Wikimedia Foundation elections/2017/Post mortem

The election ended 11 June 2017. No more votes will be accepted. The results were announced on 19 June 2017. Please consider submitting any feedback regarding the 2017 election on the election's post mortem page.

Importance of the requirements that were set out in the Call for Candidates

This is a re-post of a suggestion that was posted to the main discussion page before this post mortem page was set up.

It's really frustrating that while the call for candidates set out a very sensible bullet point list of required candidate characteristics and experience, the election procedure makes it almost impossible for voters to assess candidates against those requirements. The stated requirements were:

• Have high integrity, the ability to stand up for a point, to resist pressure, think independently and be able to reach consensus;
• Have deep subject matter expertise in technology & product, strategy, finance, board governance or have some significant professional experience in the areas important for the Board's work;
• Have reasonable management experience (teaching, people management, project management, larger non-profit board experience, especially those of international organizations, etc.);
• Have experience from other boards or management positions, especially at non-profit or FLOSS organizations;
• Add to the diversity in some ways or have a proven track of activism in the area of diversity - including gender identity and expression, geography, sexual orientation, different cultural experience, different income experience, race, religion, family status, physical and mental abilities, etc).

Unfortunately, the candidate presentation template bears very little relationship to those requirements of the job, and positively encourages candidates to talk about other things such as 'online wiki work' (read 'boasting about high edit-count') which are of much less importance for intending members of a strategic oversight board of a major international non-profit.

Even worse, it's apparently not possible to ask candidates to expand on how their characteristics and experience measure up to the detailed requirements of the job description. The only community query relating to that seems to have been eliminated at some point in favour of a variety of re-written questions, many of which (such as as "why should we trust you?") are not useful in a job-interview as they are incapable of eliciting responses which can meaningfully distinguish between the candidates.

Perhaps this could be taken into account for next time, and used to improve the procedure for the next election. MichaelMaggs (talk) 11:43, 21 May 2017 (UTC)

In principle I support MichaelMaggs' suggestions and would encourage the Elections Committee to consider how candidates for both FDC elections and WMF Board elections could have a greater variety of qualities which are relevant to the roles be highlighted to the voters. Perhaps candidates could be asked to address each point in the job descriptions. --Pine^✉ 06:34, 22 June 2017 (UTC)

I have to support MichaelMaggs here about a million times. The current template and the current voting system (see Chris' remarks below) emphasize skills and achievements that, at leat in my opinion, are irrelevant to making a good board member, and worse, that actually put people who would be good board members (or good FDC members) at a disadvantage if they don't _also_have the wiki CV to shine on the things that are being put forward. So we end up with (almost) always the same kind of results. White, male, from Global North countries, with a high edit count... Results seem to put new comers, women and people with diversity backgrounds at a (real) disadvantage. So yes, putting forward the things that *really* help those official bodies do their job well in the application and in people's mind would make a lot of sense. A better adequation between the requirements and how the candidate statement is put together makes a lot of sense. notafish }<';> 06:50, 16 July 2017 (UTC)

What to do when the number of candidates == the number of positions?

This was the case for the FDC ombudsperson this year, where there was one candidate for one person, but it's also worth thinking about more generally. In particular:

• Is it worth running the election in this case, or should the result just be announced?
• Should there be an extension to the candidature period to try to get more candidates?
  • Experience (well, mine ;)) shows that does not work. notafish }<';> 06:51, 16 July 2017 (UTC)
• Should there be a minimum % support for the candidate?
  • Yes, I think that would make sense. notafish }<';> 06:51, 16 July 2017 (UTC)

Thanks. Mike Peel (talk) 22:02, 14 July 2017 (UTC)

Thoughts from Chris K

Hello - I had a number of thoughts about the process which I've put below. While I am grateful for the work of the many volunteers and staff members involved in running the elections this year, I think there are a number of ways the election could be significantly improved. These thoughts are based on my experience as a candidate in these elections, as a facilitator of the last two previous affiliate selected board seat processes, and from being involved in countless other elections to voluntary-sector committees one way or another. Regards, Chris Keating (The Land) (talk) 09:56, 15 July 2017 (UTC)

Forward planning and consultation

The first we heard about the election was the call for candidates. There was no public announcement of timescales and no public discussion of any aspect of the running of the election. I imagine that in this case it's because there was little time between the appointment of the long-overdue Standing Elections Committee and the minimum timescale to actually run the election. If so, that's understandable, but still a shame. For the affiliate selected board seat process, there is normally consultation about how the process will be run from several months in advance of the call for candidates. Of course, that is a much simpler and lower-profile process - but all the more reason why the complex higher-profile community elections should see more consultation and public planning not less.

