Knowledge:Requests for comment/Arbitration Committee Elections December 2017

4666:

transparent gauge of sysop activity and familiarity with relevant policies, and that's pretty important. Editors editing is not a very good gauge of activity: maybe they're an active reader who only edits when there's a major revision to be made or something important to say, or mostly edits logged out and only creates an account to participate in areas like RfA or edit semiprotected pages, or I don't know what else. I think it's fair to say that if an editor is genuinely inactive, they'll simply ignore the notice or not see it in the first place, unless they're inactive because they have an axe to grind, but elections are the place to grind axes. Anyway it's unlikely that a few such editors can significantly influence this sort of election, if participation is high. I really don't think this is a problem. That being said, I'd be more likely to support an activity criterion based on total number of edits and not on a period of time, as I think volume of edits is a better indicator of intention to be a member of the community than editing recently is.

1482:. As someone who ahs run for Arbcom under the old (pre-message) system and the current one, I can attest the higher turnout is essential in making this a genuinely democratic process. Entirely appropriately, only a tiny fraction of the editor base spend time thinking about Arbcom or checking Arbcom calendar dates; but a much higher proportion are affected by Arbcom decisions including with respect to disruptive editors or long-running disputes. Encouraging a larger voter base helps encourage a greater interest in how Knowledge dispute resolution works (or doesn't). And that can only be a good thing. Also, a higher voter awareness encourages more detailed and relevant questions, and makes the Q&A session more than just a rerun of recent cases. Keep the mass message, for the sake of worthwhile Arbcom ballots. -- 3999:

editing very infrequently). Most of the time these people aren't problematic users, but users in good standing who haven't had time to edit but simply decided to show their support for other candidates who have done good work for the community. I also question how effective this measure will be: If a user is scrupulous enough to lay dormant for an entire year just to vote against people they don't like (or the reverse), then they'll surely notice a statement like this and make sure to make one or two edits to pass the requirement, rendering the whole thing moot. I don't think upping the required amount of logged actions/edits will counter this either since, once again, a user who waits an entire year just to spite-vote will probably be more than willing to edit enough to pass the requirements. Respectfully,

1045:

when balloting closes. That information is important both for editors like me who probably won't vote immediately for whatever reason, and also for editors who don't come across the notices until some time after the balloting period ends. Notices like this are vital to community function; editors concerned about having their watchlists spammed can very easily set their watchlists to ignore bot edits, and probably should. Furthermore, I don't support suppressing the notices from editors who are deemed inactive by an arbitrary standard but are otherwise eligible to vote. And as for blocked accounts, if an otherwise eligible editor is notified but is prevented from voting because they are blocked from editing at the time, well, tough. Don't do stuff that gets you blocked.

3444:

a role titled "Election Administrators", of which there were three, which sort-of evolved into the Electoral Commission. "Election coordinators" is a little more amorphous. There are always persons who take it upon themselves to do the necessary "grunt work" liking setting up this page, setting up the candidate and question pages, etc. In past years these volunteers were called Election Coordinators, and they were separate from the Election Administrators/Electoral Commissioners, who decided disputes that arose. I am not sure whether anyone actually calls themselves an "Election Coordinator" anymore. I hope that clears things up. :)

2140:

good thing and although some will nearly always find a reason to complain about something, keeping people informed should be considered necessary in order to maintain literacy of the topic. As with any other website or forum, not everyone is going to follow along the whole time so I would suggest allowing them to opt in or out if they decide how much communication they want. Maybe a couple different possibilities. One to let them know as the topic progresses if they want that and maybe a separate one that only informs them once at the end of the topic so they know what the outcome is.

1903:

community, and that includes those people who have less than 100 edits a month. While arbs don't set policy, the decisions they make have impact on every editor on Knowledge. That means giving the person who edits only on Weekends and sometimes skips a few months a voice. They have a right to have their preferences reflected on arbcom, and since it is a secret poll, I think most of them likely won't be joining any pile-ons, but will instead be voting based on their gut. Some might like it, but I don't think that is bad.

4957:, third check-box down. Let the users who keep complaining about bot edits polluting their watchlists figure out how to use that functionality, instead of imposing their niche preferences on the rest of us who are used to our watchlists performing a certain way, and (presumably) won't be able to "turn on" watchlist entries from this flagged bot. I occasionally monitor and report bots that are leaving notifications improperly (see current thread at AN about HostBot); this change will hamper that monitoring. 1177:: it will obviously done by a bot, so there can be solution for this. We usually have an option in watchlist to "hide edits by bot". I am not saying that everybody should turn it on. What I am proposing is, for this operation only, tweaks should be done somewhere so that edits by this particular bot would not show up in the watchlist (like autopatrolled users, and new pages feed). Once all the messages are sent, honourably revoke that flag from the bot. Everybody remains happy, including the bot. — 155:: In the case of proposals that change existing rules, or that seek to establish new ones, lack of consensus for a change will result in the rules from the 2016 RFC remaining in force. Some issues are not covered by the existing rules but will need to be decided one way or another for the operation of the election, in those cases it will be up to the closer to figure out a result, even if there is no clear consensus, as they have had to in the past. 305: 4694:

at all outside a particular topic. Editor X runs for Arbcom. One of said wikienemies notices, and promptly emails all the others, and/or posts on an online forum someplace. In response, other members of this group crawl out of the woodwork to cast negative votes. Now, to folks who work in uncontroversial areas, this may seem farfetched; to those who work on south Asian politics, Arab-Israeli politics, Gamergate, etc, this would be

30: 3032:. How can we be certain that this power will not abused and leveraged to remove questions which are simply inconvenient? I don't trust any proposal which proposes the creation of some new power, but which promises that it will be used only for limited purposes. That same promise was made when "extended confirmed" standard was created, and just as I anticipated, it was quickly expanded to allow virtually unrestricted use. 3373:"The mandate of the Electoral Commission is to deal with any unforeseen problems that may arise in the Arbitration Committee election process, and to adjudicate any disputes during the election. However, members of the Election Commission should intervene only when there is a problem that needs resolving, and either discussion is not working, the rules are unclear, or there isn't time for a lengthy discussion. 4775:

Maybe that's wrong, maybe I should reevaluate my priorities, maybe neo-Nazis should be made uncomfortable wherever they appear, I guess those are all matters of opinion. I think you're right that there's a debate to be had about who we consider members of this community, but I don't think we'll settle it here, and as long as we haven't settled it I don't think we should arbitrarily restrict who can vote.

