Comments for Wikipedia Co-Founder Calls for Major New Moderation Policy

Wikipedia Co-Founder Calls for Major New Moderation Policy

2009-01-24T18:38:29Z

Hot on the heels of Encyclopedia Britannica's announcement that it is moving to a more open editing system, Wikipedia too seems ready for an about face. Yesterday, the New York Times reported that Wikipedia is considering moving away from its free and open editing system to a method that delays changes from appearing on the site until an authorized user has verified them.

On Thursday Jimmy Wales proposed turning on the system of "flagged revisions" in an attempt to reduce the amount of vandalism on Wikipedia, stating that recent death announcements of Senators Edward Kennedy and Robert Byrd on Wikipedia could have been "100% prevented" by this system.

]]> What are Flagged Revisions?

According to the New York Times, flagged revisions prevent edits of 'new and anonymous' users taking effect until they have been authorized by a 'registered and reliable' user.

While it makes sense that users must be registered to authorize a revision, defining 'reliable' may be a little more difficult. Raoul Nicolas, a Wikipedia contributor since 2005 suggests that all long term contributors should be granted the privilege automatically; if you've contributed more than X edits and been an active participant on Wikipedia for X months, you should have be authorized to approve flagged revisions.

The Potential Problems of Flagged Revisions

Although Nicolas offers a logical solution to determine authority, the question remains just how many users will be deemed trustworthy. According to a study by Aaron Swartz, Jimmy Wales found that "over 50% of all the edits are done by just .7% of the users... 524 people. ... And in fact the most active 2%, which is 1400 people, have done 73.4% of all the edits." While these 1400 users would likely be considered trustworthy, just how many edits can they approve given Wikipedia has more than 150,000 edits daily?

Clearly, there are many questions still to be answered.

The flagged revisions system has been used by the German Wikipedia, although as Wales notes in his proposal, it has at times had approval delays of up to three weeks. "Our version should show very minimal delays (less than one week, hopefully a lot less) because we will be using it on a subset of articles, the boundaries of which can be adjusted over time to manage the backlog," Wales said in his argument; he also recommends a limited test period.

Wales Turns to the Wikipedia Community

Pleading with the Wikimedia Foundation and Wikipedia users, Wales ends his proposal with: "Per the poll of the English Wikipedia community and upon my personal recommendation, please turn on the flagged revisions feature as approved in the poll."

The proposal has led to a heated discussion on the site, you can participate here. While some users oppose the idea calling Wales out for his attempt to use his "~~dictatorial~~ constitutional monarch" powers, and others just rant and rave adding little to the conversation, it appears the majority of users applaud Wales on his call to action, asking only for clearer guidelines.

As Wikimedia Foundation spokesman Jay Walsh told the New York Times, "Implementing this functionality is really a volunteer community decision."

Wales, in an attempt to get the ball rolling has asked users who are opposed to the proposal to suggest alternative solutions, giving them until the 29th January to submit ideas before putting suggestions out to be voted on.

Wikipedia's editors have a knack of quickly updating important (and incidental) news largely due to the openness of the site. While the next month will give us a clue in which direction Wikipedia is heading, we can only hope the speed at which updates occur isn't so time consuming that it degrades the service used by so many.

We'll be looking at the potential impact of this dramatic proposed change in more detail over the next few days, but until then, please let us know your thoughts in the comments.

Comment from Byron on 2009-01-24

2009-01-24T20:30:56Z

Those numbers concerning the # of users who do the preponderance of the edits don't make sense....Wales claims that the most active 1,400 contributors have made 73% of all the edits??? And then in the same paragraph, there are 150,000 edits per day? That means that each of those 1400 users makes 78 edits every single day, without fail...from the beginning!!! That is highly suspect.

Unless I'm missing something. I really wasn't great with algebra, but I did pretty well in Accounting ;-)

Comment from Coleman on 2009-01-24

2009-01-24T20:56:51Z

i'm afraid that their moderation will go too far and they will quash things they just don't like. i know that wikimedia is leftist, and since I'm not, it makes me nervous

Comment from Jmartens on 2009-01-24

2009-01-24T21:05:14Z

The problem I see here is that this change will give the select group of Wikipedia posse members even more power. As a casual user, I have found that Wikipedia is anything but a crowd-sourced encyclopedia. It is run my a 'mafia' of members that consider themselves above all else.

The proposed editing rules will give these people even more power, making Wikipedia less useful, not more.

Comment from Nihiltres on 2009-01-24

2009-01-24T21:07:56Z

Hi, I'm an admin on Wikipedia. There's been a lot of confusion over the whole issue here. The issue is the installation, on the English Wikipedia, of an extension called "FlaggedRevs" (for "flagged revisions") to MediaWiki (the wiki software that Wikipedia uses) that would allow some sort of edit approval process.

Jimmy Wales has been advocating that the extension be enabled so that it can be tested before it goes into either more widespread or more permanent use. This is supported somewhat by a poll for a *trial* of the software, which came out with 60% overall support, which is low for Wikipedia, which usually tries for more along the lines of 75% or more support for big changes like this.

The extension can be configured in multiple ways which would end up placing different restrictions on which edits "go live", that is, are displayed by default for a given article. This has the potential to help prevent vandalism and other inappropriate edits.

Given the low level of consensus on the original poll, the focus is shifting to a more specific proposal for "Flagged protection", using a specific configuration where the approval process will be used only on certain articles, as an alternative to normal page protection or "locking". This new proposal looks like it will get more support, since it actually opens Wikipedia up more. Some articles which anonymous and new users currently cannot edit would be editable by them under this proposal—albeit that those edits would have to be approved, by anyone who would already be able to edit the page under the current system.

The details remain to be worked out, but it looks like this version is the version that has the most chance of being implemented.

Comment from Steven Walling on 2009-01-24

2009-01-24T23:21:01Z

Nihiltres is 100% correct in his characterization of the event so far. If you take a look at Wales' talk page (User talk:Jimbo) you can see that a likely compromise is flagged protection, i.e. just adding flagged revisions as an emergency protection measure to certain articles.

Also, it's important to note that the numbers from Aaron Swartz are from 2006, eons ago in terms of the growth of Wikipedia. The community was extremely different back then, and by May 2007 was larger in active number of editors by hundreds. The core community of very active editors has at least doubled since he did his calculations.

@Jmartens:

You're right in one sense: Wikipedia has never been a crowdsourcing experiment like most people think. It's always been tended by a dedicated community of volunteers who do most of the work.

If you take a little time to learn the customs of the site, just like you did for Twitter, then you too can be a part of that community. But you can't just barge in ignorant of how the site works and expect things to go well.

Comment from Henway on 2009-01-24

2009-01-24T23:49:22Z

This is great.

So, um.. how is Wikipedia gonna make money again?

Comment from Lidija Davis on 2009-01-24

2009-01-25T01:25:13Z

@Byron - you know, the numbers don't surprise me. 78 doesn't seem to be too many for very invested contributors. I guess it would also depend on the type of edits.

@Coleman - It will be interesting to watch. They've done a pretty good job so far, so I'm crossing my fingers they continue on that path.

@Jmartens - Yeah, it might seem that way. When contributing only occasionally, we find it difficult to understand what the users working on the site day in day out do. But my take is that it is a privilege that those who have been committed deserve - much like on any other UGC site.

@Nihiltres - Thank you for stopping by and clearing it up for us. As I mentioned in the post, ReadWriteWeb will be analyzing this further in the coming days and your input is valuable and much appreciated.

@Steve - yes, the Swartz information is a couple of years old, but I'm hoping the next few weeks will give us access to newer data. It will be interesting. And yes, I absolutely agree that you need to get to know the community of a site before jumping in.

@Henway - good question

Comment from Jmartens on 2009-01-24

2009-01-25T06:56:05Z

@Steve Great explanation and good points. I just wish everyone would communicate it like you just did. Unfortunately, what we hear in the media is that a wiki is a site that anyone can edit. That's about as true as saying anyone can contribute to Nike's next shoe design. Sure, anyone can but in order to make it happen you have to become part of a very select and exclusive community. And really, there is nothing wrong with that.

Just seems like the proposal makes the community even more exclusive, not more open.

Comment from kgyst.myopenid.com on 2009-01-25

2009-01-25T17:11:41Z

I am a very active contributor of the Hungarian Wikipedia (within the most active 20) and an occasional contributor of the English (approximately 1000-2000 edits, ~5 new articles and templates)

The daily 70 edits seem to be quite normal. In my most active periods I was over 100, and an average active moth has 1000 edits (ie. daily 30)

As an occasional contributor at the English WP I have not experienced any kind of elitism and mafia-like behavior. I think in a good article inexperienced users' contributions simply don't meet the quality of the article (in most cases these are the most viewed, best quality articles that the beginners see for the first time) and thus are deleted or re-edited. They think this is a mafia-like behavior, but in fact a simple quality management.
Finally, a document editable by anybody is re-editable for anybody, also.

The flagged revisioning, that exists in the German (and also in the Hungarian, we love the bureaucracy) WP is a catastrophe, since it removes being up to date, one of WP' main advances. The working system is maximally elitist, since while advanced users have the old method without obstacles, the beginners, who need the most positive feedback and experience, get nothing about their edits.

A flagged protection could be the solution for this problem, leaving new artices to be edited frrely.

Comment from usd6 on 2009-01-25

2009-01-26T01:23:40Z

Wikipedia The move is good or bad? Only time left to prove

Thank you for your articles, written in very good!

Comment from BanyanTree on 2009-01-26

2009-01-26T12:05:24Z

@Henway - Wikipedia is a project of the Wikimedia Foundation, a non-profit charity. It is not trying to make money. However, it does have operating costs. In the recently concluded fundraising drive the Foundation met its goal of $6 million ahead of schedule. Of this, $2 million was from foundations or major gifts and $4 million was in small amounts donated by 125,000 people. It appears that readers are willing to donate a small amount if they feel that Wikipedia is offering them a valuable service.

The flagged revision proposal has been pushed by those who are trying to reduce the incidence of seen "bad" article revisions, which are thought to reduce the credibility of the site in the eyes of the public. However, since many of the most active editors remember originally joining the site after realizing that their edits went live immediately, which is incredibly exciting in an age when everything is moderated, people are raising the specter that flagged revisions will discourage new users from joining, potentially reducing the rate at which new active members join to below the replacement rate to take over from the constant stream of older users who either burn out or just move on, which would be disastrous to the long term health of the community and site. One can understand why this issue of whether and how to implement is causing drama in the community.