Is something fundamentally wrong with Wikipedia governance processes?
The Wikipedia is often hailed as a prime example of peer production and peer governance, an example of how a community can self-govern very complex processes. Including by me.
But it is also increasingly showing the dark side and pitfalls of purely informal approaches, especially when they scale.
Wikipedia is particularly vulnerable because its work is not done in teams, but by individuals with rather weak links. At the same time it is also a very complex project, with consolidating social norms and rules, and with an elite that knows them, vs. many occasional page writers who are ignorant of them. When that system then instaures a scarcity rule, articles have to be ‘notable” or they can be deleted. It creates a serious imbalance.
While the Wikipedia remains a remarkable achievement, and escapes any easy characterization of its qualities because of its sheer vastness, there must indeed be hundreds of thousands of volunteers doing good work on articles, it has also created a power structure, but it is largely invisible, opaque, and therefore particularly vulnerable to the well-known tyranny of structurelessness.
Consider the orginal thoughts of Jo Freeman:
“Contrary to what we would like to believe, there is no such thing as a ’structureless’ group. Any group of people of whatever nature coming together for any length of time, for any purpose, will inevitably structure itself in some fashion. The structure may be flexible, it may vary over time, it may evenly or unevenly distribute tasks, power and resources over the members of the group. But it will be formed regardless of the abilities, personalities and intentions of the people involved. The very fact that we are individuals with different talents, predispositions and backgrounds makes this inevitable. Only if we refused to relate or interact on any basis whatsoever could we approximate ’structurelessness’ and that is not the nature of a human group.
Consider also this warning:
Every group of people with an unusual goal - good, bad, or silly - will trend toward the cult attractor unless they make a constant effort to resist it. You can keep your house cooler than the outdoors, but you have to run the air conditioner constantly, and as soon as you turn off the electricity - give up the fight against entropy - things will go back to “normal”.
In the same sense that every thermal differential wants to equalize itself, and every computer program wants to become a collection of ad-hoc patches, every Cause wants to be a cult. It’s a high-entropy state into which the system trends, an attractor in human psychology.
Cultishness is quantitative, not qualitative. The question is not “Cultish, yes or no?” but “How much cultishness and where?”
The Wikicult website asserts that this stage has already been reached:
With the systems, policies, procedures, committees, councils, processes and appointed authorities that run Wikipedia, a lot of intrinsic power goes around. While most serious contributors devotedly continue to contribute to the implied idealism, there are those with the communication and political skill to place themselves in the right place at the right time and establish even more apparent power. Out of these, a cabal inevitably forms; the rest, as they say, is history.
Specialized sites have sprung up, such as the Wikipedia Review, monitoring power abuse in general, or in particular cases
The Wikipedia Review offers an interesting summary of the various criticisms that have been leveled agains the Wikipedia, which I’m reproducing here below, but I’m adding links that document these processes as well. Spend some time on reading the allegations, their documentation, and make up your own mind.
My conclusion though is that major reforms will be needed to insure the Wikipedia governance is democratic and remains so.
1. Wikipedia disrespects and disregards scholars, experts, scientists, and others with special knowledge.
“Wikipedia specifically disregards authors with special knowledge, expertise, or credentials. There is no way for a real scholar to distinguish himself or herself from a random anonymous editor merely claiming scholarly credentials, and thus no claim of credentials is typically believed. Even when credentials are accepted, Wikipedia affords no special regard for expert editors contributing in their fields. This has driven most expert editors away from editing Wikipedia in their fields. Similarly, Wikipedia implements no controls that distinguish mature and educated editors from immature and uneducated ones.”
Critique of Wikipedia’s open source ideology, as opposed to free software principles
2. Wikipedia’s culture of anonymous editing and administration results in a lack of responsible authorship and management.
“Wikipedia editors may contribute as IP addresses, or as an ever-changing set of pseudonyms. There is thus no way of determining conflicts of interest, canvassing, or other misbehaviour in article editing. Wikipedia’s adminsitrators are similarly anonymous, shielding them from scrutiny for their actions. They additionally can hide the history of their editing (or that of others).”