3931:

while plenty of current editors do. (And sometimes the axe-grinding is understandable and not disruptive to the project). I recognize the good faith in which this was proposed, but the thought that disenfranchising longtime, constructive contributors whose real lives preclude them from editing for twelve short months is somehow preferable to welcoming them back to vote strikes me as unreasonable.

3061:

deserve much a reply then don't give them one, and let the community decide if that was the right course by voting for you or not. Questions from banned users, spammers, outing attempts or extreme personal harassment can be removed under IAR as being disruptive to the process. But otherwise, let's let people ask whatever they like, and Arb candidates can answer however they like too. --

4725:). In last year's election there were 1,942 votes, and the gap in net votes between the lowest-ranked elected member and the highest-ranked non-elected candidate was 180 net votes. Which means in last year's election 180 inactive editors would have had to "come out of the woodwork" for the sole purpose of voting against someone for it to have had any impact at all. There are about 3555:. It came close to having sufficient support, but the "numbers" show 58%/42% support, and I find it impossible to call that "consensus" on such a fundamental change to the status quo. In addition, the rationales of the supporters and the opposers are not significantly more persuasive than one another, so there's no reason to discount either argument. So, 4734:

be worked up enough to protest vote, and while it's not impossible I find it unlikely. I also think it's worth mentioning that the only candidate with a negative net vote was the one whose nomination statement was three words long, and who later stated in response to one of their questions that they weren't really taking it seriously.

1841:) that they got their talk page message a bit later in the process, after which they felt that candidates couldn't be bothered to answer their questions. If the notification messages are spaced out it shouldn't be for a significant amount of time to allow everyone ample time and opportunity to engage with the candidates. Respectfully, 2191:), and if other users "endorse" or agree with the statement, they would sign their name, no need to add any extra reasoning. In other words, users who sign their name under the "endorse" section are merely expressing their agreement with the statement that I wrote. No extra reasoning is required because I already provided it. 4509:, I think that this is probably an appropriate forum: its advertised on the watchlists, and has the stated purpose of considering changes to election rules. I also think having a parallel RfC would be distracting at best. Keeping all the activity in one place so it is read in context of the other proposals make sense to me. 2901:. and not only by the election committee, but also by any establshed user in good standing. Blatant harassment, baiting, and personal attacks are often disguised as questions. Unfortunately, the candidates are not permitted to defend themselves, and the lies and attacks are allowed to stand. Such comments should be removed 3210:, they need to be removed because these pages are already incredibly long, just counting the legitimate questions. If we practiced DENY and left them in place, we could easily wind up with an unnavigable mess. I suspect if anything gets inappropriately removed, enough users will be watching to ensure it's restored. 4623:

year away is enough time to be "out of touch", especially if they were very active in community discussions in the preceding years. To have any teeth I feel this proposal would need a longer time away and a requirement of more edits leading up to it, but those numbers are always going to be somewhat arbitrary.

1431:

ten edits in the last 6 months would strike me as a sensible figure, less active editors could still vote, but are unlikely to notice a talkpage message. It would also be nice if we could exclude sending messages to anyone with the retired or dead templates or publicly identified secondary accounts per

4898:

As of the time when I am posting this, the consensus is toward "yes, send the message". Some of the opposing voters are concerned about watchlists of "talk page stalkers" being flooded, while others are concerned about ability/competence of "random" editors. Who gets to vote is a different issue, but

4791:

My response to your organizing against a Neo-nazi is "yes, please keep doing that"; but the pedant in me wants to ask "but you can no longer vote against him, can you?" We're not removing a person's ability to advocate; just to vote, as a residency requirement would do in many (most?) countries where

4589:

Sorry folks, been meaning to get back on this. While I have to accept that changes to election rules have been and can be made here, I personally don't support something as serious as disenfranchisement being decided by such a small and self-selected set of editors. For that reason, I will personally

4397:

The vast majority of occasional contributors would not be disqualified by my proposal, so your oppose rationale does not actually address what I am proposing. Moreover, occasional contributors who go for a year or more without editing are going to have little to no interest in ArbCom elections. It is

4247:

I agree with filelakeshoe. While I usually do not like policy changes without some sort of proven justification, I could get behind a voting requirement for these elections, for the same reason we have them for steward and board elections. But I think it should be a meaningful standard, not something

3998:

While I think the underlying thoughts behind the statement are reasonable, I also believe that Rivertorch makes some good points. My home wiki is a smaller one, and when we have CheckUser elections people sometimes "come out of the woodwork" to vote despite the fact that they haven't been editing (or

3563:

of the debate: If this is brought up again next year, something that would be useful to people trying to form an opinion is the relative % of voters who would have been affected by this. Both to determine the seriousness of the potential problem, and the extent of the effect it would have on people's

2315:

was about inappropriate questions being asked. Following an internal discussion after we were notified of these issues, it was decided we lacked the jurisdiction to remove the questions (although it was later done in my personal capacity, from memory). Therefore, I feel we specifically allow Election

1934:

There will always be some votes that are not carefully considered, even among those who would have voted without the mass message. We could hope that some of those new voters would realise that they're not very clear on the candidates yet, and do a lot of reading to try to cast informed votes. I know

1866:

spacing is going to be an issue - too fast and you hit the API limit (and MassMessage cannot deliver the messages fast enough!), too slow and you flood peoples watchlists over several days. This year, I think it will be best to deliver over the space of 36-48 hours, and potentially straight after (or

1549:

I do not believe greater advertising would be beneficial. Users who are likely to give considered choices, with a knowledge of the candidates and why/not they would be a good choice for Arbcom, are the sort of people that should be voting. Editors that aren't aware of Arbcom elections are (generally)

679:

My support is somewhat tempered, however. Basically, I think that more participation is a good thing, so that's why I support. And my recollection is that there had been some kind of analysis that showed that first-time voters voted similarly to long-time voters (but maybe I'm wrong about that?). But

4980:

Yes. I am on the same page with you. Its not just about watching bots, but I prefer to keep track of every edit in my watchlist. Let it be by IP, or bot. So yes, I am never going to check "hide edits by bots" in my watchlist preferences. And for the same reason, I made it clear that this flag should

4774:

in the district in which I used to live, and even though I've moved away I continue to organize with activists who still live there against this person. Not that this particular individual stands any chance at election, but I personally feel that their life should be made difficult however possible.

