Re: scanning the openSUSE members

On 10/02/2015 01:05 PM, Cornelius Schumacher wrote:
> On Friday 02 October 2015 14:23:55 Richard Brown wrote:
>> We've discussed this repeatedly on this list now
> 
> Yes, but it seems there is no consensus on the topic, so maybe we do need to 
> discuss this further. To me this is important because it is about the core 
> values of the project, about our openness, and about the question who we want 
> to be part of the project and whom not.
> 
>> Membership is a key part of this projects governance, and the current
>> situation where membership votes have a terrible turn out is not
>> acceptable
> 
> Why is that not acceptable? The active people vote, the inactive don't.

I think this is a bit too simplistic. By casting a vote for the board
elections the voter has a certain influence on the direction of the
project, and we can argue about this as part of this discussion. Thus,
with a large number of inactive "openSUSE Members", per the member
definition, the problem arises is two fold.

a.) If non of the inactive members vote then those that are active have
the influence they arguably should have, but we do not know if this is
the case and thus the vote (number of voters vs. number of members)
looks non representative due to the low turn out. We can also not
differentiate between the inactive non-voters and those that are active
but choose not to vote.

b.) If everyone votes then one can argue that those that are generally
inactive have an undue influence over the direction of the project by
casting their vote.

These are just the two most obvious, from my perspective, problems that
arise by the skew of "active openSUSE Members" vs. "inactive openSUSE
Members".

The problem does not necessarily need to be solved by culling the
membership list. Other approaches may be feasible.

- Run the "activity script" in November and the top 200 contributors get
invited to vote. This may be one potential option. We can then argue
about how we determine the "top 200", or we can pick another arbitrary
number.

- We can avoid the problem by changing the governance structure,
dismember the board and have the role of the "Board Chairman" be renamed
as "Community Liaison" to the primary sponsor of the project.

- Another option is to move to a "pay for" model. Pay a yearly
membership fee and the payer is in, whether or not the person paying
contributes is immaterial. Those that pay get to vote.

- Have an entirely appointed board. The board composition can still be
determined by the current rules. The board terms can also remain in
place. New board members could be appointed/nominated by the rest of the
board, the sponsoring organization(s), or anyone in the community.
Presumably the nominees would be from a pool of most active people. The
current board members then pick from the pool of nominees.

Anyway as Richard pointed out, a change in governance model needs a vote
by "openSUSE Members" which brings us back to the original problem. Of
course we can choose to change the governance model with the current
voting structure.

> It 
> makes sense to care about the number of active people.

Agreed.

> But what does the 
> number of inactive members change?

It skews everything, see above.

> Nothing, because these are the inactive 
> ones.

From my perspective this is not correct. As I tried to explain above
inactive "openSUSE Members" do have an effect on the project in one way
or another. Thus, I think, it is fair to try to quantify that effect and
possibly reduce or avoid the effect.

> 
> Note that this is a very different kind of governance than elections for a 
> government or similar votes. A government decides for everybody, also for the 
> people who didn't vote. There a strong legitimation by good participation does 
> make a difference.
> 
> But for openSUSE, which is a volunteer project, and those who don't vote 
> because they are not interested anymore are also not affected by the results 
> of the votes,

I would argue they are. They may not care about the effects, but that is
a different discussion. The way the project represents itself,
internally and externally, the way the project conducts presence at
events etc. is representative of all those that are associated with the
project, active or not.

> because they chose to not participate anymore. In such a 
> situation the number of non-voters says nothing else than how much of a 
> history the project had.
> 

With the exception that you cannot differentiate between the people that
are active and do not vote and those that are inactive and do not vote.
Thus it is pretty much impossible to "care about the active" people and
try to understand why the choose not to vote. That is also an important
part about caring for the active people, at least from my perspective.

Later,
Robert


-- 
Robert Schweikert                   MAY THE SOURCE BE WITH YOU
Public Cloud Architect                         LINUX
rjschwei-IBi9RG/b67k@public.gmane.org
IRC: robjo