3. Wikipedia’s administrators have become an entrenched and over-powerful elite, unresponsive and harmful to authors and contributors.
“Without meaningful checks and balances on administrators, administrative abuse is the norm, rather than the exception, with blocks and bans being enforced by fiat and whim, rather than in implementation of policy. Many well-meaning editors have been banned simply on suspicion of being previously banned users, without any transgression, while others have been banned for disagreeing with a powerful admin’s editorial point of view. There is no clear-cut code of ethics for administrators, no truly independent process leading to blocks and bans, no process for appeal that is not corrupted by the imbalance of power between admin and blocked editor, and no process by which administrators are reviewed regularly for misbehaviour.”
The blog Nonbovine ruminations critically monitors Wikipedia governance
The Wikipedia has stopped growing because of the deletionists: Andrew Lih ; Slate
Wikipedia’s abusive bio-deletion process: case by Tony Judge
4. Wikipedia’s numerous policies and procedures are not enforced equally on the community, popular or powerful editors are often exempted.
“Administrators, in particular, and former administrators, are frequently allowed to trangress (or change!) Wikipedia’s numerous policies, such as those prohibiting personal attacks, prohibiting the release of personal information about editors, and those prohibiting collusion in editing.”
The undemocratic practices of its investigative committee
The badsites list of censored sites belonging to Wikipedia’s enemies
Lack of transparency and accountability
The Judd Bagley case
InformationLiberation on Wikipedia’s totalitarian universe
5. Wikipedia’s quasi-judicial body, the Arbitration Committee (ArbCom) is at best incompetent and at worst corrupt.
“ArbCom holds secret proceedings, refuses to be bound by precedent, operates on non-existant or unwritten rules, and does not allow equal access to all editors. It will reject cases that threaten to undermine the Wikipedia status quo or that would expose powerful administrators to sanction, and will move slowly or not at all (in public) on cases it is discussing in private.”
Monitoring of ArbCom’s activities
The case of the secret mailing list for top insiders
6. The Wikimedia Foundation (WMF), the organization legally responsible for Wikipedia, is opaque, is poorly managed, and is insufficiently independent from Wikipedia’s remaining founder and his business interests.
“The WMF lacks a mechanism to address the concerns of outsiders, resulting in an insular and socially irresponsible internal culture. Because of inadequate oversight and supervision, Wikimedia has hired incompetent and (in at least one case) criminal employees. Jimmy Wales for-profit business Wikia benefits in numerous ways from its association with the non-profit Wikipedia.”
Wikimedia chairwoman rejects demand for transparency
Review of the conflict of interest issue
January 7th, 2008 at 10:22 am
Good work Michel
This is a very good piece. The last point that Wikipedia’s founder is running a related for-profit business shouldn’t be given too much weight, it is kind of common with many Open Source projects. (But, as your later research (in email) shows, it may be also this aspect has gone too far and there is not a clear separation of interests between Wikia and Wikimedia Foundation.)
Apart from that the compilation of criticisms was eyeopening. I read in particular the case of http://antisocialmedia.net/. If any of the other links support this is true, it seems the Wikipedia
administration should be ousted in a popular revolt - or a fork (although brand value of Wikipedia is so high the cost of forking is almost prohibitive). [Now that I’ve thought about it overnight, maybe a large and influential enough group of people shoul approach Jimmy Wales on the issue, he should be able to understand problems if pointed out to him and it would be the most civilised way of resolving this.]
It seems my previous thoughts on Wikipedia administrators simply misunderstanding the situation and opportunities of an online medium was way too positive. I now believe Wikipedia has in fact been taken over by a bunch of incompetent fools, who are drunk on the power of administering facts on the worlds now most popular source of information.