4693:

Let me sketch out the sort of scenario I am hoping this would address. Editor X, over the course of doing good work in a controversial topic area, makes several enemies, who bear a grudge even if conflict is no longer ongoing. They do not edit regularly; indeed, they may not be interested in editing

3443:

The scrutineers are definitely separate from all the other titles or groups that you mention. My understanding is that they are stewards whose "home" wiki is not English Knowledge. There is no "election committee" unless that is what you choose to call the Electoral Commission. There also used be

3362:

If we have a system that has worked for decades, that's a reason to keep the existing system, not create a new one. Right now, removals of questions are allowed by any editor, with the usual rules about BRD, Consensus, and Dispute Resolution up to and including ANI. If there is a problem that cannot

3060:

No. From memory the message that was actually removed last year wasn't that offensive. And, bluntly, if you're running for Arbcom you will need to be able to deal with offensive questions and angry people. That doesn't mean that all questions should be treated equally - if someone's question doesn't

2470:

but really I would prefer it be left to the non-committee members running the election (i.e. the Election Commission) if at all possible. I would only like to see the Arbitration Committee involved in regulating questions asked for candidates for election to itself as a last resort, but I agree that

1885:

Have we considered how most that receive these notices are likely to be pile-on voters? This is hardly beneficial for discussion, and essentially turns it into a numbers game. The users who are likely to do the questioning, reasoning, and considered voting are those who are already going to be aware

1659:

in the encyclopedia. Furthermore, if one day there is a truly serious need (Knowledge itself being in desperate need of immediate donations, for instance), I have a strong feeling opt-outs like that would not obtain community consensus for an override, at least not quickly. In that sense, this issue

4755:

be wrong would be for me to return to vote in the same election against the same candidate X after moving to a different town, which is what I see as happening here. In any case, I guess this ultimately comes down to the philosophical question of who we consider to be members of this community; and

4750:

That's not quite an accurate analogy though, is it? As long as I live in a certain community, I have a certain connection to its government, even if I choose to neglect it; and if you manage to rouse me out of my stupor because candidate X once looked at you funny, that's silly, but not wrong. What

4733:

more than twice (some are bots or blocked) out of about 500 total unique accounts; the stats don't show how many have not edited in the past year and/or are blocked. You'd need to get nearly all of them who were reasonably active or about two-fifths of all of the editors who ever edited the page to

4622:

While I understand the thrust of this proposal I am not sure it would target the editors mentioned. For example if I was an editor that fell under points 2 or 3 all I would need to do is make any edit prior to the 28th October and then I would be able to vote anyway. I am also not sure that being a

4221:

Not sure what this really achieves. If we enact such a rule, people who were already planning to appear out of nowhere and vote in the ArbCom election will just click random article and remove some whitespace and become eligible, is there really a difference? If we are worried about inactivity then

4089:

May I point you to my RFA, in which there were multiple folks emerging from lengthy inactivity to cast votes, the nature of which I was far from the only person to object to. Whether that is disruption or the process taking its proper course might perhaps be debated, but I for one believe we may be

3460:

And for whatever its worth, I think the removal of disruptive questions is already within the scope of the Electoral Commission's authority, as is the determination of disputes regarding the removal of questions by other editors. As others have said, the authority to remove questions should not be

2279:

personal attacks, etc., the determination of what is inappropriate or off-topic is clearly to be left to the Commission. Multiple editors also emphasized their support was based on this exact wording; the Commission must decide to remove a question, not an individual commission member. Finally, it

2124:

There was concern expressed two years ago that the mass-messaging wound up being broadcast more broadly than was originally intended, and that this could have a major impact on the election results. At the end of the day, I think the feeling was that the same arbitrators were elected who would have

1917:

Sure, and I'm very much the infrequent editor myself. To clarify what I meant by pile-ons: it's fairly clear through the questions posed and their responses who has the popular support, at least in prior years. I'm hesitant to endorse anything that will bring votes that aren't carefully considered.

1794:

On the point of reaching sockpuppets, the frank answer is yes - I've no doubt there are many undetected sock puppet accounts out there, and it would be futile to try and get this bot to deliver to them. If this question relates to detected and blocked socks, I'm sure some arrangement can be reached

1790:

The main reason used last year to notify blocked users was that some of these were likely to be unblocked, and therefore eligible to vote, before the end of the period. I could look at, if desired, only delivering to users whose blocks are scheduled to expire before the end of the voting period if

1749:

I'm not sure about the particulars of the technical implementation, but if I recall correctly, last year a few notices were indeed sent to a few blocked users due to a difficult-to-avoid technicality. The effect of notifying blocked users is nullified, however, by the fact that the scrutineers will

5284:

votes? I wonder how many voters can see through the mechanics of the voting system being used, specifically that an oppose vote carries more weight than a support vote. I believe we should abolish oppose votes altogether and determine the exact voting regimen in case this change passes. I know I'm

4646:

Yes, obviously so, for any editor who is deliberately gaming the system, but the value in the suggested new rule is to get rid of deadwood, people who just don't edit anymore. Presumably anyone gaming the rule at least comes around often enough to see that an election is about to take place. I's

3955:

