Fundamental reform of Wikipedia, ideas wanted Options

[Peter Damian]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

This post has been edited by Peter Damian: Sat 12th September 2009, 12:04pm

[Nerd]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

For one, I'd definitely have named editors only, so there's some accountability.

[Hell Freezes Over]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

1. ArbCom elections with a secret ballot and candidates using their real names during the election, not just after being appointed. I'd form a committee of trusted editors tasked with ensuring the offered names are genuine. The outcome of the election alone to determine who gets appointed.

-- It would reduce the likelihood of socking to get on ArbCom (though it wouldn't eliminate it). Would make the process and outcome more professional.

-- Downside: it would discourage people who've edited in contentious areas from standing, because they'd be less likely to want to identify themselves. That would mean fewer people on ArbCom with experience of handling contentious content disputes. However, the current process already favours people who've not been involved in disputes, so I don't think we'd make that aspect any worse, and I think the benefits would outweigh the drawbacks.

2. Either (a) no more BLPs unless the subject is already the subject of a biographical article in a high-quality mainstream publication, preferably an encyclopaedia; or (b) the subject has the right of reply in a pop-up box that's visible over the article. The latter was an idea of Patrick Byrne's.

-- The first option would remove most of our BLP problems, but would leave the project a lot less informative, so I'd be more inclined to go with (b) if we could make it work.

3. Everyone who holds a position (admin, Arb, checkuser, bureaucrat) to hold it for six months of the year only, and for the rest must work as an ordinary editor. The software would automatically give and withdraw access on a six-month rotation, assuming that could be done technically.

-- Would reduce the divide between editors, admins and others. Would prevent (or at least wouldn't facilitate) the formation of an admin class that administers the project but doesn't add content.

[Kelly Martin]

All positions of authority (including GodKing) either elected by secret ballot or appointed by standing committees, and all committees with appointment power must be elected by secret ballot with members required to stand for periodic reelection.

All broad policy to be decided by majority (or, when appropriate under the ordinary rules of deliberative assemblies, a two-thirds supermajority) vote, and such voting is to be conducted by secret ballot. All votes to be held in a central place, with clear notice of the opportunity to vote given to all community members (something akin to the "orange bar of doom" used for talk page messages).

I'd also implement a jury system for arbitrating disputes.

[victim of censorship]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

I would purge the servers, sell the IT equipment, sell the domain name "Wikipedia" and take the monies and give to some inner city library in need of real books.

[Angela Kennedy]

I would say "Thanks but no thanks Jim. You got yourself into this mess."

[John Limey]

1. To tackle content issues: Create an Editorial Board (or more like several - one for each subject area) with ultimate authority over all content issues composed of people with real-world expertise and credentials (those on the board might or might not already be Wikipedians).
2. To tackle vandalism and sockpuppetry: Display IP addresses beside usernames in page histories. Require account creation to edit along with a valid email address (a very low bar, but one that provides some protection). Allow users especially concerned about their privacy to hide the IP address only by registering a non-free email address (as is done on WR).
3. To address BLP issues: Create a system with a stable, displayed version of an article (that can not be edited) and a NOINDEXed workspace that can be edited. Periodically update the displayed version from the workspace after an approval process. In other words, a much much stronger version of flagged revisions.

[Lar]

More than three things are needed but since we need to pick the top three, I'll reiterate things I've said before

1. Real names, backed by a validation system at least as good as Amazon's. Can't be perfect, doesn't have to be, but something that costs money to game is goodness.
- This solves a whole host of problems at once... COI becomes much more apparent, socking goes away, and it enforces real life responsibility. If you're not willing to be responsible for your edits, or if you think it's dangerous to edit... DON'T.

2. Flagged revisions. Not some watered down scheme like we are going to get "any minute now" but the full deal.
- This takes away a lot of the incentive to vandalize because vandalism won't be visible. That takes away a lot of the work the MMORPG-admin-wannabees do so maybe they could concentrate on writing fancruft er I mean real articles.

3. BLP opt out. If you're not in a dead tree bio, your choice if in or out.
- This takes away a lot of incentive for a lot of things and it's just a good idea.

Some of the other stuff mentioned in the thread is good too but those are my top three.

[TungstenCarbide]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address

There is only so much that can be done with the wiki model. The vast majority of the millions of articles will always be crap. Without recognizing and using expertise, very few article will ever reach a high level.

That being said, there are a few things that could help;

1) I would hire and a systems analyst (for lack of a better term). This person would be empowered to experiment, turn on and tune things like flagged revisions. They would be responsible for analyzing and measuring the results. It's easy to say 'I know all the answers'. It's hard to put one's imagination to the test, with genuine metrics, and honestly see how it matches up with reality.

2) BLP is one of the biggest problems. I'd try some things and see how they work; op-out, revision control, create and empower a sympathetic committee of editors to give personal attention to BLP victims.

3) Veropedia seemed to be a pretty good idea; take good articles and run them through some cleanup filters and publish them - revise as needed. Copy this idea with a Foundation sanctioned side project. Featured articles seems to be a logical place to start. Elect good editors for limited terms to run the project, or maybe FA writers. Pay them with the advertising revenues from the site. If the money is good enough consider hiring expert editors from academia to write an occasional article - they might be willing if their work doesn't get instantly shit on. Consider redirecting the wikipedia articles to the 'releasedpedia' articles, when they exists, with a back link for wiki editing. Nothing like a little competition to shake things up.

This post has been edited by TungstenCarbide: Sat 12th September 2009, 7:37pm

[KD Tries Again]

1. To tackle content issues: Create an Editorial Board (or more like several - one for each subject area) with ultimate authority over all content issues composed of people with real-world expertise and credentials (those on the board might or might not already be Wikipedians).

This is the bottom line, because most of the other problems would vanish if people couldn't come to Wikipedia and create whatever content they like in whatever way they like. In a sense, it would simply spoil the "fun."

But I'd also make explicit that the Board(s) would be the gatekeeper(s) for topics too.

3. To address BLP issues: Create a system with a stable, displayed version of an article (that can not be edited) and a NOINDEXed workspace that can be edited. Periodically update the displayed version from the workspace after an approval process. In other words, a much much stronger version of flagged revisions.

Also a good idea, and worth extending beyond BLP to current news in general.

This post has been edited by KD Tries Again: Sat 12th September 2009, 5:35pm

[Malleus]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

I would purge the servers, sell the IT equipment, sell the domain name "Wikipedia" and take the monies and give to some inner city library in need of real books.

As opposed to fake books do you mean? Wouldn't they be cheaper?

[Emperor]

1) Unban everybody and unprotect all articles.

2) Implement Encyc-style management system where admins are chosen by the God King, and allowed to serve as long as they encourage participation in the content-generation process.

3) Change the software so that only basic wiki code is allowed, so that source pages are human-readable and can be edited by people without advanced computer skills.

[The Wales Hunter]

I find it hard to disagree with the points raised by Hell Freezes Over/SlimVirgin.

I'd also ban minors from editing all but certain areas, and those areas would be altered so a non-minor has to check their edits before they go live, but I know I'm opening a can of worms even suggesting it.

[victim of censorship]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

I would purge the servers, sell the IT equipment, sell the domain name "Wikipedia" and take the monies and give to some inner city library in need of real books.

As opposed to fake books do you mean? Wouldn't they be cheaper?

Wikipedia = fake, false, unreliable, unaccountable.

Book = Real, reliable, true, accountable.

[Alison]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

I can think of waaay more than three, but these are the biggies for me:

• BLP opt-out. No questions. Also, bring in the 'dead tree' standard and raise the bar on BLP notability.
• Flagged revisions, as per Lar above. Real flagged revisions - not some pathetic, half-assed attempt at appeasement.
• Some sort of RL authentication. Tagging real names to editors should make people think twice about what they're saying.

[The Wales Hunter]

• BLP opt-out. No questions. Also, bring in the 'dead tree' standard and raise the bar on BLP notability.
• Flagged revisions, as per Lar above. Real flagged revisions - not some pathetic, half-assed attempt at appeasement.
• Some sort of RL authentication. Tagging real names to editors should make people think twice about what they're saying.

Agreed.

I like to call your second one "forced flag" in that the flagged version would have to be the default version and the only version available to non-logged-on users.

I've previously suggested a BLP-class editor where real-life details have to be lodged.

[Malleus]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

I can think of waaay more than three, but these are the biggies for me:

• BLP opt-out. No questions. Also, bring in the 'dead tree' standard and raise the bar on BLP notability.
• Flagged revisions, as per Lar above. Real flagged revisions - not some pathetic, half-assed attempt at appeasement.
• Some sort of RL authentication. Tagging real names to editors should make people think twice about what they're saying.

Far too sensible Alison, you'd never get a consensus for that. For one thng most of the administrators would be forced to admit to their parents that they'd been wasting their time slapping vandals instead of studying, that's why their grades were so bad.

[No Paste]

1. Flagged revisions

2. Some form of review for administrators (I'm not sure what would work best, but ArbCom is not enough)

3. An official standard for gaining consensus (Once again, I'm not sure of a good standard, but we need something)

[LessHorrid vanU]

I agree with the general BLP opt out proviso, and would extend it to bio articles of the recently dead - if a spouse/partner or child wishes for it to be taken down in the aftermath of a death then that desire should be accommodated. A 12 month moratorium should suffice, and afterward any article would be a flagged revision candidate.

Flagged revisions should be immediately applied to BLP's and bio articles relating to persons active in the 20th century (there is far too much political editing of such articles), all "proper" science related articles, all pre 20th Century historical articles, all pre 20th Century art, music and literature articles, and other "dead wood encyclopedic" articles generally.

Wikipedia articles should be tiered, so that the "fundamentals" would not only be under flagged revisions but under regular scrutiny depending on the sensitivity of the subject (political entities, be they individuals, nations, parties, theories, etc. are too prone to idealogical warring irrespective of encyclopedic worth. BLP's have been touched on. Nationalist subjects, while falling under political perview, are also edited emotionally rather than objectively by pro and anti factions), with a secondary tier (the Arts and Sciences?) being flagged revision subjects but with a greater facility to allow wider editing access, and with popular culture remaining open to all access editing.

***

The aim for all of the above is to raise the standard and thus the perceived value of the project, while allowing those subjects which will continue to attract the greater number of participants to remain open to "instant editing".

[GlassBeadGame]

Wales gives you complete charge, and you have three fundamental reforms that you are allowed to implement.

What would those be? And what are the problems or abuses that each are supposed to address?

1) Abandon a risk management policy based on immunity and accept board level responsibility for editorial policies and practices (this will result in a through reform of BLP issues, among others issues, in a manner that will not be undone at the whim of the next person with "three wishes.")

2) Engage in a thorough board development/strategic planning/capacity building process utilizing the tools and accepted practices of the wider non-profit community. Broaden the representation of the board to include the many stakeholders that are shutout by a insular "community"dominated board. Increase the level and presence of employee/agent actors responsible for assuring community compliance with board created standards, practices and policies.

3) Implement rigorous child protection measures informed by experts and practitioners in child protection.

1. ArbCom elections with a secret ballot and candidates using their real names during the election, not just after being appointed. I'd form a committee of trusted editors tasked with ensuring the offered names are genuine. The outcome of the election alone to determine who gets appointed.

-- It would reduce the likelihood of socking to get on ArbCom (though it wouldn't eliminate it). Would make the process and outcome more professional.

-- Downside: it would discourage people who've edited in contentious areas from standing, because they'd be less likely to want to identify themselves. That would mean fewer people on ArbCom with experience of handling contentious content disputes. However, the current process already favours people who've not been involved in disputes, so I don't think we'd make that aspect any worse, and I think the benefits would outweigh the drawbacks.

2. Either (a) no more BLPs unless the subject is already the subject of a biographical article in a high-quality mainstream publication, preferably an encyclopaedia; or (b) the subject has the right of reply in a pop-up box that's visible over the article. The latter was an idea of Patrick Byrne's.

-- The first option would remove most of our BLP problems, but would leave the project a lot less informative, so I'd be more inclined to go with (b) if we could make it work.

3. Everyone who holds a position (admin, Arb, checkuser, bureaucrat) to hold it for six months of the year only, and for the rest must work as an ordinary editor. The software would automatically give and withdraw access on a six-month rotation, assuming that could be done technically.

-- Would reduce the divide between editors, admins and others. Would prevent (or at least wouldn't facilitate) the formation of an admin class that administers the project but doesn't add content.

Why would anyone listen to what this vile monster has to say about "reform?"