This reminds me of what easily happens with large mailing-lists. An online medium runs the risk of becoming dominated by those who have most time to spend - such as people who are unemployed because of mental problems (this was literally the case on one list I used to frequent, I’m not kidding - of course some of the others were just weird types without a social life and therefore spending too much time creating controversy on the list) - and therefore can afford spending so much time producing nonsense that all the sane people (with day jobs) just have to surrender and go away.
One would certainly hope something is done to correct this problem, Wikipedia is too valuable to suffer such a fate. Let’s hope a group of more clueful and responsible people (The Wikipedia Review?) are willing to spend their scarce time in the battle that seems to be necessary.
January 7th, 2008 at 9:47 pm
This is a really good, measured piece with appropriate, interesting links. Well done.
Wikipedia failed because in 2005-2006, when the site suddenly lurched into the public’s consciousness, the founding fathers didn’t take appropriate measures to ensure a responsible process to cope. People like Jimmy Wales, entrepreneurs who clearly weren’t making enough money off it to be satisfied, ceased to care enough about the site itself. Opting for the glamour of “the myth” of “free collaboration” and parading around international conferences talking turkey.
Meanwhile, by mid 2006, with co-founder Larry Sanger disillusioned and seeing a grim future, the site had become infiltrated, then controlled by unaccountables who had identified Wikipedia’s status at the top of google. These newcomers understood the power that provides. The bearded idealists and the dedicated encylopedists of the early days were gradually crowded out, or forced out by serious game-players.
As the site lurched from crisis to crisis (John Siegenthaler / Essjay etc) it became apparent just how irresponsible Wikipedia was prepared to be. From top to bottom. The latest COO scandal is another example of this.
Making the rest of ask: How can “the sum of human knowledge”, which now dominates the internet and has crept into education itself, be trusted to a community that so consistently fails to show due responsibility?
January 8th, 2008 at 12:12 am
In my work on Inquiry Driven Systems it eventually became necessary to examine the blocks to inquiry that always seem to arise just as soon as any significant inquiry gets going.
A critical turning point occurs when a system equipped with an Interpretive Framework (IF) is able to reflect on its own IF to the degree that it develops a Reflective Interpretive Framework (RIF).
But if a community of inquiry obstructs or prohibits the requisite degree of critical reflection then it finds itself headed for the cul-de-sac of cult behavior.
Here is a pertinent passage from one of my working papers on the subject:
In the case of Wikipedia, we are dealing with a system that falls under the description of a Deliberately Unreflective Framework (DUF), that is, an IF in which “the rules of the IF in question forbid the act of reflecting on its form”.
January 8th, 2008 at 12:15 am
There’s a missing tag in that last post.
January 8th, 2008 at 9:26 am
For reference, below is what I wrote previously on the p2presearch mailing list on the issue of deletionism. As I wrote above, I’m now convinced that the problem is much more severe than a simple misunderstanding. (Though incompetence certainly plays a part in any case.)
**
Is the issue of creating “editions” of Wikipedia often discussed? I mean the whole question of deletionism seems to me to be a misunderstanding of the opportunities created by an internet
wikipedia. The problem deletionists try to solve is that of quality/trustability, they try to delete all such material that would not normally make it into a traditional encyclopedia, or at least material which can be suspected to be wrong or just poorly written. Yet, the great thing about wikipedia (or the internet in general) is that it is not a traditional encyclopedia - it can contain an infinite amount of information (why not have bio’s of all of us there?) and be updated fast, rather than through a slow perfectionist process.
It seems that the antagonists in the deletionist vs inclusionist debate have forgotten that they are dealing with an infinite medium, where all solutions ( -> forks) may co-exist. The sensible thing to do would be to have one “source” Wikipedia, which would strive for maximum inclusionism, and then have editions which strive for a certain treshold of quality, certain topics etc… (And there are mechanisms that can be implemented to make sure the original source is still efficiently re-used, the editions would be subsets of the inclusionist full wikipedia.)
This would be similar to how Linux distributions work: Sourceforge and the internet in general will create an endlees supply of Open Source Software, and distributions are there to filter out the true gems for easy access to the greater public. Instead now the different camps in wikipedia seem to have gotten stuck on the idea of a one true wikipedia, and battling over how that should be governed.