Regarding stale accounts, they may not necessarily be more likely to be compromised, but I submit that it would be easier to get away with compromising a stale account. After all, on an active account, the person who operates the account would likely notice the funny business, but in the case of a

2952:

Echoing AlexEng's concerns about vague criteria for removal. Other than blatant trolling, vandalism, and I suppose content which we would remove anyway (personal attacks, outing, ban evasion, etc.) I don't feel it's constructive for the election committee to be deciding what is a valid question or

2139:

As I mentioned in a previous comment I want to be clear that I am new and only created my account today and only came here because I saw the link in the top so I don't really know much about this. What I can say from personal experience is that deliberate and constructive communication is always a

1902:

I strongly suspect most people, like in real life elections, vote on their gut and for people they like and think represent them. While those of us who are the most active Wikipedians might not like that, I actually think it is a good thing in many ways: we want arbs who have the confidence of the

1581:

Per Keira1996. Asking people who know nothing about the Arbitration process, or that it even exists, to vote does not seem like it would have much of a beneficial effect to me. We can assume good faith that people will read up about the process and the role of the positions (and then read up about

1566:

Looking at some major votes (Brexit comes to mind), I am not convinced that a larger participation results in more informed votes. I remove the messages on pages of users who are not active, messages of 2015 and 2016 (sometimes more than one per year) which claim today "elections are now open". If

1430:

150 mainspace edits and not currently blocked. I'd like to see the currently active filter raised slightly, I'm sure last years test of one edit in the last twelve months will have meant the invite going to a bunch of people who retired or died almost a year ago. Either 1 edit in the last month or

1044:

An election of this sort can't be considered valid if the persons eligible to vote are not informed that the vote is occurring. Per Gerda Arendt in the section below, I support modifying the statement from "elections are now open" to "elections are open until ...", because it's not otherwise clear

4906:

this task. Because of that flag, the posting of this message should not be visible in watchlist (similar to auto-patrolled users, and new pages feed). This flag can be (honourably) revoked from the bot after task of sending this message is done. This way, everybody gets happy. Message is sent (to

4716:

Sure, I understand that scenario, but that's how elections are supposed to work. Say I find out a candidate is running in a local election in my district who I feel strongly about making sure they don't get elected. Maybe they're a neo-Nazi, maybe they drove over my lawn, maybe they wear a stupid

878:

Not only would this remind people to vote, but it will only be acted upon by people who care enough about it to follow the link and add their opinion. I see no drawbacks to the mass-invite. Editors who don't know much about it will most likely not vote anyway. But they might become interested and

250:

Very clear consensus to send the message out again. There were some suggestions for tweaking the way we did it last year - minor wording changes, trying not to send it to blocked editors, limit the number of accidental posts to odd pages, etc - but none were presented as separate suggestions, so

5453:

interacted with him (or read his answers to questions), 66.37% thought he was qualified for Arbcom. Obviously it wouldn't be desirable for someone to get 3 support votes, 1000 neutral votes and 1 oppose and win with 75% support, but that's an extreme edge case and I believe it's desirable to let

4190:

Which goes on to further the point I'm trying to make, this proposed rule is arbitrary and might potentially exclude genuine voters too and that should not happen. Also the fact we're assuming everyone with one edit in the stipulated period will be a genuine voter is also a dangerous assumption.

3930:

Inactivity is a funny thing. (1) Lack of edits don't necessarily indicate that a user is "out of touch with community proceedings". (2) I'm unaware of data suggesting that "stale" accounts are more likely to be compromised than active accounts. (3) Lots of "former editors" have no axes to grind,

3695:

I was considering this myself. This would by far be one of the less restrictive arbcom voter activity policies when compared to other language Wikipedias. If this is the first year doing it, I think activity within a year of the official "election season" is a good cutoff. This could simply be a

3606:

I will leave it to others to discuss whether the number of required edits/actions should be higher, but I believe we should at least institute this minimum requirement to prevent votes from users who are (1) out of touch with community proceedings; (2) stale accounts that are vulnerable to being

3390:

has the authority to remove a question if the current system isn't working, but so far we have not had to exercise that emergency authority. If we ever do, I fully expect a huge amount of drama from editors who are fine with "to deal with any unforeseen problems that may arise in the Arbitration

2985:

I just created an account today and only came here due to the link at the top but I echo the concerns above and only wonder what if any measures will be taken to prevent this being used to remove comments that are merely disagree or don't go along with the popular vote. There should be some very

2969:

Instead, I suggest to mark such questions, making it easier to ignore them. For example: "3 out of 5 election committee members regard this question as inappropriate. Candidates are not expected to reply. To make it easier to see which questions were posed in total, the question is not removed."

2526:

Yes, I support this idea very strongly. The Election Committee is set up to have the requisite independence and cluefulness, so I see no problem with that. And the overload of questions actually makes it harder for most members of the community to find the information that they want, so let's at

5601:

Assume I'm voting for A and B, and you are voting for C and D. If I'm opposing your candidates (a.k.a. 'tactical voting') and you remain neutral on mine, then A and B stand at 100% and C and D at 50%. Effectively, my vote carried more weight than yours. If everyone is opposing all not-supported

4469:

Though I agree in principle, I believe this is the wrong forum for this. This forum is surely for deciding on the practical running of the election, and not for changing the underlying election rules. I think such a proposal would need to go to a wider (RFC?) discussion somewhere, where I would

3340:

I understand your perspective, but I don't know how necessary it is to formalize the procedures for dealing with inappropriate questions. Some questions are obviously inappropriate and ought to be removed quickly. If the appropriateness is borderline, the elections people can discuss whether to

2090:

Technically, MassMessage is an extension, not a bot, and as nobots is not a standard feature of MediaWiki, this is why it is not respected (AFAIK - it's been a while since I read up on this, and this may have changed). Depending on how this is implemented, and what people think, either, both or

5396:

If we have candidates whom voters think are definitely not suitable for Arbcom then there needs to be a mechanism for expressing that. Also, I don't see how an oppose vote carries more weight than a support vote, given that the minimum Support level measured as Support/(Support+Oppose) is 50%.