Communication with candidates

I was surprised that no-one from the Board or the Elections Committee was in contact with candidates at any point during either election, except staff members confirming receipt of identity documents (edit: not 100% true, the EC were in touch about organising the Youtube hustings, but my points still stand). An email saying "thank you for standing, just to confirm this is the timescale, if you have questions please get in touch" would be nice. It would also be nice for candidates to receive an email with the results, possibly even slightly before they are published. We usually do this for the ASBS process, because we'd prefer candidates to find out whether they have been successful or not directly from the facilitators, rather than having to continually refresh Meta or have people pinging them on social media to tell them the results are up.

Questions

I felt the idea of "compositing" questions in the Board election was basically a good one, because question fatigue is a real thing, and some questions submitted are frankly not relevant to anyone except the person asking them. It would probably be helpful/possible for some questions to be merged/confirmed in batches, perhaps starting before the deadline for submissions, so that candidates aren't confronted with all of them at once. I also agree with sentiments from elsewhere that the merged questions turned out to be a bit bland.

I got the feeling that no-one really knew what was happening with the FDC questions. It wasn't clear whether questions were going to be merged, or whether there were cutoff dates for last submission of question/first answer. I know a number of candidates were uncertain about when/where/whether they were supposed to be answering.

Youtube hustings

We ended up having a Youtube hustings, which was good. I expect this would be valuable in future years, and more valuable with more planning and more publicity. It would also be good if in future years moderators were stricter about keeping people to time. Some candidates made no effort at all to stick to the timings we were given, which made the whole thing disjointed (and probably boring to watch).

Voting system

It has long been my view that we should change the voting system used for these elections. I can see literally no positive points about the current system, which was designed for situations when you are e.g. selecting administrators, where you are deciding whether one person should be added to an indefinitely large group of people. In these elections, it has the following flaws:

• Being a majoritarian system, it has a tendency to produce homogenous results. Imagine a situation where 2000 people favour candidates ABC and 1500 favour DEF. Under our voting system, candidates ABC will be elected and the views of the minority will be completely ignored. One might argue that our situation is so diffuse that this doesn't happen in practice, but it's still the tendency that is built into the system. A preference voting system (STV or Schulze STV) would be much better at reflecting the whole community (even if I don't believe it would have changed the result of at least the Board election this time)
• The support/oppose/neutral system results in situations where it is very difficult to not be re-elected, or very difficult to be re-elected, depending on the mood of voters. Most years there are votes for the incumbents because most people think things are going fine (this year only one person was elected to either Board or FDC who had not previously served on that committee previously). In other years (e.g. 2015) there are a couple of hundred extra "oppose" votes for the incumbents because the Board has done something controversial, and there is 100% turnover as a result. Don't forget that oppose votes typically have 4 to 5 times the power of support votes.
• Finally, voters have to make a weird decision about how to use support, neutral and oppose. If you think 2 candidates are great, 2 are good, 2 are OK and 2 are awful, what do you do? Support your top 2 and oppose everyone else - or support your top 4 and oppose the bottom 2? The latter might be the "natural" way to vote for many people, but if you do that you are harming the chances of the candidates you really want elected as you're potentially helping the people you'd rank 3rd-6th beat them.

So I would urge the Elections Committee and the Board to think about what characteristics it wants the election process to have, and then choose a system based on those characteristics, not continue using s/n/o voting.

Nature of postmortem review

Shortly after the election I asked whether there would be a structured post-mortem review, or whether I should just write my thoughts on Meta. I was advised to wait for the formal post-mortem review, which I have done. Now, I'm just posting the same thoughts on Meta. I was hoping for a more structured format, possibly with specific questions prompted by the Election Committee's own reflections on how the process went. Will the EC be posting their thoughts here, or will their review be conducted in private?

