Re: New Charter

Standards move forward in different ways.  Just as the W3C Recommendation process differs from the IETF RFCs, so too does the CG work follow a different path from Recommendations.  The most important criteria in my mind is establishing a critical mass of browser companies and application developers.  The details of the process may influence mindshare but are of far lesser import than the ideas that comprise the standard and the timing in the marketplace.

Glen and Satish have each advanced an opinion that the current document is sufficiently advanced to merit implementation and author feedback.  This is an opinion, but certainly an informed one.  Should Google be joined by one or more browser vendors, or perhaps even if not, there is an opportunity for rapid progress to be made.  I fully endorse their efforts to field test the Javascript Speech API and will offer my assistance where appropriate.  The lessons from these efforts will undoubtedly inform future work and further the prospects of a fully speech-enabled web.

That stated, I see no reason that interested parties should not proceed with plans for a W3C Working Group.  While Satish has indicated that Google will not participate at this time, my reading of Milan's, Raj's, and Jim's emails suggests that there are at least three companies interested in following that path.  There may even be an opportunity to re-engage with existing working groups such as the Multimodal Interaction Activity [1].  No doubt the members of a W3C Working Group focused on speech APIs will want to track efforts at Google, AT&T, Apple, and elsewhere as they develop a W3C Recommendation for a speech API.  I am certain that the initiative would also challenge and improve existing W3C Recommendation track documents such as EMMA and the recently published Multimodal Architecture and Interfaces Recommendation.  Surely this is a good thing.

I see no deep conflict here.  Different opinions and complementary efforts, sure, but a a richer set of concepts and a better eventual specification as a result. 

-=- Jerry

[1] https://www.w3.org/2002/mmi/Group




On Sep 18, 2012, at 8:04 PM, Matt Womer wrote:

> Just to be clear what we have is not what W3C considers a spec. CG work does not directly result in W3C Recommendations and as such are not covered under the IPR agreement covering Recommendations. I find it troubling that at this stage we, having exhausted other WG options and not creating a new WG, are going to let this "spec" exist as is without finishing the process. This seems in conflict with what I thought this group wanted.
> 
> -Matt
> 
> On Sep 18, 2012, at 7:38 PM, Glen Shires <gshires@google.com> wrote:
> 
>> Milan,
>> The charter of this CG is stated on the home page for this CG, and has not changed since the beginning of this CG. http://www.w3.org/community/speech-api/
>> 
>> The goal and scope of this Community Group is to produce a JavaScript Speech API that supports the majority of use-cases in the the Speech Incubator Group's Final Report [1], but is a simplified subset API, such as this proposal [2]. For this initial specification, we believe that a simplified subset API will accelerate implementation, interoperability testing, standardization and ultimately developer adoption. This JavaScript Speech API will enable web developers to incorporate scripts into their web pages that can generate text-to-speech output and can use speech recognition as an input for forms, continuous dictation and control. Specification of HTML markup and a network speech protocol are out-of-scope of this Community Group.
>> 
>> [1] http://www.w3.org/2005/Incubator/htmlspeech/XGR-htmlspeech/
>> [2] http://lists.w3.org/Archives/Public/public-webapps/2011OctDec/att-1696/speechapi.html
>> 
>> 
>> This is indeed a change of my opinion of what we should be doing at this point. However, if statements that I made in that prior email response were interpreted by you as re-defining the group's charter, then I apologize for the confusion. The only reference to the word "charter" that I see in that entire thread is in reference to drafting a new charter for the potential new WG.
>> 
>> With regard to the three points you made below...
>> 
>> 1. The in-progress discussions have been summarized as editor notes in the current spec: http://dvcs.w3.org/hg/speech-api/raw-file/tip/speechapi.html
>> We believe we have wrapped-up the work to the extent that compatible implementations by browser vendors are possible.
>> 
>> 2. Test Suites are nice to have, but aren't blocking us from moving forward.
>> 
>> 3. As stated previously: Google will not join a new WG at this time. We believe this is the best course of action right now for the spec for the reasons stated earlier in this thread.
>> 
>> 
>> On Tue, Sep 18, 2012 at 4:18 PM, Satish S <satish@google.com> wrote:
>> Perhaps I misunderstood, but I thought the charter was defined when the CG was formed and is the one mentioned in http://www.w3.org/community/speech-api/. That doesn't mention anything about transitioning to a WG so it is clearly not part of the charter. 
>> 
>> We did discuss about it in the CG earlier and Google's current position is after careful consideration of the great progress made in the CG the timing is right to focus on implementation and iterate based on web developer feedback. I do see us continuing to improve the spec based on  feedback we receive once there are UA implementations of the proposed API.
>> 
>> Our position is not inconsistent with the charter you proposed, but we wanted to inform in context of the new charter discussion that we won't be joining a new WG at this time.
>> 
>> Cheers
>> Satish
>> 
>> 
>> 
>> On Tue, Sep 18, 2012 at 11:52 PM, Young, Milan <Milan.Young@nuance.com> wrote:
>> Glen, I agree with what you have written below, but fail to see how this is inconsistent with the new charter.  If there is an inconsistency, please let me know so I can update the proposal.
>> 
>>  
>> 
>> I’m also still awaiting your response to each of the three points I made below.  Being that your hold the position of chair, and given the context of the statements, the promises you made were essentially interpreted as being part of the groups charter.  (In fact I believe you are aware that many of us have remained active in this group precisely because of those promises.)  It would be a violation of the CG mission for this group to proceed otherwise without explicit consent from the members.
>> 
>>  
>> 
>> Thank you
>> 
>>  
>> 
>>  
>> 
>> From: Glen Shires [mailto:gshires@google.com] 
>> Sent: Tuesday, September 18, 2012 3:33 PM
>> To: Young, Milan
>> Cc: Satish S; public-speech-api@w3.org; Ian Jacobs <ij@w3.org> (ij@w3.org)
>> Subject: Re: New Charter
>> 
>>  
>> 
>> We at Google believe the spec is ready to be implemented by browser vendors in a compatible way. The few remaining open issues are not blocking issues, and getting early feedback from web authors will provide great insight in charting the course for future work on the spec. We believe that it's most important right now for browser vendors to focus on implementation, rather than attempting to resolve the few remaining issues without feedback from web developers building real-world JavaScript applications using it.
>> 
>>  
>> 
>> /Glen Shires
>> 
>>  
>> 
>> On Tue, Sep 18, 2012 at 2:27 PM, Young, Milan <Milan.Young@nuance.com> wrote:
>> 
>>  
>> 
>> From: Satish S [mailto:satish@google.com]
>> 
>>  
>> 
>> We at Google are pleased with the current progress in this CG of the Speech JavaScript API Specification, and believe that it is nearing completion, at least an initial version that can be implemented compatibly across multiple browsers.  We believe it provides a rich toolset for web authors, and that getting early feedback from web authors is valuable. As such, we believe the major task ahead is to complete implementations and test suites.
>> 
>> [Milan] We’ve had a similar discussion on this topic before, and that thread was brought to a close with a statement from our chair Glen [1].  He made three points in that mail that I’d like to review:
>> 
>> 1)      We were planning to wrap up work before TPAC.  This statement was in turn followed up by another email from Glen in which he stated that he and Hans would work to summarize in-progress discussions as editor notes [2].  That second statement was made in the context of prioritizing wrapping up our work so that we would have time to transition into a WG within the agreed TPAC timeframe.  This is in contrast to the model where we would push out the date in hopes of achieving stability.
>> 
>> 2)      Test suites are nice to have, but they would not blocking formation of a WG.
>> 
>> 3)      Once the Speech API is adopted by a WG, we can conclude this CG.  I’d like to point out that the Speech JavaScript API is an explicit part of the new WG charter.
>> 
>>  
>> 
>> In short, your statement constitutes a major reversal of the premise of this CG.  Please clarify.
>> 
>> 
>> 
>> We wish to continue as a CG to rapidly complete this work, rather than take on the overhead of forming a new WG. We will not join a new WG at this time.
>> 
>> [Milan] Yes, there is a bit of overhead to forming a WG, but most of that has already been done for you.  As far as operating within a WG, I see significant gains to what we can accomplish, and we’ll have a standards-track document to show in the end.
>> 
>> Thanks
>> 
>> [1] http://lists.w3.org/Archives/Public/public-speech-api/2012Jun/0139.html
>> 
>> [2] http://lists.w3.org/Archives/Public/public-speech-api/2012Aug/0026.html
>> 
>>  
>> 
>> 
>> 
>> Cheers
>> Satish
>> 
>> On Tue, Sep 11, 2012 at 12:55 AM, Young, Milan <Milan.Young@nuance.com> wrote:
>> 
>> I suggest we use the attached as the first draft for our new charter.  After gathering feedback from this group, I am planning to push the document through the usual channels.
>> 
>>  
>> 
>> Note that I’ve also requested space at the upcoming TPAC, which I’m hoping will serve as our first meeting (informal or otherwise).  We’re on the waiting list for a room to free up (made the line August 7th).  I believe Matt Wormer is trying something similar, so please respond to his post [1] if you can make it.
>> 
>>  
>> 
>> Thanks
>> 
>>  
>> 
>> [1] http://lists.w3.org/Archives/Public/public-speech-api/2012Sep/0021.html
>> 
>>  
>> 
>>  
>> 
>>  
>> 
>> 
>>