RE: http://www.w3.org/TR/speech-synthesis

Dave,

I'm the chairman of the W3C Voice Browser Working Group.  

We have held several discussions about scripting langauges, including the
strengths and weaknesses of XSLT, both in the full Voice Browser Working
Group, and in the subgroup dealing with this specific issue.  After several
weeks of discussion, we needed to make a decision so our work could move
forward.  

We used a prepublished methodology for determining which scripting facility
to use with our grammar specifications.  First, every working group member
had a chance to state their position.  This gave everyone a change to
express their opinion, and to lobby for their perferred solution.  Next,
every company then voted for their perferred solution.  

As I recall, the basic reasons Workin Group members did not choose XSLT
included members feeling uncomfortable with the new XSLT, a large ramp-up
effort to use XSLT, and additional implmentation effort for XSLT when
implmenetations of other scripting languages already exist. 

We eliminated XSL from contention because only 4 of the 22 companies voted
for it.  With this little support, XSLT had little chance of of gaining
concensus from the remaining 18 companies, especially after several weeks of
discussion. 

Regards,

-Jim Larson

-----Original Message-----
From: Pawson, David [mailto:DPawson@rnib.org.uk]
Sent: Thursday, August 31, 2000 11:40 PM
To: 'www-voice@w3.org'
Subject: RE: http://www.w3.org/TR/speech-synthesis


I'd call that burying ones head in the sand,
I'm rather curious why the group are so cautious?

DaveP

> 
> Microsoft has advocated strongly to include XSL technologies 
> into the spec
> during the discussion. Unfortunately, the majority members in 
> the W3C voice
> browser working group uncharacteristically express fear, 
> uncertainties and
> doubts against W3C's own XSL standard, and have consistently voted out
> anything related to XSL, first in synthesis and this week, 
> recognition!
> 
> Microsoft also demonstrated a more powerful dialog paradigm 
> can be easily
> implemented in XSL and presented it to the working group in 
> May. Despite the
> supports from the dialog communities, this working group 
> refused to even
> consider the idea. As a result, the use of XSL in spoken language
> applications is likely to remain a Microsoft proprietary extension. We
> regret this working group's decisions to defeat W3C's own standard.
> 
> -----Original Message-----
> From: Pawson, David [mailto:DPawson@rnib.org.uk]
> Sent: Thursday, August 31, 2000 12:34 AM
> To: 'www-voice@w3.org'
> Subject: RE: http://www.w3.org/TR/speech-synthesis
> 
> 
> 
> 
> > -----Original Message-----
> > From: Kynn Bartlett [mailto:kynn-hwg@idyllmtn.com]
> > Sent: 30 August 2000 17:25
> > To: Pawson, David; 'www-voice@w3.org'
> > Subject: Re: http://www.w3.org/TR/speech-synthesis
> > 
> > 
> > At 10:02 AM +0100 8/30/00, Pawson, David wrote:
> > >There would appear to be no mention of the similar functionality
> > >available in XSL, particularly
> > >http://www.w3.org/TR/xsl/slice7.html#common-aural-properties
> > >I believe that this set of properties follow broadly the CSS-2
> > >property set, yet no mention is made of them.
> > 
> > How stable is the XSL-FO specification at this stage?  I haven't
> > been following closely enough to know if it's reached Rec yet.
> 
> No, hopefully within the year it should move to candidate,
> but rumour has it that its pretty stable.
> 
> DaveP 
>

Received on Monday, 11 September 2000 19:31:57 UTC