Requirement to know English

There's a requirement in the call for candidates:

All potential candidates must be able to communicate and discuss in English, since this is the language we use in all our board discussions.

This disqualifies most of humanity.

Now, I acknowledge, of course, that it's a practical matter. Board discussions, oral and written, are in English, and interpreting them all in real life is probably not so simple.

I also acknowledge that five out of nine of the current board members are native speakers of languages other than English. This is good, but still, they all know English, and a person who knows English, even as a second language, cannot experience how it is not to be able to read the English Wikipedia. A recent Twitter thread by the official @Wikipedia account proudly showed the many articles Wikipedia has about squirrels, but forgot to mention that most of these articles are available only in the English Wikipedia. It's too easy to forget about the privilege of having English as your fallback language, but this privilege is very much a thing—Google "english language privilege".

I'm trying to take our motto "a world in which every single human being can freely share in the sum of all knowledge" literally, and since most people of the world don't know English at all, it is a problem that these people aren't directly represented in the Board. And thus, decisions about strategy, budget, and priorities are made at best with indirect participation of people who don't know English, who can make their voice heard through liaisons or translators —who sometimes use machine translation, which is never perfect—, and at worst, without their participation at all.

We should, therefore, not only allow people who don't know English to be on the Board. We should actively encourage them to join.

Of course, I am not saying that not knowing English should be the only criterion. Over the years I met plenty of people who don't know English, but who conform to the other requirements: integrity, finance expertise, project management, NGO experience.

I don't have an easy solution to the communication problem. Ideally, all relevant written and oral communication would be interpreted from any language and to any language. It might be costly, but it's worth it because without people who don't know English, the voice of the majority of human beings is not heard in the Board.

I know very well that this sounds radical and impractical, but it would be useful to at least acknowledge the problem. --Amir E. Aharoni (talk) 11:54, 15 July 2017 (UTC)

Hello Amir E. Aharoni. I understand where you come from and wonder if the elections are the best place to enforce this. My take is that with equivalent qualifications, a person with no knowledge of English would have no way to be elected, given the constituency of our voters, who mainly come from the English Wikipedia. So I would say that this might be something to really consider in the appointment process. notafish }<';> 06:39, 16 July 2017 (UTC)

given the constituency of our voters, who mainly come from the English Wikipedia—where does this information come from? I have looked at the statistics page and I couldn't find it, but maybe I haven't looked well enough. If it's true, it is in itself very problematic that the strongest Wikipedia community reinforces itself yet more.

It's possibly true that a Board member who doesn't know English has a better chance of getting appointed rather than elected, but if you agree with the general idea that people who don't know English should be directly represented in the Board, do you also agree that the requirement to know English should be removed from the Call for candidates? If a candidate who doesn't know English has a low chance of getting elected anyway, this makes the explicit requirement all the more unnecessary. --Amir E. Aharoni (talk) 08:07, 16 July 2017 (UTC)

And here we come to the next problem of anglophone arrogance: Not even the election messages were translated in any language when they were deployed to the different projects, and that must be a strict No-Go! I start a new thread about that. Grüße vom Sänger ♫^(Reden) 08:21, 16 July 2017 (UTC)

+1 It is the sum of all knowledge, not the sum of all knowledge in English. By inviting and electing only fluent English speakers anglo-centrism on Wikipedia is amplified and institutionalised. To be honest, I don't know how a Board should work where members cannot speak each other's languages. But I also think that this is a solvable challenge. --Pgallert (talk) 08:48, 18 July 2017 (UTC)

For the sake of precision and fairness, the requirement is actually not for fluent, but for basic English, and there's even a commitment to help improve it. But still, most people don't know any English, and pre-filtering them is not quite right. --Amir E. Aharoni (talk) 20:54, 19 July 2017 (UTC)