January 8th, 2008 at 10:22 am
Uhoh, this is really grave. I read yesterday the mail thread starting with Wikimedia chairwoman rejects demand for transparency. I thought, surely that isn’t so, this is some out of context quote by people who just want to make baseless criticism on one of our favorite projects. I was stunned to really read what the WMF chairwoman has to say…
So it is true, Wikipedia and its foundation are led by people who have completely abandoned the core ideals that brought it to its current glory.
Reading the thread further one can see a couple comments to the sentiment of
…but I fear that those refer more to the accident of appointing a convicted criminal as COO, rather than any possible shortcomings of the remaining Wikipedia power holders. So I’m not too optimstic that a sudden change towards the positive is already taking place.
However, Florence Devouard, the current WMF chairwoman has decided to not run for another term for the office and thus the search is on for a new chairperson. I think this is a good time to ask: What can we do to save Wikipedia?
January 8th, 2008 at 4:26 pm
Michel et al.,
Just from my cursory scans of these pages, it looks like the P2P group has a lot of solid sociotech concepts to offer but that it is just a bit shy of grasping the Realpolitik that determines the real dynamics of Wikipedia.
I invite the memebers of the P2P group to initiate threads at The Wikipedia Review where these isues might be more lesiurely discussed.
January 12th, 2008 at 6:45 pm
One of the “services” that Wikipedia “provides” is to show us everything that can wrong with a Peer Ideal Project (PIP). I think that it’s fair to say that anything that can go wrong with a PIP has already gone wrong or will eventually go wrong at Wikipedia.
The uses of adversity are quite well known to the experimental mind, and information about worst case scenarios is extremely useful to those who know how to learn and adapt. Sadly, all too sadly, there is no critical mass of experimental minds at Wikipedia who are capable of reflecting on the symptoms of their current dysfunctionality and who, at the same time, have the power to correct their increasingly deflected course.
NB. In the interests of encouraging wider dialogue, I am dual-posting my comments on this thread at The Wikipedia Review.
January 13th, 2008 at 5:45 pm
I wrote about this on my blog too: http://openlife.cc/node/194
January 13th, 2008 at 6:34 pm
Without makeing a judgement about issues, named above one must also see that wikipedia has powerfull enemies. Beside an army of small contributors unhappy, when their wrong articles have been rejected or corrected there are vast corporate interests, which would like wikipedia to support them like pharma industry, food industry, etc. Last but not least the are numerous regimes still in power, which must fear encyclopedical knowledge about their past and today.
Wikipedias’ structure must resist these forces.
January 14th, 2008 at 8:32 am
Dear Jon: concerning your invitations to contribute to the Wikipedia Review thread, I tried, but both my emails were refused …
Somebody on that board says: “why did you wait so long” … well, this is not entirely true, I have documented criticisms about the Wikipedia for process for quite a long time already, they are also available in my delicious tag here at http://del.icio.us/mbauwens/Wikipedia. I have also on occasion refracted these arguments in this very blog,
But it is true that I crossed some kind of treshold. I think there are systems that can be repaired, and other that can’t. My feeling after going over the evidence is that the dysfunctional process is probably beyond reform, and that the deletionist power grab is too entrenched. Before, I was of the opinion that the dysfunctions were part of a broadly healthy ecosystem that could repair itself from within.
I keep an open mind though.
Michel
January 14th, 2008 at 9:54 am
Hi Michel,
I think the article is good, and both accurate and worrying.
I started a chat about it here which you’d be most welcome to contribute to, if you’ve got the time and energy.
Wikback.com is much more directly connected to the ‘wiki-powerful’ - having been setup by a current member of the ‘Arb Com’ - and it seems to finds criticism of wikipedia harder to digest than the plain speaking at Wikipedia Review - but it’s doing ok, and registration shouldn’t be overly onerous should you be interested in discussing anything further.
thanks again for the interesting article,
best,
PM
January 14th, 2008 at 10:09 am
“…the deletionist power grab is too entrenched.”
