User:Bluerasberry/2019 asbs commentary

From Wikipedia, the free encyclopedia
The Signpost



2019 asbs commentary

2019 Wikimedia Affiliate Selected Board Seats Election Debrief and Commentary

The last issue of The Signpost reported the "2019 Wikimedia Affiliate Selected Board Seats Election Results". The election facilitators have also published a debriefing which lists individual decision points made in organizing the election. You are invited to discuss these and any other issues at the election talk page. The following commentary gives my interpretation on that debriefing. The better parts may come from the other election facilitators. Feel free to direct comments on the other parts to this author or in the comments section below.

In addition to the debriefing, all the ballots, the notes of election facilitator meetings, and the code of the algorithm which tabulated voting results are all available. Wiki elections are as transparent as possible and we strive to identify whatever is private.

Exclude non-community election governance[edit]

Support independent elections! A comparison to the 1934 Italian general election is overly dramatic but the scary image of this compromised election office triggers the public desire to remove all barriers to independent elections.

The Affiliate-selected board election seats the chosen candidates of Wikimedia affiliate organizations on the Wikimedia Foundation board of Trustees. The intent is of the election is to support the Wikimedia Foundation in fulfilling its mission by electing trustees whom the community and its affiliate organizations have selected as aligned with Wikimedia community values and culture. For the election to work as intended, the Wikimedia affiliate organizations and the Wikimedia community need to maintain control of the election without external disruption.

For this 2019 election the staff of the Wikimedia Foundation operated the poll, claimed rank and power above Wikimedia community facilitators, enforced private conversation about matters of public interest, and made demands of time and labor for volunteer election coordinators that they do things against their consent. Let there be no misunderstanding - the way that the Wikimedia Foundation operated the polls was inappropriate and incompatible with any reasonable understanding of what an independent election is. Please, no one blame anyone, but the Wikimedia Foundation is asserting power and control over this election which is incompatible with community governance. A public community conversation is required about what happened in this election and how future elections will proceed.

An election poll is supposed to collect the ballots and turn them over to the election facilitators. This is a global concept and everyone understands this. Some things which polling operators do not do include looking at the ballots, calculating a result, identifying potential problems with ballots, suggesting to throw out any ballots, or seeking to persuade or compel the election facilitators to call the election in a certain way. The Wikimedia Foundation in official capacity did all these things in this election. All of this is taboo, insane, completely bonkers, and inexplicable. Let's put all this behind us, and for future elections, use an independent polling station and operator which will present ballots to the Election Facilitators without pre-interpretation and advocacy.

Perhaps it is better to not talk about this considering the history of unconstructive tension between the Wikimedia community and the Wikimedia Foundation. If anyone wants to take action now, assert the necessity of operating the polls independently of Wikimedia Foundation staff involvement and confirm Wikimedia Foundation support for this practice.

Support newer and smaller organizations in voting[edit]

142 Wikimedia movement affiliates were eligible to vote in this election and 120 did.
Eligibility requires an organization to be registered with the Affiliations Committee and up to date with reporting.

Some organizations cast ballots which seemed unusual as described in the Election Faciliator notes and as can be seen in the published ballots. For example, what was the intent of a ballot which votes for three people, and ranks them as choices "#1 - person A, #2 - person B, and #11 - person C"? By the preset rules of this election, the facilitators interpreted this ballot where person C was their third preference over all candidates not named. Maybe this was the intent, but perhaps it was not. Perhaps there was a misunderstanding of the process. Why not just vote 1, 2, 3, rather than 1, 2, 11?