Support from me, too. If we take the international approach seriously, we need to overcome the requirement to speak English. Of course we help board members to improve their language skills, but in the end that also increases their workload and it's only possible with an adequate time commitment from their side. The challenge is to find new ways for information and exchange within the board or other committees like the FDC, and to be honest, that hasn't been discussed so far. A change will take its time to be prepared (and won't be an simple thing to do), and I assume that it won't come for free. But even if "since this is the language we use in all our board discussions" is true for today, it doesn't have to be the only option for tomorrow. I'm glad that you raised the issue here. Alice Wiegand (talk) 20:11, 26 July 2017 (UTC)

This is a good discussion to have. Do we have examples we can refer to of specific NGOs or bodies that deal with multiple members with different first lanaguages/sole language speakers? I am sure the UN must provide examples as well as international development organisations. How do they get around this - use of translated documents and translaters? What problems does this create beyond a resource cost? Leela0808 (talk) 15:49, 25 August 2017 (UTC)

There has to be a requirement to publish election mass messages in at least 20-30 languages simultaneous

Unless there are at least translations to the languages of the 20-30 biggest (in terms of community, Botpedias like the Cebuano one could be discarded) projects, it cannot be announced properly, and no english announcement should be made beforehand. English alone is far from enough, and posting English texts to, say, the German, French, Russian, Chinese Wikipedia is extremely rude. WMF has more than enough resources (volunteers and/or money) to make sure this can be accomplished. This time there was nearly no time for volunteers for translation, and no proper attempts for other solutions were made by the election committee, see for example this thread. This must never happen again. Grüße vom Sänger ♫^(Reden) 08:21, 16 July 2017 (UTC)

+0.5. Not quite +1 as I think making 20-30 languages mandatory would be too ambitious. But I think reviewing the call for candidates, and creating space for it to be translated into a range of languages, would be really beneficial. Chris Keating (The Land) (talk) 09:44, 16 July 2017 (UTC)

If I had only asked for 5-10, I would (rightfully) be called a arrogant western/northern guy from one of the big languages. Of course German, French, Spanish, Japanese, Russian, Italian, Chinese, Portuguese, Persian and Polish as the ten biggest active communities (after English) in WP should be taken, but the rather lush availability of millions of bucks and/or many volunteer translators should make this enterprise rather easy. Grüße vom Sänger ♫^(Reden) 08:27, 21 July 2017 (UTC)

Maybe we have to discuss and explore different paths for translation overall. It's not that anybody doesn't want to have translations in time. Especially with meta-WMF-issues like elections, policies and more my perception is

• the interest of community translators is limited
• the language used in the originals is an obstacle for those translators who are interested

I say that as someone who is working on translations from time to time and who struggles a lot with long sentences, undefined terms and unclear messages. At the same time I remember the approaches to get some of the documents translated by professionals. That's where I realized that our terminology is special and you can't learn it with a youtube video. So professional translators can to a lot - which will cost a lot. And their output still needs a huge portion of love and rework by community members. What's missing is a new approach to make translations more interesting. Not sure how to do that, but pretty sure that we can't resolve it without community involvement. Alice Wiegand (talk) 20:36, 26 July 2017 (UTC)

The use of English in the original call for candidates can be improved, to be honest. It's easier to translate simple, clear English. The call for candidates is not simple or clear. What does "We would like the Board members to be a competent representation of our communities and projects" mean? Chris Keating (The Land) (talk) 20:46, 26 July 2017 (UTC)

Thoughts from Pgallert (talk)

1. Nomination period notification: I received a private email on the eve of the end of the nomination period, asking me to run. Without it I would have missed the deadline. Not sure how prominently it was announced but I didn't see it.
2. Voting system: I support the criticism of Chris K above, particularly because a tactical vote carries more weight than one that just supports the favourite candidate(s).
3. Interaction with/about candidates: In the questions there was a lot less Jumping-Through-Hoops than in the 2015 elections, which is good. Still, two questions ('Why should you be on the Board?' and 'What are your priorities?') were completely superfluous, because that information should be in the candidate statements. I still feel that questions to individual candidates should be allowed, as there might be certain issues that apply only to one or few candidates. I also feel awkward about voters' guides and think they should not be on-wiki: This year, one guide praised me more than I deserve, and one, I believe, criticised me unfairly, so for me it was 50/50. But I would have prefered to be confronted with both views in form of a question or discussion.

Cheers, Pgallert (talk) 08:39, 18 July 2017 (UTC)