Re: VoiceXML 2.1: Organization of the document

* Matt Oshry wrote:
>The Voice Browser Working Group considered the pros and cons of
>authoring a specification that duplicates all of the functionality in
>the previous version of VoiceXML 2.0 vs. only including the set of new
>features (aka the diff). The VBWG has chosen and continues to choose the
>latter approach for VoiceXML 2.1. The meager size of the diff did not
>warrant the effort involved in duplicating the 2.0 spec. In addition,
>the VBWG desires to make the importance of backwards compatibility clear
>by forcing readers to refer back to the VoiceXML 2.0 specification as
>the source of truth for the majority of functionality that comprises
>both VoiceXML 2.0 and VoiceXML 2.1.

I think it's vital to the success of VoiceXML 2.1 that the community has
access to an up to date and complete specification. Errors in VoiceXML
2.0 have been identified since the publication of that document in March
2004 and the Working Group is required to eventually publish a revised
edition of that specification. With the current organisation people have
to constantly switch between the two specifications and possible the two
errata documents we'll eventually end up to find the definition of a
certain feature, which makes using the specification as a reference, and
authors are more likely to consult other resources which often have some
errors they'd then adopt in their content and implementations. As I
pointed out in the original issue

other W3C Working Groups tried to make similar "diff" specifications,
but had to conclude that complete specifications are of considerable
more value to the community and changed their documents accordingly. If
Merging the two documents requires significant effort, it's better to do
it just once than to require any reader to merge them in their daily use
of the documents. Since I raised the issue in June 2004 the Group also
had more than enough time to do it. Considering that other members of
the VoiceXML community

agree a complete specification would be of considerable more value, I
don't think I can the accept the resolution as it stands. If there are
concerns with respect to backwards-compatibility, a better approach to
make this clear is to discuss this in a dedicated section rather than
the current form of reference.
Björn Höhrmann · ·
Weinh. Str. 22 · Telefon: +49(0)621/4309674 ·
68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · 

Received on Monday, 13 February 2006 13:06:40 UTC