W3C home > Mailing lists > Public > www-voice@w3.org > January to March 2004

Re: clarification vxml 2.0

From: Jeff Kusnitz <jk@us.ibm.com>
Date: Sat, 20 Mar 2004 14:35:06 -0800
To: "janine graves" <j9graves@comcast.net>
Cc: www-voice@w3.org
Message-ID: <OFA955B305.B539603C-ON88256E5D.007B7492-88256E5D.007BFB8C@us.ibm.com>
Janine Graves wrote on 03/20/2004 06:06:58 AM:

> I have been putting together a course to teach VXML 2.0, but have 
> found a few inconsistencies in the spec, and would like these 
> cleared up, if possible. 

> 1) With the builtin grammar type “boolean”, what is returned, is it 
> the Boolean value true/false, or is it the string ‘true’/’false’ ?
> this is very important when comparing the returned result to see if 
> caller confirmed or not. 

It's true or false, not "true" or "false" (as the text says, "The result
is ECMAScript true for affirmative" ...)

> 2) With the <say-as interpret-as=”…”> element, what value should be 
> used for interpret-as?

It will probably depend on the vendor's support of builtins, as they're
optional.  We support "vxml:type" on our platform.

> 3) Furthermore, where are the values for the format attribute of 
> say-as defined?

The values for the format attribute are likely vendor/platform-specific.

Thanks,
Jeff
Received on Saturday, 20 March 2004 17:35:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 October 2006 12:48:59 GMT