649:

A handful of people may have complained about receiving the talk page invites, but people will complain about everything... including the complaints of other people. I think we are better served if more Wikipedians exercise their right to have a say in the governance of their community. I also

5010:

bots to malfunction, but it happens. In my role I end up watching a lot of blocked users' pages, and if the bot is placing notices where it shouldn't I'll probably be one of the first to see it, and flag it quickly so that it can be fixed quickly. I won't see it at all if we hide the notices.

4665:

I hope nobody minds I bulleted the discussion mini-threads above, to keep things clean. Demoting a sysop for lengthy inactivity is very much a different thing from excluding editors from community processes who haven't edited recently. Sysops performing admin actions is a very good and highly

3380:

In addition, while the Electoral Commission is not responsible for logistics of the election, the Commissioners should also help ensure that preparations for the election—such as setting up the relevant pages, posting notices of the election in the appropriate places, and asking the Office to

2109:

I'm amazed that people are voting against people being notified that they are eligible to vote in these elections. Most of the arguments are along the lines of "they won't be able to make a good decision about the candidates", which seems outrageous - why shouldn't they be able to make a fair

4418:

Without evidence to show how this is a problem, this just seems to disenfranchise editors who might have interest in the project, but for whatever reason, don't' edit (e.g. real life commitments, preferring to just read, lack of access to reliable sources in their area of interest, etc.)----

390:

that they didn't like having their user talk page or their watchlists spammed with these messages. However, I think greater participation in the election is a good thing, and the numbers clearly show that mass messages are effective at informing eligible voters that an election is happening.

2953:

not based on standards defined in this way (basically not defined at all). Although I would trust the committee to make such a judgement, I would prefer if the criteria were more fulsome in advance of endorsing this statement. I would likely reconsider if such criteria were written first.

1375:. It is important that ordinary, behind-the-scenes editors be given the chance to participate. Since they do most of the real work on Knowledge, they can bring a fresh and rational perspective, unlike the groupthinkers who spend all their time haunting the ANI and ArbCom discussion pages. 5710:, Kelapstick (who got modest support but a relatively low number of opposes) would not have been elected, Gamaliel would have been elected to a 2-year term rather than 1-year. And Rich Farmbrough, who polled the fourth highest number of opposes, would have been elected for a 1-year term. 1567:

they seem necessary please find a wording that doesn't grow stale in a few weeks. - I'd prefer a notice similar to the information that an RfA is open, or time for monument images. - Side note: can we we edit supports and opposes separately, please? You see above what happens if not. --

906:

election measurably benefits, encouraging participation is vital for the ongoing health of our processes. I feel confident that there will be thoughtful, informed votes cast in this election by people who wouldn't have done so had they not been drawn in by the previous years' messages.

5519:

No, the current system works fine. If we were struggling to fill spots because hardly anyone made the threshold then fine, but that hasn't been a problem in any recent elections. I also like the way the current system allows a voter to say "none of the above" by opposing everyone. --

5692:, had there been no Oppose votes allowed and the result determined solely by Support votes (which is really all that would be left), we'd have voted in the same set of ragamuffins and ne'er-do-wells ;-) The only difference would have been in the relative placings of Ks0stm and DGG. 2207:

It's probably possible to filter the list of users (based on some automated criteria like blocked users, bots, etc.) and to only send it to people in the SecurePoll lists, etc. If there was a ticket in Phabricator with suggested requirements with some advance notice, we can do it.

3841:

One current problem with the current rules is the possibility of having your account stop editing, creating a new account a couple months later, and using your first account only for voting in ArbCom elections - which you do from an IP address you never use for your new account.

5751:

It's certainly not a problem-free approach, I'll happily concede to that. But neither is counting support votes only. Just out of curiosity, how do you feel about support / oppose on RfA nominations? Is it different because it isn't purely a numbers game like ACE? Respectfully,

3479:

That is my understanding as well. I was on the Electoral Commission the last two years and all three of us strongly agreed that using that power was to be a last resort. I would like to see more voters who are committed to rejecting any commissioner who doesn't feel that way.

573:

The more participation we get the more accurate the results are likely to be, true we may get some people just choosing randomly from the list of available candidates but I would hope the people voting after reading the statements etc outweighs their affect on the results.

2054:

gets checked? If so, was that a conscious decision, or just the way it was implemented? I could see a case for making mass-delivery messages go through nobots because they are so much more important, but I strongly lean in the opposite camp (i.e. that someone who slapped

859:

The mass-notifications taking up my watchlist was a major issue and if I'm being honest annoyed the life out of me however as noted above we should get wider participation instead of a selected few so despite my groans as a whole the mass-notifications are a good thing.

4931:

The wording is a bit confusing here, but it seems that the user is proposing that the account sending the massmessage be flagged as a bot to avoid spamming watchlists with the ArbCom election notifications. That sounds reasonable to me. Maybe it's already happening? --

3607:

compromised; (3) former editors with an axe to grind against one of the candidates. Quite simply, if an editor wishes to exercise his rightful voice in matters of community governance, this editor should be expected to demonstrate some involvement in the community.

3341:

remove them or not. In theory I can understand why question-removals might become controversial and a distraction to the elections process, but in practice we've been holding these elections for more than a decade now and I don't recall many such problems. Regards,

161:: This RfC is divided into portions, each of which contains a discussion point for the community. The standard RfC structure will be used, in which any user may make a general statement that other users may endorse if they so agree. The points will be listed in the 5630:

You're right, but with a different voting system certain tactical options can be removed, or at least be made explicit. For instance, giving an amount of "tickets" equal to the number of available positions, and then counting total votes only. Or, someone made the

3807:

Somebody who isn't part of the community shouldn't be choosing ArbCom members for people who are. The standard here is extremely low, I don't think there will be anyone who can claim to be an active participant in the community who doesn't make any edits at all.

