Election leaks

Leak last year likely to produce changes for handling next board election

With the next election for the Wikimedia Foundation Board of Trustees looming in June, a discussion about administering the election prompted the revelation that in last year's race, election results were leaked while voting was underway. Although the outcome was not affected, this brought into question the appropriateness of the disclosure and how to ensure fairly administered elections in the future.

The issue came up after Lodewijk Gelauff, one of the stewards who handle user access levels on different Wikimedia sites, raised the question of how this year's elections would be run. With the matter of planning for the next election on the table, previous procedures were also revisited. As a consequence, some of the rumors circulating during the last election resurfaced and it became officially known that preliminary election results were released to the board while voting was still ongoing. Angela Beesley, the former trustee whose board seat the election was held to fill, called the situation "a disgrace."

As indicated, rumors about leaked election results were in the air last September in the days leading up to the official announcement of the final results. The Signpost was aware of some of the rumors at the time, but did not have sufficient concrete facts on the record to justify reporting it then. The rumors were fueled when Jimmy Wales took the unusual step of writing to several Wikimedia mailing lists with an endorsement of two candidates, Kat Walsh and Oscar van Dillen. (These two, as it turned out, would finish second and third behind Erik Möller, with whom Wales earlier had some public differences.)

Although others were also publicly endorsing candidates — Möller had a list of people endorsing him, while also identifying several candidates he favored besides himself — Wales's endorsement drew a mixed reception. Some, including David Mestel, criticized this as an unseemly interjection into the election process, a viewpoint Wales also acknowledged. As others like Gerard Meijssen argued, the message may not really have done these candidates a service — for example, Walsh expressed that she was not comfortable with being endorsed in this fashion, whether or not it affected her prospects.

The endorsement seems not to have played a significant role in the outcome, as the order in which these candidates ultimately finished did not change from the leaked preliminary results. Later, with Möller on it, the board decided to add both Walsh and van Dillen as part of the long-awaited expansion of the board.

Now that the information has become public knowledge, Dariusz Siedlecki, one of the election officials, confirmed that a board member did inquire about the results a little more than a week before voting ended, offering a promise not to reveal them. As Siedlecki described the sequence of events, the election officials decided to inform all of the continuing board members, apparently under the impression that they lacked the authority to withhold this information from the board. Wales sent his message the following day.

Details for this year's election have not yet been set, although it appears that some changes will be in store. Developer Tim Starling, who originally wrote the software used to administer Wikimedia board elections, suggested that the vote should be run by an independent group on outside servers, with no possibility of access for volunteers or staff members who might have an emotional stake in the results.


