Re: Fwd: NomCom Volunteer List

[Michael StJohns <mstjohns@mindspring.com>]

For your edification - the probability tables are

1       0.024713
2       0.091699
3       0.193938
4       0.258857
5       0.227795
6       0.133814
7       0.051799
8       0.012641
9       0.001756
10      0.000105


In other words, the probability of having exactly 4 members of the Nomcom 
from the 48 cisco/nortel/nokia folks is about 25.9%.  The chance of having 
at least 5 members is about 42%....

Later, Mike


At 18:51 10/6/2003, Michael StJohns wrote:

> > From: Richard Draves <richdr@microsoft.com>
> > To: IETF-Announce: ;
> > Subject: NomCom Volunteer List
> >
> > At this time, 2727bis has not yet been approved as an RFC. Hence I will
> > use RFC 2727's eligibility criteria (attended 2 of the last 3 IETF
> > meetings) when performing the random selection.
>
>
>
> >  I will not check the
> > primary affiliation of the volunteers.
>
> Umm... BULLSHIT!   On this list of 113 volunteers, 16 come from Cisco, 19 
> from nokia and 13 from Nortel.    48 in total or about 42% of the total 
> list.    That means it wouldn't be unusual for 6 of the 10 to come from 
> this group.  And it wouldn't be all that unusual for 3 or 4 to come from 
> Nokia or Cisco. (Statistically speaking)
>
> This was supposed to be fixed and it was the IESGs job to get this 
> document through the process prior to this beginning....
>
> At that, what's in the document still isn't a fix - we could still end up 
> with 6 or more from this group...  could we at least keep them from all 
> being from one organization - please.
>
> Mike
>
>
>
>
>
> > If the RFC 2777 algorithm selects
> > an ineligible volunteer, I will repeat the method until I have ten
> > eligible and willing volunteers.