W3C home > Mailing lists > Public > public-speech-api@w3.org > June 2012

RE: The SpeechRecognition.maxNBest attribute

From: Young, Milan <Milan.Young@nuance.com>
Date: Tue, 19 Jun 2012 18:30:36 +0000
To: Hans Wennborg <hwennborg@google.com>
CC: "olli@pettay.fi" <olli@pettay.fi>, "public-speech-api@w3.org" <public-speech-api@w3.org>
Message-ID: <B236B24082A4094A85003E8FFB8DDC3C1A47455C@SOM-EXCH04.nuance.com>
I don't have the web background that you folks possess, so I'm happy to be told I'm wrong.  But my thinking is that they put short in the IDL spec for a reason, and this seems like a perfect use case.  Using short communicates to the developer that this is an array index, not some unbounded quantity like a duration.


-----Original Message-----
From: Hans Wennborg [mailto:hwennborg@google.com] 
Sent: Tuesday, June 19, 2012 10:36 AM
To: Young, Milan
Cc: olli@pettay.fi; public-speech-api@w3.org
Subject: Re: The SpeechRecognition.maxNBest attribute

On Tue, Jun 19, 2012 at 3:55 PM, Young, Milan <Milan.Young@nuance.com> wrote:
> Maybe I'm missing something here, but a short is going to give us 65k possibilities.  That's the maximum size of an array in many languages.
> If a recognition engine can't put the right answer into one of those slots it has bigger problems.

I think Olli's point (please correct me if I'm wrong) was that it's common to use long by default unless there is a specific reason not to.

Received on Tuesday, 19 June 2012 18:31:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:02:27 UTC