Re: Issue-151

Apologies for not being on the call Nick. As I understand the situation and what I understand you to be saying below is that the current text doesn't carry any language requiring implementation of the UGE API.  It tells you how to do it if you want to.

I think Shane wants language that it *MUST* be implemented.  Perhaps that wouldn't be in the TPE, but rather the TCS document for what is required of a compliant user agent, I don't know.

I don't understand how as the wiki asserts the  "API would be implicitly mandatory for implementers of the TPE specification" based on the existing language.
 
If the current draft is in fact silent on the issue and an implementer could therefore opt not to implement the UGE API, I *could* live with that.

However, from what I've understood Shane to say, he believes that the current text does require implementation. I just don't see how that is the case.   If silence can be interpreted this way, then I think we need instead to explicitly say implementation of the UGE AP is optional.

Regards,
John



On Nov 20, 2013, at 3:09 PM, Nicholas Doty <npdoty@w3.org> wrote:

> Hi John and Shane,
> 
> As I tried to describe on today's and last week's calls, the editors' draft text on this topic doesn't have any explicit text about requiring implementation of this section of the specification. Similarly, I don't believe it has any sentences in other sections saying, for example, 'Implementers MUST implement this section.' Implementation of any and all sections is voluntary, of course, as we just develop Recommendations.
> 
> User agent implementers who wish to implement the full TPE specification would, based on the current text and as I understand it, need to instantiate the relevant JavaScript methods and respond to them. User agent implementers won't be implementing all sections -- for example, they won't implement the server-side tracking status resource behavior, since that's server-side. Similarly, non-JavaScript User Agents won't be able to implement a JavaScript API and users who configure their browsers to disable JavaScript won't have those methods available. In the past, participants have noted that we might want to mark this feature "At Risk" when we got to the Call-for-Implementations stage, since there have been some doubts about likelihood of implementation and adoption.
> 
> As requested on the call today, I've set up a wiki page to track relevant proposals:
> http://www.w3.org/wiki/Privacy/TPWG/Proposals_on_UA_requirement_to_handle_exceptions
> 
> I'm now trying to add John's and Shane's proposals (as I understand them) and would welcome any clarifications, corrections or additions.
> 
> Thanks,
> Nick
> 
> On November 20, 2013, at 10:43 AM, Shane M Wiley <wileys@yahoo-inc.com> wrote:
> 
>> John,
>>  
>> I believe Nick captured this best on the call today.  Perhaps he can (please) chime in here.  J
>>  
>> - Shane
>>  
>> From: John Simpson [mailto:john@consumerwatchdog.org] 
>> Sent: Wednesday, November 20, 2013 11:34 AM
>> To: Shane M Wiley
>> Cc: public-tracking@w3.org (public-tracking@w3.org)
>> Subject: Re: Issue-151
>>  
>> Hi Shane,
>>  
>> I'm sorry, but I am confused.  Can you point me to the text to which you are referring.  It seemed to me the version to which Nick referred me, was silent on the issue.  As you know, I strongly believe that implementation should be OPTIONAL,  but I think we both agree that we ought to be talking about the same text.
>>  
>> I'm not sure we are.  Can you point me to the text you are citing?
>>  
>> If you are referring to the current editors' text, I don't see how it does what you want and makes implementation of the UGE API a MUST
>>  
>> Apologies that I could not make the call this morning.
>>  
>> Regards,
>> John
>>  
>>  
>>  
>> On Nov 20, 2013, at 10:02 AM, Shane M Wiley <wileys@yahoo-inc.com> wrote:
>> 
>> 
>> Proposal:  Maintain the existing “MUST” position in the text.
>