Some of the unusual ballots came from Azerbaijan, Art+Feminism, Iraq, Punjab, Uganda, Perú, Nepal, Nigeria, and Sri Lanka. All ballots in this election are public. The ballots of these organizations were an essential part of determining this election where every individual vote set the outcome. Anyone, anywhere in the world can make friends globally in the Wikimedia community by identifying ways for everybody to take advantage of the shared Wikimedia infrastructure, and thus to take advantage of the institutional benefits we collectively offer. Voting is a way for small wiki organizations to have outsized influence in the direction of Wikimedia governance and should be relatively easy in comparison to other high-impact activities. Hopefully, voting inspires awareness of our international connections, and is part of other constructive collaboration such as developing wiki content together and encouraging more diverse communities to apply for grants or whatever they need to develop. An affiliate organization with the right to vote is one that has recently demonstrated interest to do more, and those organizations which did not vote likely would benefit from offers of friendly collaboration. Reach out whenever possible!

The organizations which were eligible to vote but did not are Basque, Commons Photographers, Egypt, Hindi language, Jenga (Swahili language), Wiktionary, West Bengal, Wikiesfera, Wikimaps, Belarus, Ghana, China, Tool Developers, Srpska, Bolivia, WikiWomen, Grŵp Defnyddwyr Cymuned Wicimedia Cymru, and the India technical group. There were no clear reasons given why those affiliates chose not to vote.

All Wikimedia chapters voted. These user groups which either did not vote or cast unusual ballots were all new to the election. The election facilitators recommend that the Wikimedia community organize to provide sufficient support for each of these organizations to exercise their vote in a thoughtful way. That support probably includes special outreach for the election, and general on-going support for small wiki organizations.

Discuss the calculation system[edit]

The calculation of Droop quota for election results is not easy to understand and still has room for subjective decisions. Both technical and non-technical discussion would increase clarity!

This election used single transferable vote (STV) with Droop quota. There is a history in the wiki community of using progressive voting systems to more completely represent the wishes of voters. Beyond wanting single transferable vote, there is less wiki community discussion.

"Single transferable vote" gives clearer outcomes when there are large numbers of votes. This election had 122 votes. There was a clear first-place winner.

The second place winner came down to three candidates. If any of these three had received one additional vote, that person would have been elected. There are even different ways to calculate a single transferable vote. In this election, a different STV calculation on these same ballots could have produced a different result.

Please organize a conversation reviewing the calculation method and the code which attempts to produce that method. It is beyond the scope of election facilitators to design the process for calculating the result, and instead, during an election the facilitators should be applying a well-discussed calculation system which already exists.

Start the election on time[edit]

The Wikimedia Foundation sponsored the translation of this schedule graphic into 11 languages.

Customarily election results are posted two months before Wikimania. The annual conference allows newly elected board members to quickly meet in person with representatives from many Wikimedia affiliates. Every Wikimania conference includes a meeting of the Board of Trustees, so candidates elected through this process will attend their first board meeting at Wikimania. As Wikimania occurs during the summer, it influences the schedule of the election.

Election planning begins late in the year before the election and people get particularly excited at the start of the new year. The Election Facilitators recommend making every effort to start formal election planning early, typically in January.

This election cycle began late because the Board of Trustees proposed a radical change to the election format to allow Wikimedia user groups to vote. Although the board expressed a wish for this in November 2018, there was no confirmation until February 2019. The facilitators could not reasonably start planning the logistics of the election before March 2019, cutting two months from the anticipated schedule.

For future elections, we strongly encourage making every effort to start on time! For this 2019 election, your facilitators met weekly with video chats for one hour each for three months to develop the process and cover as many details as possible. Please show respect for the process and give time for the Wiki community to participate. Six months at a casual pace is about the right amount of time, not four months as happened this year.

Discuss and raise concerns on wiki[edit]

At the top of every Wikimedia page there is a link to a discussion page!

More than 10,000 people likely directly discussed this election, and those people represented the several million people who are stakeholders in Wikimedia affiliate organizations. Among all this global discussion, only about 5 people commented on the election in the official discussion channel at the election talk page on Meta-Wiki.

Feel free to discuss the election off-wiki in the online and offline places which are right for your community. If anyone actually identifies a concern or wants to express an idea in the global record, then please either post that into Meta or establish a communication channel for your organization's voting delegate to bring that concern into the permanent, public, online, global record.

