Re: [rtcweb] Unacceptable - (Re: Straw Poll on Video Codec Alternatives)

[Richard Barnes <[e-mail]>]

On Tue, Dec 10, 2013 at 10:29 AM, Dave Crocker <[e-mail]> wrote:

> On 12/10/2013 7:24 AM, Adam Roach wrote:
>
>> On 12/10/13 08:46, Dave Crocker wrote:
>>
>>> As I noted, 2/3 of the choices were positive.  That biases response
>>> patterns.
>>>
>>
>> I don't think we need to go around accusing the chairs of biasing
>> towards success as if it's something bad.
>>
>
>
> Wow.  That misunderstanding of my note is even more impressive than
> thinking that I'm merely expressing an opinion, rather than citing
> methodological fact.
>
> The term 'bias' is a term of art in research methodology.  It pertains to
> procedural problems that produce distorted results.  The fact that an error
> is present in methodology is independent of how or why it was introduced.
>
> In particular, I was not making assertion about the "intent" of the folk
> creating and issuing the survey.  I frankly don't care what their intent
> is, since ad hominem factors are nearly always supposed to be out of scope
> for IETF discussions.


Dave, the salient question here is whether whatever biases are in the straw
poll will affect the outcome in a way that is (1) material and (2)
detrimental.  As Adam notes, the worst case is that we think there's more
support than there actually is.  I'm sure the subsequent consensus call
will control for this.  And I am far from convinced that there is any
likelihood of a material impact from the straw poll's layout.  People here
are smart enough to choose wisely.

Your observations of the empirical facts on methodology-induced bias are
noted.  Your opinion that those facts have some bearing on the process here
is, as far as I can tell, in the rough.

--Richard




>
>
>
> d/
> --
> Dave Crocker
> Brandenburg InternetWorking
> bbiw.net
>
> _______________________________________________
> rtcweb mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/rtcweb
>