Re: Clarify definition of resultIndex

I've updated the spec with this change:
https://dvcs.w3.org/hg/speech-api/rev/633e735be33e

As always, the current draft spec is at:
http://dvcs.w3.org/hg/speech-api/raw-file/tip/speechapi.html

On Fri, Sep 7, 2012 at 8:46 AM, Glen Shires <gshires@google.com> wrote:

> Yes.
>
>
> On Fri, Sep 7, 2012 at 2:00 AM, Satish S <satish@google.com> wrote:
>
>> > this constraint is already inherent in the spec because only a single
>> final result may be returned in continuous mode (5.1.1)
>>
>> I guess you meant to say "only a single final result may be returned in
>> non-continuous mode"?
>>
>> Cheers
>> Satish
>>
>>
>>
>> On Fri, Sep 7, 2012 at 9:26 AM, Glen Shires <gshires@google.com> wrote:
>>
>>> I propose eliminating from the definition of resultIndex the sentence:
>>>     "When continuous was false, the resultIndex must always be 0."
>>>
>>> While this is true when returning the final result, this constraint is
>>> already inherent in the spec because only a single final result may be
>>> returned in continuous mode (5.1.1) and the results list must never contain
>>> interim results before final results (5.1.8).
>>>
>>> However, it is not necessarily true when returning interim results (for
>>> example, when only some of the interim results need to be overwritten).
>>>
>>> If there's no disagreement, I'll remove this sentence from the spec on
>>> Tuesday.
>>>
>>> /Glen Shires
>>>
>>
>>
>