User:Ad Huikeshoven/The IGC bootstrap problem

The IGC bootstrap problem

We are all screwed by skewed distributions in our Movement

Establishing an (Interim) Global Council is not a green field operation. No green field approach will lead to an outcome with support beyond the people directed involved in its creation. This approach is workable (and recommendable) as far as you are willing to involve many voices from around the globe, for example by interviewing hundreds of them. An all volunteer Movement Charter committee will not have the time or capacity to do this. Hiring external research consultants might be necessary to get the task executed. The money is available in the coffins of the WMF, as hardly any penny was spent on research during strategy formation in the years 2018-2019, while a multi million dollar sum was earmarked for this purpose.

The Wikimedia Movement has evolved in twenty years to what it is. Codification of the evolved informal power structure would be the other extreme. Many people will recognize the outcome, and support it, because it is what is now. A small group (about 200 participants in Global Conversations and SWAN meetings) might want change. Most of them are solitary individuals and not representative of any entity.

Whatever you measure in the Wikimedia Movement, the distribution is heavily skewed, and most likely exponentially distributed. Out of 900 Wikimedia content project wikis just 3 of them make up for half of all active users. About two thirds of the Wikimedia project wikis combined (or 600 out of 900) make up for only 1% of all active editors. The language distribution is skewed as well. Three languages make up for half of active users. Forty languages make up for 95% of all active users.

I propose two example compositions. Assuming elections in which voters identify with candidates speaking the same language, I state hypotheses for their likely outcome.

Maximize diversity, electoral districts per language - Chamber One

Suppose you create a “diverse” Global Council of forty seats, with one seat per language, you would indeed capture 95% of all active users. The majority of small languages would “represent” only 8% of all active users. The 5% of users not represented on this Global Council are distributed over 260 languages. Their under of not representation can only be remedied by expanding the Global Council to include all 300 languages. However, the users speaking language not represented on this Global Council can be comforted by pointing out the “small” languages have a majority. Adding more seats to a one seat per language council will only further increase the dominance of small languages on this Global Council, and likely won’t bring in extra perspectives. On the other hand, a majority of those 40 seats will be European languages. Even some European languages are small languages. You have to add more seats to remove the European majority.

Maximize proportional representation - Chamber Two

Suppose you create a Global Council in which voters elect forty candidates from a single list. As per assumptions voters elect only candidates speaking the same language, the number of seats gained per language will be proportional to the distribution of active users over the languages. The most likely outcome will be that 32 out of the 40 seats will be from the top ten languages. (And Polish, Dutch, Ukrainian and Hebrew) aren’t in the top ten. The eight remaining seats - given the ranked voting system, and depending on preferences of not top ten speakers - can end up anywhere, and maybe at languages 11 to 18 in size. Those 18 different languages will represent at maximum 86% of all active users. You can add another 20 seats, hoping to seat another 20 different languages, bringing the number of represented users to 95%. In this expanded Global Council of 60 seats, the total number of different languages will be around 40, and the 32 seats of the top ten languages still have a majority. Most likely there will be a wide spread of opinions in the major languages on nearly all topics for discussions, so it will be unlikely the speakers of the top ten languages will ever vote as a bloc to get a majority. (The outcome of this projection excludes the influence of multilingual wikis, see below).

Ensuring equity in decision making

The Wikimedia Movement is in need of a delicate balance between diversity, representation of small languages, and hence underrepresented voices, and the vast majority of people who contribute to a handful of Wikimedia content project wikis. A bicameral council might be the solution, consisting of Chamber One and Chamber Two. Any decision needs consent by both chambers. Decision making in both chambers is by majority rule.

What about the representation of the multilingual wikis?

I did illustrate the outcome given the distribution of users over languages, and based on a strong assumption of user preference for candidates.

• Chamber One will have one seat per language. For each seat or language there will be a list of candidates from the native speakers of this language. Voters will be eligible to vote for the languages on which they have more than a minimum number of edits. This chamber can optionally be expanded with a number of multilingual seats, with a list of candidates who are active on multilingual wikis. Voters will be eligible to vote for multilingual candidates if they have more than a minimum number of edits on multilingual wikis.
• Chamber Two will be elected from candidates from a single list. Voters will be eligible if they have more than a minimum number of global edits. Chamber Two naturally includes candidates and voters from multilingual wikis.

What about the tech and developer community?

Most of them are employees of the Wikimedia Foundation and Wikimedia Deutschland. And there are about a couple of hundred volunteer developers active, probably less. If they have the numbers one of them might make it to Chamber Two, or a seat can be reserved for them in Chamber One.

What about the Board of Trustees, C-levels, and staff of the Wikimedia Foundation?

As, if and when they meet the eligibility criteria for candidate or voter, they can participate as candidate or voter in elections for Chamber One and Chamber Two. I won’t suggest to reserve specific seats for the WMF on the Global Council, to keep checks in place. The Global Council will be like a legislature, and the Board of Trustees of the Wikimedia Foundation for a large part as executive power.

What about the Wikimedia affiliates?

It is hard to see where they fit in. Affiliates exist in all sorts, shapes and sizes. There are geographically or regionally based affiliates, more or less respecting country or national borders. There are language based affiliates. And there are a small number of cross sectional and truly thematic (not language based) affiliates.