I would argue that the deletionist / inclusionist debate much loved by Wikipedians is of little real consequence to the debate.
The key issues are those around themes of defamation, misinformation, corruption, unaccountability and an all pervasive resistance to responsible practices.
January 14th, 2008 at 12:48 pm
My second objection against presented arguments is that most of them seam to lead to one unsatified wikipedia user.
January 14th, 2008 at 4:36 pm
I think that it is very important to distinguish Open Source projects in the software domain from so-called “Open Source” projects in other domains, especially non-technical domains like amateur journalism and general information almanacks of the Wikipedia variety.
The discipline of programming and the rigor of technical documentation maintain a constant reality check on Open Source software projects. No one there decries the brand of expertise that can prove itself in practice. It is not incidental that GFDL was to designed to suit just those domains, not the kinds of domains where the infantile whims of Power Rangerâ„¢ fanatics can defeat the good faith efforts of ethical reporters and lifelong scholars.
January 15th, 2008 at 9:20 am
How can we turn it into a positive ‘recommendations’ list for big online communities? What would be the lessons learned?
January 15th, 2008 at 10:07 am
Hmm - just tossing an idea - a system of that size easily can reach and surpass the complexity of a state. So how about using some of the classical means like tripartition of power (http://en.wikipedia.org/wiki/Separation_of_powers
for example?
January 16th, 2008 at 11:56 am
Hi Zbigniew,
I think that one of the remaining and crucial advantages of peer governance is that it is a permission-less system, i.e. you can undertake actions and produce freely, and the validation process is a posteriori. I think the problem with Wikipedia is the validation process which doesn’t require expertise and thus it becomes a battle of wills. In the chosen condition of artificial scarcity (deletionism) it would therefore be important to find mechanisms that are both democratic but also include a modicum of expertise, as is done in the free software field. Formal democratic governance is useful in all areas of scarcity, where resources have to be allocated, but it is equally important to learn to profit from abundance, and use probabilistic production where we can.
Michel
January 23rd, 2008 at 5:30 pm
Wikipedia also doesn’t want articles on uncomfortable subjects, whether or not they’ve been covered by major media outlets:
http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Deletionist_versus_Inclusionist_Controversy
January 24th, 2008 at 9:10 am
[…] Jon: Wikipedia also doesn’t want articles on uncomfortable subjects, whether or not… […]
January 28th, 2008 at 5:29 pm
Michel, I found this critique very generative. I do hope that it engenders productive discussion.
However, I think you are missing a larger discourse on group dynamics and organizational behavior. There is a body of socio-technical and socio-psychological work from the 1940s and 1950s that has demonstrated what Jo Freeman claims about groups and structure. This work builds on the work of Melanie Klein, Wilfred Bion, Elliott Jaques, A.K. Rice, Fred and Merrilyn Emery and many, many others. References to this line of work can be found at http://www.akriceinstitute.org .
I think that your arguments would be stronger if the claims about group dynamics and behavior referenced the available literature. Jo Freeman’s article is a good one, but it contains no references. She was not the first to notice our human group behaviors. From a historical perspective, Sigmund Freud described humans as herd animals in (Group Psychology and the Analysis of the Ego), and I’m thinkin’ he was not the first.
C. Fred Alford provides an analysis of the group perspective in our political life in Group Psychology and Political Theory.
I find that balancing my desire to just do things myself, with my life as part of many groups, is an ongoing challenge. It’s often easier to try to ignore it. It’s not a great tactic.
You are right: more open discussion of how we actually work in groups and how authority and labor need to be divided to get work done in groups is required, and it will be very beneficial. As I said above, I hope we can have productive conversations.