5448:

is cited as an example where the election would have turned out differently, and it actually illustrates why I disagree with this statement: Presumably, Kelapstick wasn't widely known which is reflected by the high number of Neutral votes, but of the people who

4568: 2229:

The way this has previously worked is that the WMF generate the list of eligible voters, and I merely applied an inactivity sieve to this - in theory, bots and ineligible users should already be removed, as they "should" be filtered out by the WMF script.

242: 4564: 318:

is open from Monday, 00:00, 20 November through Sunday, 23:59, 3 December to all unblocked users who have registered an account before Saturday, 00:00, 28 October 2017 and have made at least 150 mainspace edits before Wednesday, 00:00, 1 November 2017.

758:

Support with the proviso that notifications are made before the beginning of the voting period. This will prevent users who have already voted from receiving a message. (I distinctly remember discussing this before, but I can't find any record of it.)

2255: 4847: 4104:

I appreciate that reference, thanks. It's kind of a known problem at RfA that inactive users with a grudge pop in to oppose candidacies, but they do that without being notified by a bot. How do they know? My assumption has to be that they're active

3710:

Only those who are active should have voting rights, to prevent canvassing, and to make sure that only editors who are "in touch" with the community and who would actually be affected by ArbCom are voting. German Knowledge's RFA standards (and even

2510:-- only those questions genuinely unrelated to the process should be removed. The moment we start removing questions for any other reason means we enter a grey area where genuine questions may be removed, which is not a good thing for this process. 2485:

I support this up to a point: clearly inappropriate/offensive questions should be removed, but critical questions (even if POINTy) shouldn't be removed, and it's then up to the candidates to decide whether they want to answer them or not. Thanks.

3014:

Is there a "meh" option? We don't need an RfC to authorize people to remove blatant trolling. If a question isn't blatant trolling, but is just kind of stupid, meh. The job does involve dealing with dumb/inappropriate/annoying questions a lot...

205:: Normally, these RfCs begin on September 1 and last for 30 days. However, these elections are volunteer-run, and this year, the RfC is starting a bit behind schedule. In order to preserve the timetable above, we'll see whether a consensus after 3421: 3416: 3411: 3406: 3401: 1209:

As well as improving participation, this has had a positive effect on the quality of discussions around the ArbCom elections and the committee more generally by diluting the impact of the group of editors who tend to engage with/obsess over it.

469:

Unfortunately, there isn't an easy way to get a list of those who have voted from the SecurePoll system used in this election, so this feature is not possible without manually checking each entry, which I'm sure you can agree is not workable.

4867: 2289: 268: 3427:

Request for clarification: I keep seeing phrases like "election committee members" and "election coordinators". are these different from the electoral commission or the scrutineers? If so, who are the members and where is their page?

4371:

I read the proposal. I understood the proposal. I considered the proposal. And I made my decision. Knowledge very much depends on its occasional contributors (in aggregate, they do far more for Knowledge than do the obsessive and

3746:

It seems unfair to allow people to !vote etc who haven't been here for a year or more, Ofcourse not everyone can edit here 24/7 / 365 I get that but simply put if you want to !vote you should be an active member of the community.

4792:

democratic elections occur. Still, I think we can agree to disagree on whether a year's absence from Knowledge should operate in the same way: I'd be the first to acknowledge that my experience with such folks is influencing me.

3573: 3539: 3363:

be resolved the usual way, there is the Electoral Commission, which I have been a member of for the last two years. And of course there are always the scrutineers to double check that the electoral commission has not gone crazy.

1299:

Zero issues with this; if I don't want to vote, a simple revert of the message takes three seconds. Just as long as it's not going to socks or block/banned accounts, that's the only case I would think it's a waste of resources.

4149:

Current levels of inactivity should never undermine prior experience. I mean if the returning editor shows a lack of understanding, sure, but it makes no sense to senselessly punish someone because they just chose to return.

3827:, because I agree that inactive editors can come back to pursue grudges, and it's such an easy standard to meet. But as noted by some of the opposing editors, it's also ridiculously easy to game, just by making one edit. -- 2186:

This RfC was actually not intended to use the support/oppose/discussion breakup – it just kind of happened. Instead, if another editor had an alternative view, they would submit a new statement with a level 3 header (e.g.

1582:

the candidates), but I doubt most users who aren't as concerned with governance will go to such lengths. Encouraging people to vote on instinct is bad, especially when they make up a significant proportion of the votes. –

2312: 3274:

I would like a clear definition of who would be allowed to remove a question, whether one person could do so or whether you would need two or more people to agree to the removal, and the proposed criteria for removal.

4307:

Is there any evidence that this (1 through 3 above) is a problem that needs a solution? Even if this is a reasonable proposal—which others have made a good case that it is not—there should be a problem it solves.

5602:

candidates then it is fair, but if some apply the 'tactical voting' (without really having something to say against other candidates, just to strengthen their own stance), and some don't, then it is not fair. --

1838: 4599: 4501: 566: 4884: 1471: 2181: 4518: 2125:

been elected by the smaller voter group, somewhat reducing the worries that arose at the time. I think the revised distribution that was agreed upon last year was generally accepted as a reasonable one.

556:

I want to add that greater participation reduces the possibility of organized groups successfully pursuing a political platform - when very few vote, it's easier for political canvassing to win the day.

5719: 5701: 1681:

Please note that the ArbCom does not organize or run the election process, which is done by non-arbitrator community volunteers, and had nothing to do with the decision to start posting these messages.

664:

Sending notices to people who are eligible to vote to let them know that they can vote is something that was resolved many years ago by most voting systems, hopefully we can catch up with them! Thanks.

