As a result of the call to discuss the DA board elections resulting from this thread: http://groups.drupal.org/node/199178, we agreed to post a couple of specific follow up threads on topics that needed further resolution, so I'm posting on behalf of those who were on the call (FTR the call's etherpad is here: http://lullapad.com/5TCVHgaCYV):
Voting eligibility
Question: Who gets to vote [DA Members? Drupal.org users? Some other parameters?]
Answer: Basically, narrowed it down to two choices for this year's election, and these are generally for practical reasons:
1) Drupal Association Individual Membership
2) Drupal.org user account
Arguments:
- This does not remotely represent the ENTIRE community, but are two reasonable ways we can "vet" voters
- There are many other factors that are important in defining "Community", but the consensus was we shouldn't attempt to do this right now. Instead, we should be clear that we are defining community in a particular context, for this election, and that this definition could shift over time.
- Look at this as a definition of voter eligibility rather than who is a community member, at least for this initial year for the "at-large" positions.
- Drupal.org user account is more open and inclusive, but a less controlled set of users.
- DA membership is a more controlled set of users (hence less risk for vote tampering), but is more exclusive and also may leave out people who can't afford a membership. Possible in future to set option of a sliding scale cost for membership, or of free memberships to certain subsets of the community.
Suggestion to settle this by a poll or similar on d.o. (ie. this discussion thread). So...please discuss!
Comments
I think the impact of the
I think the impact of the cost issue is smaller than folks think. It's not like a $30 fee to vote, it's a $30 donation to an organization that funds a lot of great work.
I've read zzolo's detailed rundown on relative salaries around the world, but I think they overgeneralize from the statistics.
I will cover a portion of membership fees for people who request it via my d.o contact page and state:
I'll pay up to $150, so this would cover at least 5 people and maybe more.
certifiedtorock.com/u/36762/
I think that's a really
I think that's a really awesome gesture, and I hope people take you up on it. But I can definitely see that some people might feeling uncomfortable asking for "charity", so that doesn't really addresses the representation concern.
I think if we want to limit voting to DA individual members only, it should be for one of the solid reasons outlined in the previous thread, which were:
... but definitely not because $30 doesn't have a large impact for at least some people, especially people outside of North America, who we're hoping to engage.
However, there were ideas with some support such as http://groups.drupal.org/node/199178#comment-658003 about providing a "sliding scale" fee for those for whom $30 is a lot of money. I have doubts about whether that's implementable within the election timeframe, however. :\ (Anyone have experience with this + Ubercart?)
@greggles $30 can be a big
@greggles
$30 can be a big chunk of a monthly salary in some countries. In the Philippines its not uncommon to hire someone full time, outsourcing for example, for $300/month and they feed their whole family on that many times.
I made a quick analysis of the countries of the DA members, plus a few suggestions about how the fee could be different between high, mid and low income countries in http://groups.drupal.org/node/199178#comment-658593.
That comment also have suggestion on sponsoring memberships and a few other things.
--
/thomas
www.drupalmill.com
Some metrics that I pulled up
Some metrics that I pulled up tonight from Donna's questions on IRC, which might or might be useful when evaluating the question of whether Drupal.org accounts should be used as a means of vetting voters.
TL;DR: 742,323 (68%) of all Drupal.org user accounts have logged in at least once and are not banned (spambots). Of those, 278,805 (37.5%) have logged in in the past year, and 218,775 (78.5%) of those are new accounts from the past year, vs. 60,030 (27.5%) accounts from 2010 or earlier.
Number of all user accounts in the Drupal.org database: 1,089,986
SELECT COUNT(u.uid)FROM users u;
Number of total "valid" (non-blocked, logged in at some point) users in Drupal.org database: 742,323
SELECT COUNT(u.uid)FROM users u
WHERE u.status = 1
AND u.login > u.created;
Number of valid users who logged in since January 1, 2011: 278,805
SELECT COUNT(u.uid)FROM users u
WHERE u.login > 1293840000
AND u.status = 1;
Number of valid users who are new since January 1, 2011: 218,775 (that means 60,030 accounts that were active in 2011 were from before 2011)
SELECT COUNT(u.uid)FROM users u
WHERE u.login > 1293840000
AND u.created > 1293840000
AND status = 1;
Unfortunately, Drupal doesn't log (to my knowledge anyway) how many times/when a person has logged in, just the last login timestamp.
Oh, and for comparison
Oh, and for comparison purposes, there are currently 1,693 active Drupal Association individual members, which you can see for yourself at https://association.drupal.org/membership/individuals (I copy/pasted into vi and turned line numbers on).
This total represents 0.006% of all Drupal.org accounts active since 2011. And while I can understand that creating a visceral reaction to the level of unfairness this would entail, bear in mind that there are some arguments for it.
The figures, however, as well as the concerns about international accessibility, definitely push me more into wanting to figure out a way to define voting rights beyond individual members. However, I don't have any real bright ideas atm about what that means, other than really "dumb," but inclusive, metrics like "did they log in to drupal.org recently"?
This total represents 0.006%
Think that is a little low, should be 0.6%. Still small, but at least 100 times bigger :)
--
/thomas
www.drupalmill.com
Ah, I was doing 1,693
Ah, I was doing 1,693 (members that have renewed in the past year) / 278,805 (non-spam bot drupal.org accounts who have logged in in the past year) == 0.0060723447
In any event. Tiny. :)
Thanks for these data!
Angie, thanks for this! Some solid data to work with really helps when trying to grapple with these questions.
Voting eligibility
First answer that comes to mind would be to let only members of DA vote on elections of DA board, which sounds logical.
And it would be the case if the board in its future work make decisions only affecting members of DA.
If it was an organization which has a list of members, and this members vote to choose amongst themselves a board of directors which will lead this organization and make decisions to the benefit of members of this organization.
DA however is a bit different. As decisions made by DA board affect much bigger community than DA itself, I think that this bigger community should have a representative on the board and should have an ability to choose representative whom they want to see. How exactly this "bigger community" will be defined it's another question. Considering tight deadlines for this elections suggestion to count eligible for vote active drupal.org accounts seems fine. Active could mean something like:
- d.o account created not less than 2 month ago
- last login date is not more then 2 month ago
Focusing first on criteria, second on individual options
With this decision, and with all the others in the elections, I'd like us to focus first on the question of criteria.
According to the Drupal Association bylaws, the Drupal community should elect two at large directors. So our challenge - and, yes, it's a big one - is to answer: what do we mean by the Drupal community?
I hear a broad consensus that the Drupal community is very large and diverse, ranging from developers and documenters to users, with everyone in between, comprising - at least - hundreds of thousands of people around the world.
What approach should we take in trying to parse "community"? I want to try to tease out here a few different ideas that I'm hearing at least implied in discussion so far.
Broad and inclusive
The first is that we should be as broad and inclusive as possible. We're talking about the community, and no one should be left out. This approach treats these elections like, say, civic elections. Everyone in the city should get a vote.
Key aims here are: no one should be left out. If you're part of the community, you should get your vote.
Cross section/representative sample
In this approach, we don't aim to include everyone, but instead come up with some indicators that help get a cross section of the community. Yeah, tonnes of people are left out, but that's fine--the point is not to include everyone, but to get a good sample.
Key aims here are: capture a cross section that is representative or reflective of the community as a whole. For example, it shouldn't be biased towards capturing developers more than trainers or people in North America more than those in Europe or Africa.
Organizational
In this approach, the Drupal Association is viewed as an organization. Yes, the Drupal community is much bigger than the DA, but in electing DA directors it's those with a direct organizational tie who should participate.
Key aims here are: ensure that those associated with the Drupal Association participate in decision making.
There are lots of other options that could be taken, like "Leaders" (we should have identified leaders in the community play do the selecting) or "Delegates" (at large directors should be selected by a group of delegates from Drupal groups and entities) but I haven't yet heard these raised or implied.
Before we can compare options, we need to clarify which of these very different approaches to parsing "community" we have in mind.
So what is it ? ;)
Given:A DA membership
Given:
...I would prefer this to be open to as wide a swath of people as possible. Therefore, I'd recommend "drupal.org account holder, logged in in past year, registered before nominations open" as the bar, which would put our constituency somewhere in the neighbourhood of ~280,000.
While this is certainly not nearly as big as the entire Drupal community, I can't envision a better system for alleviating voter fraud, and preventing people completely disconnected from Drupal contributors (without whom Drupal simply wouldn't exist) to have say over how money is spent that highly impacts said contributors.
So, for the purposes of the DA elections, I believe we should go with "Cross section/representative sample," but be as "Broad and inclusive" as possible with it.
It's 0.6%, right? (you have
It's 0.6%, right? (you have to multiply the decimal value by 100 to get the percent) but yes, very small.
I don't see why this would be hard to change in either direction: from d.o members to only DA members or vice versa or some other criteria. Each year is a new one and we can learn from the first.
I don't like this proposal as much as DA members, but think the criteria you've set could work and wouldn't raise an argument to try to block them.
An interesting statistic will be: what percent of people eligible to vote actually do :)
certifiedtorock.com/u/36762/
Oh hai I fail at math. :P
Oh hai I fail at math. :P
Adopted recommendation: valid, active drupal.org accounts
As of the meeting today, we updated the Draft community election process to make the recommendation "Voting will be open to all drupal.org account holders who have logged in during the past year and whose accounts were registered before nominations opened (February 8, 2012)." There didn't seem to be much contention around this suggestion, while there were pretty sizable concerns around the option to use DA memberships.