+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.
This kind of misses the point. The issue isn't that some of the Board members knew the results, but that those results were then distributed quite widely outside of the Board, including to some of the candidates, and to people who are known to have changed their votes after learning the results. Angela. 15:41, 11 April 2007 (UTC)[reply]
How widely, by whom, to whom, and who changed their votes? I try to limit my reporting to facts I know and have checked, and I don't have those kinds of details, certainly not on the record.
In any case, I'd still say the connection between the leak and Jimmy's endorsement (and more generally the sense that it was okay for board members, despite their personal interests in the outcome, to have this level of involvement in election administration) is more of a story than that. That a rumor mill among Wikipedians operates much like rumor mills elsewhere is not particularly surprising, unless there's evidence that the rumors were being spread in a concerted campaign to influence the election—above and beyond the publicly visible endorsements. --Michael Snow 16:04, 11 April 2007 (UTC)[reply]
I'd say it's a hell of story. Currently it is in vogue in the press to talk about Wikipedia's quality but it seems likely that it's governance may be the next thing to come under scrutiny. Consider:
  • Two resignations from the board in 9 months. Danny's resignation from employment. All three of these resignations appeared not to be for personal or external reasons but as a direct result of (perceived) problems with the running of the foundation.
  • Public bickering on foundation-l between the Chairwoman and an ex-employee hinting at questionable travel expenses (this will be seen in the light of five pages of auditor's comments about the need for better developed internal controls at the Foundation in their report to June 30, 2006).
  • Apparent attempt by Jimbo to shape the board election results without disclosing he was aware of results so far.
  • Possible conflict of interest between for-profit entities and non-profit Wikimedia Foundation. For instance, I read one commentator calling Jimbo ordering nofollow onto English Wikipedia days after announcing his own search engine a conflict of interest.
And with new elections coming in July, I am not sure we are done with the bickering. I'm not judging whether these should be big issues. But I am warning that that the press has a skill of blowing exactly these sorts of things into big issues. The Foundation needs to be ready to answer queries like these next time it is running a fund-raising campaign. Pcb21 Pete 14:27, 13 April 2007 (UTC)[reply]
There was only one resignation from the Board (Angela's). The other was Brad's, who was never on the Board. Some of the underlying issues must remain confidential but, yes, we do have reasonable answers to questions that may come up.
Thanks for the responses, Erik. Sorry for the mistake about Brad, you are absolutely right of course. I guess the very phrase "must remain confidential" grates on the nerves of Wikipedians who'd love to have the open/wiki way for everything but I understand when needs must. Pcb21 Pete 16:35, 16 April 2007 (UTC)[reply]
As for COIs, we have a COI policy and Board members regularly recuse themselves from decision making processes they have a vested interest in. The notion that the nofollow decision is a COI is an utterly ridiculous fantasy.--Eloquence* 03:54, 15 April 2007 (UTC)[reply]
That you needed to write "regularly recuse themselves" rather than "occasionally recuse themselves" itself suggests that preventing COI will be a more important issue for this non-profit than for many others. OTOH, I'm pleased there is a policy :-). The point about nofollow was more about what I'd read amongst some blogosphere opinion rather than my personal view. (Actually on that point my view on that is that Jimbo did make a mistake, and I hope one day it will be reversed, but it wasn't due to his own interests that he made it!.) Pcb21 Pete 16:35, 16 April 2007 (UTC)[reply]
I think the point is that it's a "regular" practice for board members to recuse themselves—as opposed to recusing themselves only "occasionally", which might imply that the policy wasn't being followed. I don't read it as a comment on how frequently conflicts of interest arise. --Michael Snow 17:16, 16 April 2007 (UTC)[reply]
Indeed. COIs are not anything "bad" and indeed very common in the non-profit world (where you have plenty of volunteers with personal business interests); it is the recognition thereof that matters.--Eloquence* 03:10, 17 April 2007 (UTC)[reply]
Thanks Erik. At this point I want to make clear I was trying to be helpful about what might come down the pipe, particularly with the elections later in the year. That point may have been lost in my subsequent posts. My father happens to be the chairman of a non-profit whose goal is to advise non-profits on achieving their goals, so I've ended up knowing a few interesting tidbits in this area by proxy. Preventing COIs from having a detrimental effect on a non-profit requires a lot of vigilance as the problems can sometimes be subtle. "Spinning off" (used in a very loose sense in the case of Wikia etc) a for-profit from a non-profit can be a particularly difficult area. I recognise that the board are all volunteers overwhelmingly doing their best for the foundation for which us regular editors should be grateful. Pcb21 Pete 10:52, 17 April 2007 (UTC)[reply]
Btw, If anyone else is still reading you can read the foundation's public COI policy. Pcb21 Pete 10:52, 17 April 2007 (UTC)[reply]
It was two resignations from the Board, not one - me and Tim Shell. Angela.
A little over a month ago I wrote '"I'd say it's a hell of story. Currently it is in vogue in the press to talk about Wikipedia's quality but it seems likely that it's governance may be the next thing to come under scrutiny.'". FWIW, the New York Times thought the situation over at the Mozilla Corporation was worth a story yesterday. Mentions Wikipedia in passing. [1]
Well it took longer than I thought... but mistakes made long ago about lack of internal controls are now having damaging consequences to the project, regardless of whether any allegations are true. Pcb21 Pete 12:48, 5 March 2008 (UTC)[reply]



       

The Signpost · written by many · served by Sinepost V0.9 · 🄯 CC-BY-SA 4.0