February 1st, 2008 at 2:31 am
[…] Is something fundamentally wrong with Wikipedia governance processes? » P2P Foundation - The Wikipedia is often hailed as a prime example of peer production and peer governance, an example of how a community can self-govern very complex processes. Including by me. But it is also increasingly showing the dark side and pitfalls of purely inform […]
February 11th, 2008 at 3:53 am
Assume that (our understanding of) information in Wikipedia has undergone the trajectory of categorization. In other words, it has undergone a trajectory from binary (something either is/isn’t information, so all information is equally subject to change in Wikipedia) to spectral (some information is more x [insert metric, e.g. complexity] than other information, so some information is more subject to change than other information) to typological (there are different types of information, so different types of information exhibit different subjectivities to change). As Wikipedia’s governance system has emerged, the entry for science has become increasingly difficult to edit whereas that for social graph (which, would you believe, references this blog) still maintains its malleability. (If subjectivity to change sounds too subjective, use the word ‘resilience’ and check the history of an entry and track how often changes are made. Actually, because some changes are just copy edits, it would be most beautiful to visualize the evolution of a Wikipedia entry – incorporating date of edit and quantity of characters added/removed, among other variables. Or this could be translated into music, as would be possible with emails. Or into whatever, when you think about it.)
But how to typologize information? Specifically, which typology of information correlates with resilience within the context of Wikipedia? One possibility is information hierarchy, which splits the enchilada into data, information, knowledge, and wisdom. Considering that data is used to generate information, which is used to create knowledge, and so on, wisdom may be the least likely to change. But Wikipedia is about information. Hmm. Another possibility is Funtowicz & Ravetz’s post-normal science, which splits it up into applied science, professional consultancy, and post-normal science (see image). But again, Wikipedia is about information; not science. (I’m aware that I’m littering this post with Wikipedia links, but I won’t it so easy this time: if you don’t know the difference between information and science, check Wikipedia.) Still, these typologies may be illuminating. Re: DIKW, could information be categorized into concentric types? Re: post-normal science, F&R’s axes of choice are decision stakes and systems uncertainty; what would be the relevant axes for information?
Wikipedia has been criticized for advertising itself as the encyclopedia anyone can edit, but becoming increasingly resilient; I’d argue that only certain types of information are increasingly resistant. At the net level, there’s a tradeoff between editability/dynamism and quality/governance. But if you look at the distribution of resilience across entries, there certainly is one.
February 12th, 2008 at 7:26 pm
Stephanie,
Students of information theory know that that information and control are intimately related, sometimes in sync and sometimes dually, if not “duelly”, in a reciprocal trade-off with each other.
In our study of Wikipediatrics it is well that we keep both sides of this coin in mind.
Those who do will almost immediately recognize that the controllers of Wikiputia are far more interested in amplifying their control over people, their beliefs, and their consequential actions than they are in augmenting people’s information.
Just a little something to think about.
May 15th, 2008 at 3:22 pm
Wikipedia is indeed a wonderful running-experiment in social dynamics.
I will point out a fundamental error in your analysis though. Wikipedia is in no sense “structure”-less nor “informal”. It has had from day one a very clearly and explicitly defined social structure explicitly facilitated by its formal (technical) mechanism: utter egalitarianism.
What it lacked was a formal Social Hierarchy.
What we are seeing is the creation of Hierarchical structures, via the age-old mechanism of esoteric sects/subcultures. “Four legs good, two legs BETTER.”
To be clear: you are confusing a flat structure with absence of structure; you need to reword your core thesis to address the creation of a hierarchy (elitism), not of structure.
May 21st, 2008 at 2:03 am
I do not regard the Wikipedia as structureless, but in fact I hold it as a problem that the non-expert editors (but with experience in the norms and rules), can, through notability and deletionism, hold power of more expert subject authors.
As for your argument, that is exacly the argument of Jo Freeman in her theory of structurelessness, that it indeed creates hidden structures. See the beginning of the article.
So I’m not sure where you located that fundamental flaw,
Michel
May 22nd, 2008 at 3:54 pm
Wikipedia —
It’s Not Rocket Science —
It’s Racket Science —
Jon Awbrey
June 6th, 2008 at 6:39 am
[…] Link: Is something fundamentally wrong with Wikipedia governance processes? » P2P Foundation […]