Wikipedia talk:Arbitration/Policy/Petition (May 2023)

Regarding the amendment process in general

The requirement that 100 editors in good standing must sign a petition requesting the amendment is quite the unusual and archaic-sounding provision (see Wikipedia:Arbitration/Policy#Ratification and amendment). It dates back to Wikipedia:Arbitration/Policy/Update and ratification#Ratification referendum from 2011. Nowhere else on the English Wikipedia is this kind of ratification process required to change a policy.

I suppose the intent is to make it more difficult to change Wikipedia:Arbitration/Policy compared to other policies. But I wonder if even that goal is accomplished: the policy goes on to state that proposed amendments will enter into force once it receives majority support, with at least one hundred editors voting in favour of adopting it. Suppose there is a proposed amendment that becomes quite controversial, and 101 editors vote to support it, and 100 editors vote to oppose it. In such a case, it seems to me that both approval conditions are satisfied: the proposal has majority support, and at least 100 editors have voted in favor of adopting it.

However, nowhere else on the English Wikipedia do we decide things by simple majorities. We decide things by consensus. Even in contexts where we do make decisions through a kind of quasi-vote, a supermajority is typically required. For example, at WP:RFA, typically a 65% support percentage is needed to pass. Unless the arguments of a nontrivial portion of the 100 opposing editors were somehow deeply unprincipled, I would expect such a closely divided discussion to result in no consensus.

It seems to me that the arbitration policy may inadvertently be easier to amend than other Wikipedia policies. I hope I'm wrong. I hope that common sense will prevail, and we will determine the outcome of the second phase of this process just like we do normal policy RfCs. Mz7 (talk) 09:03, 11 May 2023 (UTC)

@Mz7 I think the petition part is useful to avoid bringing frivolous votes to the community. As far as the other part of this being a majority vote, arbitration related issues is one of the primary things that is decided by votes. We install committee members by vote with 50%+1 being sufficient to get on the committee. Then committee members issue binding resolutions - sometimes on the entire community - by 50%+1 votes as well. — xaosflux ^Talk 09:45, 11 May 2023 (UTC)

If that's the case, then this needs to end immediately. I'm not worried so much about the bureaucratic 100 person gatekeeping process. But Wikipedia policy is adopted by WP:CON. period. That includes arbcom policy. Sorry, but some animals are NOT more equal than others. - jc37 11:36, 11 May 2023 (UTC)

Arbitration is one of the means whereby we break the deadlocks that come from the consensus process. Nothing stops the community from deciding that a process should be decided in some way other than consensus. It has decided on an arbitration process that does not rely on consensus, and an amendment process that does not rely on consensus. That is the community's choice. It is in its way not so different from deciding, as it once did, that appeals from ArbCom should be decided by Jimbo, rather than by consensus. Wehwalt (talk) 13:02, 11 May 2023 (UTC)

You are not the first to comment on this particular point in the policy. Barkeep in the discussion that spawned this one also has his eyes on it. I'm sure others have seen it even before him (I'm pretty sure I've privately mused on it). I think this is a question worth reviewing, because I think the perception is that ARBPOL is too time consuming to amend generally right now (vice that ARBPOL is too hard to amend; I think it's in a pretty good spot on that measure). Adjusting how this process works to be less votey could be time consuming in its own way, as it might invite a general stream of change requests from anyone and everyone unhappy with the policy or some decision ArbCom makes.

There are some other comments made elsewhere in this discussion about other things that aren't quite right in the policy, so a general review of the policy may be in order. There may be other sore thumbs in it for committee work or otherwise. One thing I've been thinking about is the harassment RFC that occurred that hasn't been integrated in either policy or procedure in some way. Izno (talk) 15:46, 11 May 2023 (UTC)

One straightforward way to improve the amendment procedure would simply be to make the support threshold higher for the second phase. Either make it like a standard policy RfC where the consensus is determined by the judgment of an uninvolved closer, or make it like RfA where something like a 75% support percentage is needed to pass the amendment unambiguously and 65%–75% is a discretionary zone. Mz7 (talk) 16:46, 11 May 2023 (UTC)

Much higher, in the case where ArbCom asks to amend the document (8, not 100). Keep in mind there are two ways to start this process. Izno (talk) 16:48, 11 May 2023 (UTC)

Personally I think the low % threshold but relatively high editor threshold is a nice balance and is itself an important check on the committee. But yes I am concerned about what happens if there was a 4 day discussion that was at 100-100. Does the next vote decide what happens? If it's opposed can two more supports cause it to automatically pass? I think we've been lucky there haven't been these edge cases to date - in part because for a long time (but not any more) 100 editors was itself a really high threshold - but they should be addressed. Barkeep49 (talk) 16:48, 11 May 2023 (UTC)

If 100 was considered a lot of editors in 2011, what would be the 2023 analogue? What if we made the support threshold 50%+1 plus 200 editors in support? But then what if 10 years down the line, the population of our community changes such that 200 editors becomes too few again (or god forbid too many...). The 100-editor threshold is good at making sure that a discussion has thorough community participation (e.g. it isn't just a run-of-the-mill talk page discussion with a dozen editors), but it's bad at being a gauge for whether consensus exists. Mz7 (talk) 17:01, 11 May 2023 (UTC)

I actually wasn't proposing to change that, just noting why safeguards around the % are more important now than in the past. But I mean one could ask the same thing about 100 vs 50 vs 200 when it was passed. One way to address this would be to do it as some % of the electorate in the last ACE (e.g. 10% would get us in the 150 range ATM). Still the safeguards around the vote feel like a large problem and this feels more link something to tinker at the edges. Barkeep49 (talk) 17:06, 11 May 2023 (UTC)

I think there should be a minimum period the discussion is kept open, e.g. 14 days. If at the end of the 14 days there are not enough votes for either support or oppose, it is kept open for another 7 days for as many 7 day increments as needed. Whether the petition succeeds or fails it should eventually come to an end with enough participation.

I'd also support raising the minimum vote threshold to 150 or 10% of the ArbCom participation as Barkeep mentioned and requiring 60% support. I might support making this an RfA like system with crat closure, but I think the threshold should be lower than RfA since it is easier to oppose a policy than a person. I'm somewhat opposed to making this a regular consensus based process since as Wehwalt mentions ArbCom is where consensus deadlocks are broken. If there's any controversy over the process or closure, it's going to be impossible to resolve. Galobtter (talk) 19:11, 11 May 2023 (UTC)

No. If you want to borrow it, we already have a process for that, it's called a "crat chat". - jc37 19:28, 11 May 2023 (UTC)

I also don't see how anyone can meaningfully weigh the arguments for something like this. There's no higher policy to weigh arguments against. Arbitration works however the community best thinks it should. RfA already has a problem where crat closures can feel somewhat arbitrary because it's hard to say what's a good or bad argument for supporting or opposing someone being an admin. Galobtter (talk) 19:16, 11 May 2023 (UTC)

I understand the want to add minimum threshholds. But that can still be achieved and stay within policy.

Just say that a minimum of X number of participants in the discussion are required, and that Z number of supports are the minimum required, but that the discussion still needs to have a consensus determined for closure. If you want to add further gatekeepers and require a bureaucrat to close, so be it.

But all policy on Wikipedia is subject to the same standard, and that fundamental standard is WP:CON.

Arbcom only exists as a body due to both User:Jimbo Wales and the community, ceding authority to it to determine outcomes concerning users' behaviour-related issues. Let's not go down the path of a star chamber. - jc37 18:01, 11 May 2023 (UTC)

• The 2011 amendment, initiated by Arbcom, was declared closed after two weeks; there were more than 100 in favor and a majority voting yes. The 2019 amendment, also initiated by Arbcom, was closed after one week, with a majority, numbering in excess of 100, for. Following the rules and precedent would seem to dictate that the question be closed at the conclusion of the one-week period in which it attains 100 votes. If it becomes apparent that it will not attract 100 votes or a majority of those cast, possibly it can boldly be closed. I do not think there is provision for any requirements for passage beyond the majority of votes cast, with the those in favor numbering at least 100, nor should any additional requirements be instituted at this late date.--Wehwalt (talk) 22:13, 11 May 2023 (UTC)

  Even if we called either of those a consensus to violate the policy, consensus can change, or honestly in this case, consensus can be followed. As always, IWANTIT doesn't trump policy (except when it does of course : )

  A system was invented. Cool. That system violates policy. - jc37 22:51, 11 May 2023 (UTC)

  So does the system that elects arbitration members by that logic, since we vote in that as well. I do not think your argument is well-founded. Izno (talk) 22:52, 11 May 2023 (UTC)

  That system is policy. You can't arbitrarily claim that one policy overrides another. Galobtter (talk) 22:56, 11 May 2023 (UTC)

  Actually, WP:CON is the policy that allows for (most of) the others. Without community consensus there is no policy. The exceptions of course being things like legal and licensing. - jc37 23:05, 11 May 2023 (UTC)

  Also the policy that allows one unelected person to override an elected Arbitration Committee, without any regard for consensus. Wehwalt (talk) 22:58, 11 May 2023 (UTC)

So what is it that you're afraid of that you're hiding behind this one-sided voting process?

Someone to show up with an actual policy argument that the proposers and their friends want to ignore? Wikipedia's big enough that 100 people can show up to say or do pretty much whatever they want.

And if you don't think so, you weren't watching AMC and gamestop stock online gamesmanship - GameStop short squeeze.

Any system that sets a threshhold that does not allow for opposition to affect the outcome is honestly totalitarian.

WP:CON means that every voice gets heard and weighed.

You all are concerned about User:Jimbo Wales being a safety valve to stop arbcom. And yet on the converse are supporting a system where opposition voices are ignored in an Arbcom policy proposal. (rolling my eyes) No. That doesn't have power grab, and isolation from accountability, written on it at all. - jc37 23:14, 11 May 2023 (UTC)

You should review Wikipedia:Arbitration/Policy/Petition (May 2023)#Amendment process clearly. Izno (talk) 23:15, 11 May 2023 (UTC)

Thank you very much for the link.

The problem is part 2. It doesn't say it's a consensus of support, it just says support.

So (extreme example obviously), 1 million could say no, but as long as 100 say yes, it passes. - jc37 23:35, 11 May 2023 (UTC)

it must receive majority support. Izno (talk) 23:36, 11 May 2023 (UTC)

WP:NOTAVOTE. I think a good comparison is the banning of an editor. Arbcom votes, the community doe so through consensus. - jc37 23:39, 11 May 2023 (UTC)

An essay vice what is today a policy? You sure that's the one you want to pick? :) And as I said earlier, WP:ACE is also a vote by the community. Anyway, I get the sense that this is something you would prefer to change also, which was the point of this section. Thanks for your comments. Izno (talk) 23:41, 11 May 2023 (UTC)

Not at all. we use a voting hybrid when determining whether to grant tools or responsibilities to an editor. We went to voting with scrutinizers (ACE), because of the combined problems of size/scale, and of sockpuppeteering. And even that method isn't a straight for or against vote for a single thing.

The problem here is that arbcom's remit is editor behavior. Content (and that includes policy), belongs to the community. there have been discussions to cede some of that policy-making responsibility to arbcom, with mixed results.

I think I understand the goal here being to have gatekeepers, and to try to simplify. Gatekeepers I understand. But policy - especially policy that decides how arbcom works - should very much be consensual discussion by the community. Simplifying the process to voting because it's easier? That's not grounds to violate policy. - jc37 23:53, 11 May 2023 (UTC)

Simplifying the process to voting because it's easier? That's not grounds to violate policy. You can argue that if/when there is a discussion about changing the lines of interest (again, which is the point of the section, so it is ontopic). To be clear though, in the context of the discussion on the subject page of this talk page, the current policy is accurately reflected at Wikipedia:Arbitration/Policy/Petition (May 2023)#Amendment process. Izno (talk) 23:57, 11 May 2023 (UTC)

Well this is just the gatekeeper stage for this..

But my initial comment above was that - if we've moved away from WP:CON, then that needs to be fixed.

Now if you tell me that that my concern of making certain that arbcom policy follows WP:CON cannot be fixed by holding a standard RFC, then there's another real issue to take a look at. Arbcom is accountable to the community - not the other way round. - jc37 00:03, 12 May 2023 (UTC)

Indeed, the current arbitration policy cannot be amended in the way you propose until that section changes, which, yes, would include the discussion to change that section. With that, I think I have other things to do. I hope things are sufficiently clear to you about the current policy. Izno (talk) 00:06, 12 May 2023 (UTC)

Izno, I've always appreciated your willingness to discuss. And in case it isn't clear, nothing the above indicates that I don't respect the hard working editors who have volunteered their time to serve on the committee. I do. Very much so. And have said so many times over the years.

But checks and balances and accountability rules are not made for those we trust. I think it's fair to say that in early America George Washington was very respected. Some were thinking of making him King. But the founders there created the checks and balances for a time when it wasn't people like George Washington.

Anyway, since I've apparently been dismissed from this discussion, I guess I should consider what I should do as well. How much do I care about this volunteer project that we're watching people try to burn things down, while general attendance to discussions is obviously way down. Google is already placing imdb links ahead of Wikipedia in search results. How long until they, or Amazon, or Microsoft, do a fork of Wikipedia into their cloudspace, and then this will become what most big tech things become. AOL, prodigy, geocities, myspace, this isn't like it's a new thing.

What Wikipedia has, what makes it special are it's 2 fundamentals: 1.) this project is to build a free encyclopedia that anyone can edit; and 2.) Things are resolved through collegiate, consensual discussion.

You can choose to not believe me, or to not care. But changing those fundamentals will kill the project.

But I guess your mileage does vary... - jc37 00:22, 12 May 2023 (UTC)

I have not suggested that a change here would be unwelcome. As you'll see above, I myself have thought "that looks weird" about this point of ARBPOL. :) I look forward to the lively discussion I'm sure we'll all have outside the context of this one for changing it. Izno (talk) 00:26, 12 May 2023 (UTC)

lol - no fair making me laugh. This is supposed to be seriouz biznezz : ) - jc37 00:29, 12 May 2023 (UTC)

"editors in good standing"

What does that mean? QuicoleJR (talk) 19:25, 11 May 2023 (UTC)

not blocked. nableezy - 19:30, 11 May 2023 (UTC)

That's it? QuicoleJR (talk) 19:47, 11 May 2023 (UTC)

Do you have an actual concern, and, if so, how does that concern pertain to the actual proposal? Izno (talk) 19:59, 11 May 2023 (UTC)

It seems perfectly valid for someone to ask for clarification. Phil Bridger (talk) 20:02, 11 May 2023 (UTC)

I think the question is what does it mean in the context of who can sign the petition. nableezy - 20:22, 11 May 2023 (UTC)

Yup. Phil Bridger (talk) 20:02, 11 May 2023 (UTC)

Ok. QuicoleJR (talk) 20:21, 11 May 2023 (UTC)

Is an editor with a total of one contribution to WP, being a vote in this petition, an "editor in good standing"? What about an IP editor with many more contributions, like me? 172.195.96.244 (talk) 15:48, 12 May 2023 (UTC)

I find this more difficult than I expected, since "editor in good standing" is nowhere defined that I can see. I would like to see some minimal qualification so as to prevent people coming in and establishing accounts (or voting as IPs) because they have heard about it somewhere on the internet, but have had no editing history. That being said, it does not look like there were any restrictions on who could vote in the ratification votes in 2011 or 2019, but then again, no IPs or people without editing histories attempted to vote that I can see.Wehwalt (talk) 16:10, 12 May 2023 (UTC)

I also found we have no definition for it with a general search. In the context of amending ARBPOL, the ACE definition has a nice symmetry. (WP:ACE2022: Registered before 2 months ago, 150 mainspace edits before 1 month ago, 10 edits within the past year before 1 month ago, and unblocked. And not socking at the time of their vote.) Izno (talk) 16:59, 12 May 2023 (UTC)

Yeah, we definitely should institute minimum voting requirements like for that of ACE; that's something I saw too - ARBPOL doesn't as of now even block voting from IPs. Galobtter (talk) 17:49, 12 May 2023 (UTC)

I agree, if the petition qualifies, the voter criteria and polling duration should be the same as an arbcom election. Sandizer (talk) 11:24, 13 May 2023 (UTC)

I've removed the vote and issued a pair of CU blocks; the names whiff of one of our project space LTAs, and either way this fails WP:PROJSOCK by a mile. Izno (talk) 16:45, 12 May 2023 (UTC)

Based on the discussion, I suggest the following qualifications, based on the time of initiation of this process. This is otherwise taken straight from the ARBCOM election voter qualifications in the last election:

An editor is eligible to vote who:

1. (i) has registered an account before Friday 00:00, 10 March 2023

   (ii) has made at least 150 mainspace edits before Monday 00:00, 9 April 2023

   (iii) has made at least 10 live edits (in any namespace) in the one year prior to Monday 00:00, 9 April 2023 and,

   (iv) is not blocked from the English Wikipedia at the time of their vote.

Obviously we would not ask voters to prove their credentials, but votes can be challenged and stricken.--Wehwalt (talk) 11:47, 13 May 2023 (UTC)

I know that there are editors whose views on contributions by IPs are, shall we say, rather unfavourable. I know that there are Wikipedians who view that editing from an IP makes the contributor unworthy of being considered part of the Wikipedia community. I know that a lot of contributions from IPs are vandalism, harmful, or inappropriate. Despite all this, as an IP editor, I find the idea that, by definition, the phrase "editor in good standing" cannot be applied to any IP editor to be offensive. Please try to accept and incorporate the idea that IP editors can be Wikipedians and worthy of being recognised for our humanity, rather than defined out as de facto unpersons. 172.195.96.244 (talk) 04:37, 15 May 2023 (UTC)