1. Firstly, the language based affiliates can be leading in organizing election campaigns for a seat in Chamber One. If for their language a seat is reserved, they can be active in recruiting suitable candidates. If for their language no seat is reserved, they can campaign for a candidate in Chamber Two, by trying to convince speakers of other small languages to vote for their candidate, or conversely, convince their constituents to vote for a good candidate from another small language.
2. Secondly, the regionally based affiliates. Most of them serve a very small number of languages. They can be considered as advocates for those languages. The argument follows the first point above as many times as languages spoken within the region.
3. Thirdly, the small number of cross sectional and truly thematic (not language based) affiliates, such as Wikimed, WikiEdu, and so on. Those affiliates are most visibly executing programmatic activities with an impact on wiki, and their program leaders will have great visibility within the global Wikimedia Movement. Leaders of those groups have made it through election to gain a seat on the Board of Trustees of the Wikimedia Foundation. They might well become very successful candidates in an election for Chamber Two. No need to reserve specific seats for them.

So, no reserved seats for the affiliates. For some, the same arguments apply as for the Wikimedia Foundation: they operate more likely as executive power, not to be conflated with the Global Council seen as legislature.

Judiciary powers anyone?

It is up to the Global Council when elected to create rules for, and establish an internal justice system, after passage of time, and conflicts between parties emerge, that is when they differ in explanation of rules enacted by the Global Council. As long as there are no rules enacted by the Global Council, there will be no such conflicts, and there will be no need for a judiciary.

How to solve the IGC bootstrap problem?

Determine the number of seats for Chamber One (42) and Chamber Two (60). Determine which languages will have a seat in Chamber One, including how many multilingual seats (2). Determine the number of seats (60) for Chamber Elect Chamber One and Chamber Two as per above description. This will be an *Interim* Global Council, with their term ending at Wikimania 2022. Their mandate will be to draft a Movement Charter, and oversee implementation of Wikimedia 2030 Movement Strategy recommendations. The IGC will have no legal entity. The Chambers will have the right to delegate preparatory work for tasks to committees consisting of elected members from their Chambers. (For example, Language Committee, Funds Dissemination Committee, or Affiliates Committee.) This might be a single Chamber committee or a Joined Committee, with members from both chambers. They might hire external expertise to the amount that is granted by the Wikimedia Foundation for this purpose. Both chambers can elect a chair to preside over their meetings. The Global Council might elect a (small) “Executive Council” to do whatever is necessary to do between meetings of the Chambers. The “Executive Council” is accountable to both Chambers. The task or mandate of the “Executive Council” is to be determined by the Global Council. The “Executive Council” might gain the prerogative to propose resolutions to the Global Council.

I can imagine both chambers conclude in their first session to delegate the preparation of a draft Movement Charter to a Joined Committee, and elect or appoint members to the Joined Committee. I can imagine both chambers conclude in their first session to elect or appoint an “Executive Council” to oversee the implementation of strategy recommendations, and report progress to quarterly meetings of both Chambers, including recommendations in resolutions for next steps.

How to get a Global Conversation or SWAN meeting to agree with the above?

Participants are either from a big language or from a small language. Those from the big languages will probably favor the Chamber Two part. Those from the small languages will probably favor the Chamber One part. There will be a handful voices of very small languages who will say they are underrepresented in the above constellation. Please reassure them that small languages have a majority in Chamber One. If this doesn’t convince them, you can tell them this is an Interim Global Council, and the number of seats in the Global Council isn’t determined yet. Smaller numbers will increase the influence of big languages, greater numbers will increase the influence of small languages. (The Movement Charter should not fix the numbers of the Chambers in absolute terms, but to be renewed every election based on the distribution of active contributors.)

There is no alternative. An initiative to create an association of affiliates (Wikimedia Chapters) has failed in the past. The idea to establish a legal entity has been rejected. People do like appointments to the IGC because of speed. They failed to propose a selection method. You might handpick 40 people for Chamber One from the forty biggest languages, one per language. Plus one from Commons and one from Wikidata. How to select by hand candidates for Chamber Two is nearly impossible without doing an election. So, I am in favor of elections for Chamber Two - and you can do an election for Chamber One at the same time. This takes some time to accomplish. Of course, it will be very unlikely the elected Interim Global Council suddenly decides to have a completely different structure for the Global Council itself, but it is not impossible.

Ratification of the Movement Charter

The Interim Global Council will decide by majority in both chambers in first reading. The IGC will be dissolved and a Global Council will be elected. The Global Council will decide by qualitative majority (two thirds) in both chambers to ratify the Movement Charter. The election of new Global Council members by active Wikimedia contributor will give the voters indirect influence on the ratification of the Movement Charter.

Bottomline

The way you pick and structure the Interim Global Council might determine to a large extent how the Global Council is structured, and those structuring rules written out, will be an integral part of the Movement Charter. I welcome other (re)solutions to the IGC bootstrap problem, and allow me to challenge your ideas about how they a) balance the interest of the small languages versus the interest of the big languages b) balance the interest of the on wiki contributors to the interest of offwiki (affiliate) volunteers.

Statements of Conflict of Interest The author is a former member of the Audit Committee of the Wikimedia Foundation (2007-2012), a former board member of Wikimedia Nederland (2012-2016) and a former chair of the Affiliate Selected Board Seats (ASBS) 2019 election facilitators. You might conclude that in the above I fully sided with “the community”, that is all active Wikimedia content project wiki contributors, which is per my intention. Ad Huikeshoven (talk) 12:12, 26 February 2021 (UTC)