3265: 5178: 3076: 3006: 1719: 5644: 5625: 5611: 4902:

The point of this statement is: I think we should send the mass message with a particular criteria for recipients. To avoid the watchlist spamming, we can give the bot some sort of "flag", or access

1978: 387: 330:. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose 479: 2595: 1705: 1691: 551: 457: 5406: 5072: 4580: 4059: 2740: 1104: 726: 375:), sending it only to editors who have edited at least once in the 12 months before the start of the nomination phase (6 November), as opposed to every eligible voter. The turnout was still high: 5022: 4994: 4138: 1450: 962: 256: 4044: 4029: 2306: 1669: 1634: 1142: 1090: 5805: 5672:

Well I thought the principal unfairness (as perceived by me) should go first. If there is no consensus for doing away with the opposes, suggesting a specific system does not make much sense. --

5217: 4821: 4801: 4786: 4765: 4745: 4707: 4407: 4392: 4366: 4331: 4317: 4120: 4099: 3989: 3965: 3350: 3024: 2986:

specific criteria for when this can be used such as Vandalism, trolling, etc. It shouldn't be an open ended invitation for anyone to remove a comment they disagreement on their own discretion.

2029: 1944: 1929: 1912: 1255: 5775: 5761: 5591: 5440:

I don't agree with the reasoning here. Firstly it's incorrect to say that oppose votes carry more weight than support votes since the end calculation does not weigh the votes differently (see

4656: 4479: 4302: 3590: 2368: 1033:

I think this communication is useful, however I don't think it is warranted for fairly inactive users and should have a cut off of 1 (first choice) or 3 (second choice) years of inactivity. —

644: 5681: 5255: 5125: 4274: 3877: 3741: 3519: 3505: 3320: 2878: 2768: 1876: 1612: 1544: 1232: 1190: 5557: 5391: 5054: 4463: 3802: 3705: 3302: 2893: 2820: 2568: 2398: 2149: 2015: 1767: 1424: 1294: 897: 792: 617: 522: 5492: 5463: 5419: 5172: 5111: 5097: 5040: 4920: 4429: 4242: 4008: 3863: 3690: 3652: 3635: 3251: 2754: 2680: 2452:

Per Amortias, above, but on balance this seems like a sensible way to keep discussions on track. The "following discussion among the committee" caveat is essential to my support, mind you.

2200: 1576: 1028: 999: 753: 659: 5365: 5327: 3904: 3849: 3836: 3727: 3489: 3470: 3453: 3055: 2979: 2864: 2554: 2536: 2384: 1964: 1744: 1504: 1384: 1204: 1056: 778: 697: 5477: 5435: 5377: 5351: 5157: 5143: 4968: 4941: 4348: 4212: 4185: 4081: 3950: 3891: 3760: 3676: 3041: 2964: 2726: 2702: 2653: 2635: 2582: 2521: 2495: 2167: 1324: 1167: 1128: 873: 854: 820: 708: 674: 5543: 5514: 5313: 4257: 3774: 3192: 3135: 2781: 2416: 2239: 2100: 2068:

In any case I think a bit of pedagogy wouldn't hurt: writing about the notification process and its opt-out somewhere on the ArbCom election pages (at the bottom of an obscure page, but

1595: 1561: 1517: 1491: 1245: 1158: 1039: 587: 285: 5571: 5529: 5202: 5083: 4457: 3918: 3818: 3788: 3616: 3178: 3160: 3070: 2914: 2849: 2834: 2795: 2667: 2616: 2463: 2447: 2348: 1850: 1791:

desired? Unfortunately, with a messaging run such as this, context tends to be lost relating to the delivery - the bot last year purely looked at edit count and account age from memory.

1409: 1353: 1338: 1269: 985: 916: 806: 603: 372: 360: 5666: 4617: 4608:

The last sentence of the instructions here says "The results will determine the structure, rules, and procedures for the election." This is germane and within the scope of the notice.

2947: 2658:

Absolutely, although I think the removal criteria should be intentionally narrowly tailored per above comments, or (perhaps less preferably) some avenue for appeal should be provided.

2480: 2134: 1367: 1219: 630: 504: 4677: 4443: 4288: 3229: 2686: 1008: 346: 4557:

In the case of proposals that change existing rules, or that seek to establish new ones, lack of consensus for a change will result in the rules from the 2016 RFC remaining in force.

3437: 3284: 2119: 1897: 1886:

of the Arbcom Elections -- there is no need to reach out to those. Increasing participation is not in and of itself a good thing, I feel, unless it actually benefits the discussion.

1804: 1783:

Whether I take a role in this years election or not, I'll be happy to fire the bot up again to send out notification again this year - I could probably spread them out a bit more as

947: 835: 739: 5745: 4171: 3681:

In principle I'd like no restrictions at all, but after personal experience with folks emerging from lengthy inactivity to resurrect old personal feuds, I think this is reasonable.

1072: 419: 4637: 3111: 2970:

Marking inappropriate questions would even allow for automatically hiding them on the klick of a button. Also agree with Alex, can see no real danger that is not already covered.--

2217: 2078: 1821: 199:

There is one topic from last year that was raised on the talk page. Anyone is free to raise any new topics that they feel need to be addressed by adding them as level two headers.

5294: 213:, an uninvolved editor(s) will evaluate the RfC and determine whether a consensus has developed. The results will determine the structure, rules, and procedures for the election. 87: 2329: 217:

Use the following format below; post a new statement at the BOTTOM of the section in which you want to make a statement. Endorse by adding a hash symbol (#) and your signature.

5766:

No, RfA is fine imho because there is no minimum or maximum number to be elected. There I find that a well-motivated oppose rightfully counts more than a 'support, why not'. --

255:. Minor tweaks/clarifications to last year's message that don't alter this decision or the general thrust of the message can be discussed prior to sending the message out at 4590:

not take a side in making such a change here - obviously, that doesn't stop the rest of you from going ahead with it, but I do hope I'm not the only one who thinks this way.

2269:

Allow Electoral Commission members to remove offensive (eg. WP:POLEMIC-style statements) or off-topic questions from question pages, following discussion among the EC members

2926:

I am concerned with the broadness of the removal criteria, which sounds like it opens the possibility of abuse. Moreover, I don't find the change necessary. Why not simply

1077:

Sure, but Gerda's suggestion on the wording is a good one. As for the question of filtering out recipients who've already voted, it seems better to me to send the messages

1817:. Could the notifications at least be spread out over a couple of days? Or maybe a sitenotice or watchlist notice with css to display only to extended-confirmed users? — 440:

07:26, 10 September 2017 (UTC) --- BUT I hope there is some technical means of not sending the message out after one has voted, which is a waste of time. Happened to me.

