the existing semantics of P3P 1.0
   * In order to avoid complicated and confusing semantics (e.g., what a 
consent means if opt-in, opt-out,
     and required things are mixed).
- MTS will draft a proposal. Once this is agreed within the working group, 
we will distribute it

Received on Thursday, 15 May 2003 11:58:33 UTC