RE: [arch] confidental flag

Hi Dirk,
The MMI Working Group has discussed your questions. We agree that it would be good to clarify that the interpretation of "confidential" is application-specific.  
However, there are two possible eventualities.
1. If we do get two implementations of the "confidential" attribute, the feature will be included in the spec, and the next version of the spec (the Proposed Recommendation) will include clarifying language that emphasizes that the meaning of "confidential " is application-specific.
2. If we don't get two implementations of the "confidential" attribute,  the feature will be removed from the spec altogether, and consequently we won't need any clarifications.
Thanks again for your comments, and also thanks for your quick response to our proposed resolutions. As with your earlier questions, please let us know before June 15 whether or not you accept this proposal.

Best regards,
Debbie Dahl

> -----Original Message-----
> From: Dr. Dirk Schnelle-Walka [mailto:dirk.schnelle@jvoicexml.org]
> Sent: Monday, June 11, 2012 4:14 AM
> To: Deborah Dahl
> Cc: www-multimodal@w3.org
> Subject: Re: [arch] confidental flag
> 
> Hey Debbie,
> 
> I am not sure if this really removes my problems with the interpretation
> of the specification. Although it is considered to be at risk of
> removal, the specification should be clear about the intention. Maybe,
> you should add that the implementation is application specific. I really
> miss that in the specification.
> Also, I am not sure if all those possible behaviors (I mentioned some
> possible interpretations in my first email) can be toggled by a simple
> switch. Especially, if you want to support more than one of those
> behaviors at the same time.
> 
> >From that point of view this solution is not acceptable, sorry.
> 
> Maybe, adding that hint that the interpretation of the confidential is
> application specific and further hints can be submitted in the data
> field could help to overcome this weakness. As a consequence the hint
> about logging should be removed since this is only one possible
> interpretation. Maybe, it could still serve as an example?
> 
> Best,
> Dirk
> 
> 
> > Hi Dirk,
> > We've discussed your question about the confidential flag in the Working
> > Group. Our intention was that the exact interpretation of the confidential
> > flag would be implementation-specific. However, it's also worth pointing
> out
> > that this feature was called out as "at risk" in the Candidate
> > Recommendation spec, due to a potential lack of implementations. If we
> don't
> > get two implementations of this feature it will be removed from the spec.
> > Since we are tracking all comment threads at this point, please let us know
> > by June 15 if this resolution is acceptable, otherwise we will assume that
> > you agree with our decision (but it is very helpful to have an explicit
> > response).
> > Best regards,
> > Debbie Dahl, MMI WG Chair
> >
> >
> 
>

Received on Monday, 11 June 2012 16:24:31 UTC