5707: 5689: 3970:

That might be true. Then again, stale accounts may be less likely to be compromised in the first place. In any event, I think this is a solution in search of a problem.

2003: 400: 376: 368: 364: 5454:

people who have legitimate concerns voice those concerns by opposing, instead of voting for another candidate to make sure they get more support votes. Respectfully,

4448:

No. People's motivation comes and goes. That's the nature of volunteering. If you skew the community toward hardcore editors then you are out of touch with reality.

825:

I'm also supporting in my capacity as an editor. I have always felt that ArbCom should be based upon the will of the community rather than a smaller interest group.

1529:

I don't think greater participation is a good thing. I am specifically opposed to spamming user talk pages but I generally don't think asking for voters is useful.

3909:

A good suggestion. It happens too often that people who have lost interest in Wikipedi still come out of the woodwork to vote on various discussions and elections.

3256:

I also fear there would be knee-jerk oppose votes for not answering all the questions, however trollish - I've certainly seen RfA opposes for that kind of reason.

1455:

If possible we should do two versions of the message, one for people who also qualified to vote last year and the other for people who are newly qualified voters.

1147:

As a candidate in 2015, I had people phone me as a result of this message. It engaged a lot of people who would not have otherwise known. I strongly support it.

5800: 4376:

drama lovers), and we should not go down the path of disenfranchising them. Everything here eventually seems to go down a slippery slope; the rapid expansion of

1814: 145: 80: 4925: 1787:

suggests below - I would probably move away from using mass-message this year however (as you lose some control over spacing of messages and throttling these).

635:

The more participation the better. I often worry about discussions that only take place among people who happen to watch AN/VP/etc. to know they're happening.

5006:

I see what you're saying, but my point is I want to monitor the placement of this notice as well. I think you found the thread about HostBot already: I don't

4947: 3621: 5616:

There's always going to be tactical voting, and it can swing things either way - like, for example, supporting everyone except the one you want to oppose.

3924: 1696:

While that may be true, there seem to be several past and present ArbCom members !voting in the support section for this statement (and none in this one).

5299: 5231: 5577: 2334: 5333: 4129:, but a moment's thought will probably make clear how an editor with an animus against another can use their watchlist to be notified of a future RfA. 3086: 2920: 2403:

On the assumption that those removing the content should be those uninvolved in the process this year (not anyone running in this batch of elections).

990:

Yes, please message us. In general, for any vote, I think it makes sense to message any user who was active since the previous vote of the same type.--

101: 744:

I support notifying all eligible voters. Encouraging engagement and participation in community voting processes is healthy for the Knowledge project.

5272: 405: 5498: 5468:

I strongly oppose this idea. As noted above, users need to be able to register opposition, and to do so in a manner distinct from being neutral. --

2839:

It makes sense to have somebody with the power to remove clearly inappropriate questions, and the election committee are the best people to do so.

2745:

This was in fact done last year. I think it's a good idea and is likely to be necessary. I think we can trust the commissioners not to abuse this.

2110:

assessment of the candidates based on their election statements? If this is a concern, then can we ask better questions of the candidates? Thanks.

1523: 4070:

Is there any actual problem that this solves? Have any of our processes ever been disrupted by a sudden wave of inactive editors casting ballots?

1955:

Is it possible to instead of sending out a mass talk page message, place a notification on a user's Watchlist, like I had that lead to this RFC?

5077:

No, this would require special software changes for something so minor. The mass-message extension is the best way to deliver bulk messages. —

5370:

We need some distinction between an outright oppose and a n "I haven't had time to examine all the candidates, so I don't know abut this one".

4717:

hat, whatever. I'm going out and rallying everyone I know who is eligible to vote to go vote against this person. The security against this is

2786:

Yes, support per Chris Troutman, however, Beyond My Ken raises a valid point that there should be some oversight or consensus when it is done.

253:

Send the notification to everyone who is eligible to vote and who has made at least one edit in the 12 months before the sending of the message

5599:

A support vote carries more weight if accompanied by opposing all or some other candidates, compared to remaining neutral on other candidates.

4017:

also requires a minimum of 25 support votes, I'm not sure that people who "come out of the woodwork" on your wiki is a good example here. --

2707:

This sounds like a good idea which will make the process simpler and more 'friendly' for candidates. I also support Amortias' suggestion.

5382:

For much the same reason we can oppose an RFA. We should also be able to clearly oppose a candidate we think is unfit for this position.

5339:

No. It does not allow voters to express disapproval of a candidate (and there have been several clearly unqualified ones in the past). --

3604:

An editor is eligible to vote who: has made at least one edit or logged action between 00:00, 28 October 2016 and 00:00, 28 October 2017.

4222:

we need to set the bar a lot higher than one edit, and I would oppose a similar proposal with a higher threshold for other reasons. –

1163:

Essential to broaden the voter base and prevent the election from being dominated by super-power-users and those with axes to grind.

452: 435: 4035:

How come? My home wiki only has about 100 active editors, most of which don't participate in the meta side of things. Respectfully,

1550:

not the sort that would be making this considered choice in voting, and would vote purely on a basic impression of the candidate.

797:

Information is needed before participation is possible. Small cliques operating and self-selecting in the dark are to be avoided.

4357:. I understand and accept most of the responses in this section, but you don't seem to grasp at all what this proposal is about. 2905:. In many cases, Arbcom elections are more toxic than RfA, and this is not a way of encouraging candidates to run for office. 2051: 1625: 4176:

I don't think we can draft a workable criterion for establishing whether the returning editor lacks sufficient understanding.

3712: 4726: 4339:. This is just a way to eliminate the voting rights of those normal people who are not part of the addict class of editors. 1999: 972:

I look upon higher participation as a favorable outcome. The spam caused by the mass message is minimal and non-disruptive.

731:

I support this for sure - the community at large should be informed that it's happening so they can voice their opinion. --

5267: