Wikipedia's administrative system: An MMORPG rotten to the core
The past week or so has unintendedly resulted in a clash between me and what I call the administrative system of Wikipedia, and I couldn't help but notice a number of flaws of process. There is no cabal - But there are factions on Wikipedia. A faction is any number of individuals who collaborate in order to achieve a common goal. The administrative system increasingly shows the sign of such a faction.
In the middle ages, noblemen were to be tried by their fellow noblemen for their crimes, which resulted in hardly anyone ever getting convicted. Similiarly, the clerical system had jurisdiction over itself - and even gross sexual abuse often went on for years without anyone intervening - if at all.
On Wikipedia, administrators are given authority and technical provisions to punish disruptive users. Some SysOps would point out that the purpose of a block is not to punish, but to prevent - But since a block is not an effective prevention, as it can be evaded at the leisure of any somewhat tech-savy user, it can only serve as punishment and to get rid of unwelcome or different POV editors.
If blocked, in error or abuse, or even legitimately, an editor has basically no chance to revoke the block. He can no longer edit the relevant pages such as the Administrator's Noticeboard for Intervention. Thus, even though the policy explicitely and uncompromisingly forbids abusing blocked or disruptive users, regardless of their past, an user, once blocked, is realistically considered free game for all sorts of false accusations, abuses, personal attacks and libel.
Because of the factionalizing mechanics of the administrative system, a blocked user has basically no chance to get an abusive ban lifted. The policy quite clearly states that the originally blocked user is meant to lift blocks he invoked himself. An abusive sysop will simply not do so. And nobody else will. WikiEN-l is a mailing list mostly read by the administrative system itself, who again, naturally, almost always will side with the administrator they know rather than a random editor.
RFComments don't work, neither do RFMediations - especially when blocked. The RfArbitration system is limited by the number of cases the ArbCom can deal with in any given timeframe, and is made a mockery of justice when a blocked user is denied the ability to defend himself. And because the ArbCom members are themselves also part of the administrative system, they again tend side with problematic or long standing administrators much more easily than with problematic or new editors.
If the wrongly blocked user then decides to ignore the block, and proceeds with the trivial technical task of evading the block, in order to protect himself from false accusations, personal attacks, or - dear lord - continuing to edit mainspace articles or talk pages, this is actually seen as an invitation to administrators to resort to personal attacks, vandalize the user at will - more even than before he becomes "free game", easily resulting in blocks lasting to the end of time, literally forcing an user that wants to contribute to Wikipedia to evade or leave the project.
Ignorance and indifference to policy are normal in today's administrative system of Wikipedia. Completely unfair and unwarranted blocks occur on a daily basis, personal attacks, vandalizations of the userpages of unpopular editors - a strong pecking order. The policies are ignorable, the only thing that counts is "Don't piss off the sysops that have real power".
All of this also undermines the basic content policies of Wikipedia. Adherents of a different POV can be removed easily (although no administrator would ever be caught dead with banning an user on the basis of POV). Not to mention it removes strong editors, and replaces them with weak ones compliant to the POV of increasingly sociopathic administrators having an interest in the article.
Currently Wikipedia is too much of an MMORPG: If one editor has 500 edits, and another has 20'000, and they have a dispute, the editor with 20'000 edits always wins - especially so when he is a sysop himself, the chances for which increase almost exponentially with the number of posts. This is not a way to find consensus: Consensus must be found in debate. And debate is the hardest of all ways to fight an opposite POV - It is almost always the more successful strategy to get the opposite side blocked than to actually get down and find this sacred consensus (or compromise).
Why are administrators needed in the first place? "To combat vandalism" would most answer. But interestingly, popups, applications, bots, etc are all much more effective than the administrative power to block and to protect an article - These are either trivially easy circumvented by a determined vandal, or undermine the most basic idea of Wikipedia: That everyone can edit it - the very idea that made Wikipedia so popular, and what it is today. Regular editors can patrol RecentChanges just as much as administrators can. The only real necessity for privileged users is the management of the front page - And it doesn't take thousands of administrators to do that.
The solution to the system is to simply remove all administrative privileges, short of maybe a dozen "hard core" people needed to manage the front page, and instructed not to use their powers for anything in the mainspace. For everything else, regular users are just as well equipped for dealing with vandalism and other disruptive occurances as administrators. And for NPOV: It does not ever serve the neutrality of an article if one party is equipped with the power to block the other party.
Would the removal of all administrators result in more edit warring? Yes of course. Because no longer would one party simply get the opposite party blocked. But edit warring is not effective, it is indeed mutually assured destruction: If two parties simply revert each other all the time, they would never get anywhere - Ultimately even the most stubborn POV warrior realizes he will have to arrange with the other side. Personal attacks? Don't listen to it and grow a skin. Copyvios? Give regular users the ability to remove pictures. Delete and undelete? Merely a technical problem.
Everything is better than compromising the sanity and integrity of the "Encyclopedia" Wikipedia.
On 6/17/06, Dabljuh samw@student.ethz.ch wrote:
If blocked, in error or abuse, or even legitimately, an editor has basically no chance to >revoke the block. He can no longer edit the relevant pages such as the Administrator's >Noticeboard for Intervention. Thus, even though the policy explicitely and >uncompromisingly forbids abusing blocked or disruptive users, regardless of their past, an >user, once blocked, is realistically considered free game for all sorts of false accusations, >abuses, personal attacks and libel.
That simply isn't the case
Because of the factionalizing mechanics of the administrative system, a blocked user >has basically no chance to get an abusive ban lifted. The policy quite clearly states that >the originally blocked user is meant to lift blocks he invoked himself. An abusive sysop >will simply not do so. And nobody else will. WikiEN-l is a mailing list mostly read by the >administrative system itself, who again, naturally, almost always will side with the >administrator they know rather than a random editor.
In most cases the block will simply expire.
If the wrongly blocked user then decides to ignore the block, and proceeds with the trivial >technical task of evading the block, in order to protect himself from false accusations, >personal attacks, or - dear lord - continuing to edit mainspace articles or talk pages, this >is actually seen as an invitation to administrators to resort to personal attacks, vandalize >the user at will - more even than before he becomes "free game", easily resulting in >blocks lasting to the end of time, literally forcing an user that wants to contribute to >Wikipedia to evade or leave the project.
Whixh is why the sane option is the wait out the block
Ignorance and indifference to policy are normal in today's administrative system of >Wikipedia.
Not true.
Completely unfair and unwarranted blocks occur on a daily basis,
personal attacks, >vandalizations of the userpages of unpopular editors - a strong pecking order.
Not true
The policies are ignorable, the only thing that counts is "Don't piss
off the sysops that >have real power".
If you manage to piss off sysops (multiple) it suggests you are doing something wrong
All of this also undermines the basic content policies of Wikipedia. Adherents of a >different POV can be removed easily (although no administrator would ever be caught >dead with banning an user on the basis of POV). Not to mention it removes strong editors, >and replaces them with weak ones compliant to the POV of increasingly sociopathic >administrators having an interest in the article.
There is a faint posibilty that argument might hold up if all admins had the same POV. They do not.
Currently Wikipedia is too much of an MMORPG: If one editor has 500 edits, and another >has 20'000, and they have a dispute, the editor with 20'000 edits always wins - especially >so when he is a sysop himself, the chances for which increase almost exponentially with >the number of posts.
That would be because they are generaly right.
This is not a way to find consensus: Consensus must be found in
debate. And debate is >the hardest of all ways to fight an opposite POV - It is almost always the more successful >strategy to get the opposite side blocked than to actually get down and find this sacred
consensus (or compromise).
There is no compromise on NPOV.
Why are administrators needed in the first place? "To combat vandalism" would most >answer.
Dealing with copyvios, stuff in the mediawiki namespace, dealing with the results of XFDs and various other ods and ends.
But interestingly, popups, applications, bots, etc are all much more
effective than the >administrative power to block and to protect an article
False. If I block you you have to mess around chnageing IP. If I use a range block you have to get a new ISP.
If I hard protect there is nothing a vandle can do.
These are either trivially easy circumvented by a determined vandal,
So are bots. In fact it would be generaly posible to know out even determided vandels but it is rarely worth the extra effort to do more than repeatedly spaping them down.
or undermine the most basic idea of Wikipedia: That everyone can edit
it - the very idea >that made Wikipedia so popular, and what it is today.
Take away blocks and you very soon find that no one can edit. Ever hear of spambots?
Regular editors can patrol RecentChanges just as much as administrators can.
They can only revet. They can't stop vandles.
The only real necessity for privileged users is the management of the
front page - And it >doesn't take thousands of administrators to do that.
So you support the protection of the front page. So much for your anyone can edit stance.
The solution to the system is to simply remove all administrative privileges, short of >maybe a dozen "hard core" people needed to manage the front page, and instructed not >to use their powers for anything in the mainspace.
The main page is in the main namespace.
Would the removal of all administrators result in more edit warring? Yes of course. >Because no longer would one party simply get the opposite party blocked. But edit >warring is not effective, it is indeed mutually assured destruction:
I'm failing to see any destruction
If two parties simply revert each other all the time, they would
never get anywhere - >Ultimately even the most stubborn POV warrior realizes he will have to arrange with the >other side.
Why? If the other side is older than him he just has to wait.
Personal attacks? Don't listen to it and grow a skin. Copyvios? Give regular users the >ability to remove pictures.
Um no. how do we prevent them from removeing pictures we don't want them to remove?
Delete and undelete? Merely a technical problem.
huh?
Please take the problems a bit more serious.
They are there - some things may be in fact different, but outright denying the very possibility that there could be problems, like you do, denies you the ability to look for solutions and instate change before it is too late.
The revocation of all admin privileges (and a modification of user privileges to ensure the continued functionality of Wikipedia) is certainly not a popular option, although I would say it would work. Very differently from how it does now, but ultimately better because it would more resemble a scholastic system of peers, rather than a game of block-a-mole - or so I claim.
If you have better options to remedy perceived problems, please bring them forward. But outright denial does not help.
On 6/17/06, Dabljuh samw@student.ethz.ch wrote:
Wikipedia's administrative system: An MMORPG rotten to the core
I assume you are referring to this? http://en.wikipedia.org/wiki/Wikipedia:Requests_for_arbitration#Dabljuh Which is interesting reading.
Garion
On Sat, 17 Jun 2006 23:04:20 +0200 Garion1000 garion1000@gmail.com wrote:
I assume you are referring to this? http://en.wikipedia.org/wiki/Wikipedia:Requests_for_arbitration#Dabljuh Which is interesting reading.
Garion
And that is exactly what I am talking about. You reduce the argument to "who" is talking, not on the argument itself. This sort of ad hominem is not just typical, it is downright the "Wikipedia" way of thinking these days.
On 6/18/06, Dabljuh dabljuh@gmx.net wrote:
On Sat, 17 Jun 2006 23:04:20 +0200 Garion1000 garion1000@gmail.com wrote:
I assume you are referring to this? http://en.wikipedia.org/wiki/Wikipedia:Requests_for_arbitration#Dabljuh Which is interesting reading.
Garion
And that is exactly what I am talking about. You reduce the argument to "who" is talking, not on the argument itself. This sort of ad hominem is not just typical, it is downright the "Wikipedia" way of thinking these days.
Maybe, but when someone I never saw before on the mailing list (or I forgot) does a long e-mail like this on the mailing list. And when that user very recently had a RFAR filed against him and is now banned indefinitely. It's very hard to take that user or the e-mail seriously. Especially when you read the case and see the diffs. That btw has got nothng to do with editcount.
Garion
Maybe, but when someone I never saw before on the mailing list (or I forgot) does a long e-mail like this on the mailing list. And when that user very recently had a RFAR filed against him and is now banned indefinitely. It's very hard to take that user or the e-mail seriously. Especially when you read the case and see the diffs. That btw has got nothng to do with editcount.
Who, do you think, would be most suitable at pointing out flaws in Wikipedia? Someone who was confronted with a number of them in rapid succession, or someone who is part of the problem? Of course my vision could well be distorted by my experience. But during all of this I have gotten the support of various other editors, who were equally disappointed with "how stuff works" here. I can assure you, the entire procedure that I have been victim of has been highly controversial, and that is not simply my personal, disgruntled opinion.
Read my other post on edit count vs "MMORPG" aspects.
Wow...that was a pointless read. You can claim anything you want, but you have to back it up with real cases.
mboverload
On 6/17/06, Dabljuh samw@student.ethz.ch wrote:
Currently Wikipedia is too much of an MMORPG: If one editor has 500 edits, and another has 20'000, and they have a dispute, the editor with 20'000 edits always wins - especially so when he is a sysop himself, the chances for which increase almost exponentially with the number of posts. This is not a way to find consensus: Consensus must be found in debate. And debate is the hardest of all ways to fight an opposite POV - It is almost always the more successful strategy to get the opposite side blocked than to actually get down and find this sacred consensus (or compromise).
I just wanted to address this paragraph in particular.
Edit counts mean absolutely nothing in day-to-day interactions except when a user is very very very new (i.e., they have less than 50 edits or so). I think it is a safe generalization to say that very new users are indeed not often valued as much as long-time contributors. There are sane reasons for that, even if we do want to say that new users are always welcome.
If one could actually correlate edit counts to success on the Wiki — something for which I imagine there is a high possibility, even if it has not here been proven — it is likely because a user with a large number of edits has invested more time here, understands how things work a bit better, and understands what sorts of arguments do and do not register. (New and outraged users usually think that decrying "censorship" is a useful strategy, for some reason; in reality, it just makes them look like total nutballs who can't construct a convincing argument.) It is also the case that users who stick around and edit a lot are likely to be noticed and recognized by others as good editors, and others are more likely to be sympathetic to them.
It is not the case the edit counts, as raw numbers, mean much. I do not know my own edit count nor anyone else's who has been around enough for me to notice them positively. I suspect others' experiences are similar. The only place where I have ever seen quantitative edit counts discussed and compared is the Request for Adminship page, where it is used as an acknowledged rough metric for editing experience.
Just my two cents. I think there is a minority of Wikipedia admins who actually do make things miserable for others in various ways, and I think it is very possible that certain variations of "groupthink" do end up having unpleasant effects. (The neverending search for "potentially offensive usernames" is a ridiculous and current example of this, in my opinion. If it is not offensive enough to generate numerous recommendations to change it, it is not offensive enough for me to give a damn, personally.)
But beyond those feelings, I do not think that Wikipedia is "rotten to the core", and I do not think that most administrators abuse their powers in content disputes, which seems to be the chief allegation in this screed.
FF
On 6/17/06, Fastfission fastfission@gmail.com wrote:
If one could actually correlate edit counts to success on the Wiki — something for which I imagine there is a high possibility, even if it has not here been proven — it is likely because a user with a large number of edits has invested more time here, understands how things work a bit better, and understands what sorts of arguments do and do not register. (New and outraged users usually think that decrying "censorship" is a useful strategy, for some reason; in reality, it just makes them look like total nutballs who can't construct a convincing argument.) It is also the case that users who stick around and edit a lot are likely to be noticed and recognized by others as good editors, and others are more likely to be sympathetic to them.
There is also the issue that someone with 20,000 would have rather a large stake in the project.
On Jun 17, 2006, at 2:49 PM, geni wrote:
There is also the issue that someone with 20,000 would have rather a large stake in the project. --
Being one that "broke" 20,000 edits early this month, I would argue that the more you edit and contribute to the project, the more you appreciate its value, and the more your tolerance for abuse (read personal attacks, circumventing blocks, etc.) diminishes. That is a good thing. If some users (usually those advocating a specific POV such as Mr. Dabljuh) find this intolerance too much to bear, he may be better starting a blog to promote his POV, rather that try to change the system/policy/community consensus to fit his views.
-- Jossi
When I speak of edit count, this is metaphorical. I do not mean so much the actual edit count, but more so a hypothetical value, to force MMORPG terminology on this, we could call it "experience points", XP, that are probably better described as the number of people in your social network, as well as their individual "XPs". And admin privileges are just (massive) bonus points, that by themselves already require a large amount of "XP".
Consider the process of becoming an administrator: You simply need a lot of people vouching for you, and other people will follow the lead. People who you have fought in edit wars will seldomly vouch for you, but instead call against this. This could be described as "negative" XP. So a good strategy (for becoming an administrator) may be to keep out of controversial articles and instead make many small and large improvements to various articles while building up this social network. People don't become sysop or anything for quality work, but for networking.
Of course trying to display the administrative System as a single faction would be false. It is much more cliquish, but every admin has his network. Once gained adminship, one is relatively free to abuse people at controversial articles - regardless of scholastic merit, since there is the clique to protect him and decry anyone attacking him.
Consider barnstars - by themselves a nice gesture and so on, it can be interpreted as scratching each other's back, making each other look more well respected. This concept of "Wikilove" has become increasingly creepy to me over the years, to be honest. Eventually the administrators turn out to be an old boys network scratching each other's backs every opportunity by inflating each other's apparent XPs.
And not disciplining administrative abuses is just obvious - anything else would award both parties massive negative XP.
Compared to these XP, "being a great writer" or "being a good scholar" regarding a topic completely fade when it comes to a content dispute.
On 6/17/06, Dabljuh dabljuh@gmx.net wrote:
Consider the process of becoming an administrator: You simply need a lot of people vouching for you, and other people will follow the lead. People who you have fought in edit wars will seldomly vouch for you, but instead call against this. This could be described as "negative" XP. So a good strategy (for becoming an administrator) may be to keep out of controversial articles and instead make many small and large improvements to various articles while building up this social network. People don't become sysop or anything for quality work, but for networking.
Not really. They become an admin for being around for a while and not doing anything people regard as a significant negative.
Of course trying to display the administrative System as a single faction would be false. It is much more cliquish, but every admin has his network. Once gained adminship, one is relatively free to abuse people at controversial articles - regardless of scholastic merit, since there is the clique to protect him and decry anyone attacking him.
Considering the number of admins I've blocked at various times that seems unlikely.
Consider barnstars - by themselves a nice gesture and so on, it can be interpreted as scratching each other's back, making each other look more well respected. This concept of "Wikilove" has become increasingly creepy to me over the years, to be honest.
Nothing saying you have to have barnstars on your userpage. I don't.
Eventually the administrators turn out to be an old boys network scratching each other's backs every opportunity by inflating each other's apparent XPs.
Considering the average age of wikipedians that seems unlikely
Compared to these XP, "being a great writer" or "being a good scholar" regarding a topic completely fade when it comes to a content dispute.
False. You do however have to demostrate these abilities and go on demonstrateing them every day.
On 6/17/06, Dabljuh samw@student.ethz.ch wrote: ....
Currently Wikipedia is too much of an MMORPG: If one editor has 500 edits, and another has 20'000, and they have a dispute, the editor with 20'000 edits always wins - especially so when he is a sysop himself, the chances for which increase almost exponentially with the number of posts. This is not a way to find consensus: Consensus must be found in debate. And debate is the hardest of all ways to fight an opposite POV - It is almost always the more successful strategy to get the opposite side blocked than to actually get down and find this sacred consensus (or compromise).
....
Actually, this isn't true. One of the more interesting results of various RFA statistical analyses is that chances of being nominated for / passing RFA increase polynomically (ie, greater than merely linear, but almost certainly not an exponential) with edit count up until about 3/4000, at which point it increase linearly and at some point around twice the first threshold begins to fall.
~maru