The election facilitators cannot watch all social media everywhere! If you want recognition of your comment, then post it in the right place. Please do not ask for facilitators to come to your off-wiki channel. Also this is a wiki-style election and the facilitators are not the election authorities or leaders, but only volunteer moderators who act at the behest of the community. Anyone who reads a question or request can answer it. Be bold and do the wiki response where anyone can help anyone else in public on the wiki.

Affiliate selected board seats process need control of election data by Election Facilitators[edit]

The election facilitators should hand out and collect the (digital) ballots.

Previous ASBS process used the private chapter wiki as a ballot box. This year the voter base was expanded to all affiliates, included user groups. As said, the chapter wiki is a private wiki, and not open for user groups. The WMF offered to use their Qualtrics subscription to collect the vote. Due to time constraints and the lack of alternatives the Election Facilitators accepted their offer. While Qualtrics works to collect votes, the Election Facilitators do recommend looking for other alternatives that might be more suitable for elections. The Election Facilitators also recommend that in a next election the Election Facilitators retain control over the election data, including handing out ballots and collecting them.

Fund the election in proportion with the election stakes[edit]

Please recognize that this election seats 2 of 10 positions on the Wikimedia Foundation Board of Trustees. The WMF has an annual budget of US$100,000,000, so the people in these seats will help direct the flow of $300,000,000. These seats have financial value, and staff at the WMF have a conflict of interest in matters which affect the outcome of this election. As a result, the Wikimedia community must design and ensure an election to be beyond any suspicion and reproach. There should be a public discussion about what resources are required to operate a fair and safe election which produces outcomes which everyone can trust. The Wikimedia community platform has the potential to be extraordinarily trustworthy and transparent about election security. Considering the importance of this election and what is at stake, and that to operate the election we need labor of various kinds, we should publicly discuss how to appropriately resource the election. Tasks which could be done with paid professional services include (1) quick translations to underserved languages where it would be a burden on the volunteer community to produce (2) secure operation of voting software which everyone can trust as an honest broker of ballots (3) some professional technical design of some interfaces and announcements to make this multi-lingual process more broadly accessible.

This author has asked the WMF to publish an account of their financial and labor investment in this election. When it appears, it will be made available through the 2019 election mainpage and announced on the talk page. The desirable accounting record would include the following:

  1. count of the staff hours invested in the election
  2. an approximation of the monetary value of those hours, based on whatever information is public or reasonable
  3. count of the number of staff who intervened in the election, including for design, coordination, software operation, producing text, or any other roles
  4. cost of consultants for translation or any other labor
  5. any cost for software, such as for the Qualtrics voting subscription or operation

Consider the weight of the votes[edit]

Wikimedia Deutschland publishes a nice annual report.
For new Wikimedia organizations, sometimes a coffee stain is their first publication.

There were 122 votes in this election. Wikimedia Deutchland's 2018 annual report claims 70,000 registered members, 80 staff, and an annual budget of US$10,000,000. There are other voting organizations which have only 10 members who chat in their local coffeehouse at their own expense. For this election each of these organizations had one vote.

This election set a precedent of equal weight for votes, but the precedent is not necessarily the rule. For this practice and any other election details, avoid treating the precedent as a rule and instead value Wikimedia community conversation as the respected authority. Ultimately this election is for affiliates and collectively the organizations determine the rules, but in practice, the affiliates want this election to proceed according to the established online discussion of people who propose and build consensus for policy on wiki.

Wikimedia Projects love emulating the United Nations, except we put every organization on the Security Council and give all nations veto power. The charming enthusiasm for this inclusive philosophy is a major draw to wiki participation, but we should balance and protect it with good discussions about justice and fairness. If you see a practice which merits discussion, then discuss it! Start talking now and plan the next election.

Make your own recommendation[edit]

It will never be easier for Signpost readers to make election recommendations and draft policy and guidance than it is now. Comment on this election at meta:Talk:Affiliate-selected Board seats/2019. Propose, start the conversation, and build consensus for whatever will work best. This discussion may well be where the next round of election facilitators begin their reviewing.