Re: Suggested response to the Yandex "cannot iive with loosening of TAG participation requiremens"

On 2015-04-13 08:36, David Singer wrote:
>> On Apr 13, 2015, at 14:29 , Daniel Glazman <daniel.glazman@disruptive-innovations.com> wrote:
>>
>> On 13/04/15 00:43, Stephen Zilles wrote:
>>
>>> The first version (that which is now in the Draft Process 2015 document)
>>>
>>> “A Member organization is permitted at most one participant on the TAG
>>> except when having more than one participant is caused by a change of
>>> affiliation of an existing participant. At the completion of the next
>>> regularly scheduled election for the TAG, the Member organization must
>>> have returned to having at most one participant.”
>> I also have concerns about this, but for reasons I think are entirely
>> different from the one behind Chaals' objection.
>>
>> 1. the last sentence is unclear. I could understand the intent if it
>>    meant "the next election for the extra seat under consideration" but
>>    if that's really any election, it means we're back to the previous
>>    situation where a TAG member has to resign at some point to comply
>>    with rules we found counter-productive and over-rigid enough to
>>    update them. In short, this does not solve the issue.
> I don’t understand why this doesn’t solve the issue. It doesn’t matter which seats are considered.
>
> What we wanted was that if someone changed allegiance, we (a) did not force a special election but (b) any >1 seat count for their employer would last no longer than until the next election. How the company resolves this — resignation, who resigns, when, term endings, etc. — is up to them.

In the current process, there doesn't have to be an election if someone 
resigns: "When an elected seat on either the AB or TAG is vacated, the 
seat is filled at the next regularly scheduled election for the group 
unless the group Chair requests that W3C hold an election before then".

I think the difference in the proposed process change is whether someone 
has to resign immediately or can wait until the next election.   I don't 
see supporting that change as agreeing the old rule was inappropriate.  
I thought it was just reaching out to some people who were unhappy with 
the old rule and offering a compromise.

The restriction to only one person from the same employer could avoid 
unintended bias.  No matter how much someone cares about the good of the 
Web and acting as an individual (not for their employer), it wouldn't be 
surprising if people who worked on some particular product at some 
particular company care about that or at the least are exposed to ideas 
and issues that are widespread at their employer. So, that could limit 
the diversity of views.

Beyond that, there is a small set of large companies with a very large 
number of participants across the many Working Groups in W3C. In an 
election of the 400 members, name recognition value of being from one of 
those companies may be fairly high.  It may be harder for someone from a 
company less well known to get elected if the entire group could be 
composed of employees from the top few companies in W3C 
participation.    Those companies do have people with years of valuable 
experience and obviously make major contributions to W3C, but given the 
TAG and AB are so small, allowing multiple people from the same company 
could limit representation from the wider membership.



>
> Say, for example, in year X, A from the company is elected. In X+1 B from another company is elected, and then changes allegiance to the same company as A. Since A’s term expires at the next election, the company will, with no special thought, return to having one member — B.
>
> Do we need the obvious stated, that you cannot stand for election if, if you are elected, the rule would be violated?
>
>> 2. this does not change the center of gravity of the TAG. The TAG is
>>    a group of individuals that we all value for their expertise, not
>>    a group of employees of their employers. As a matter of fact, they
>>    remain on the TAG if they become unaffiliated…
> That’s the theory, and I encourage you to read the previous comments to understand the concerns about this in practice.
>
> (Since the previous discussion, New Scientist has published an edition that focuses on ‘belief’. In it they report on research in which volunteers were given moderately difficult tests. Without any influence, they get X% right. When the group answering is ‘seeded’ with volunteers deliberately answering wrongly, the percentage of correct answers plummets to around X/3%. Ah, they went with the crowd, you say. No, they honestly believed their answer, and that they had not been influenced by others.)
>
>> Sorry if my comments above arrive late, I was burried under work and
>> then sick.
> I am sure thoughtful comments are helpful at any time!  Glad you are feeling better enough to comment on W3C processes…
>
> David Singer
> Manager, Software Standards, Apple Inc.
>