- Administrators think they are better than everyone else (you ever hear an admin apologize?). But most of them are just unemployed people with lots of time on their hands.
- With today's tools, even a moron can make a huge number of edits each day. Problem is, the morons do exactly that.
- The mass of editors are unsophisticated, unable to understand that the worldview they and their friends hold is pretty much limited to newspaper readers in Anglo-America.
- The policy on sources is absurd: journalists are considered good sources, but they use much lower quality sources than what would be accepted on WP. As if the average journalist is a good filter for information quality!
- The average editor thinks of WP as a mass of weeds that needs to be hacked, not as a garden that should be grown.
Wednesday, December 30, 2009
Lots of problems
Lots of problems; all recognized, many times, by others:
Monday, November 23, 2009
Speeding up
There's been a bit of discussion lately about why so many editors are leaving the project. Here's one possible reason: the editing tools have become too powerful. Many of the more inferior editors appear to pride themselves on the number of edits they make, and these powerful tools allow them to vastly increase their rate of "productivity." What kind of quality can one expect from an editor who dashes madly about, making hundreds of edits a day? What kind of edit can one expect from an editor who is just spending seconds on each "contribution"? Mostly what one sees are reverts or deletions, since adding content takes a fair amount of time.
By and large, deletionists don't write articles--their primary activity is to delete the work of other people, and they do this not out of principle, but simply because this is the fastest way they can increase their edit count. Deletion is not in itself bad, but it is bad when done without sufficient thought. In most cases, thinking over a bad piece of text will enable an editor to rewrite it. WP's improvement requires thoughtful rewriting, not frantic deletion.
It would be nice to see a change in attitude toward edit counts--if there were a bit of a stigma to generating hundreds of edits a day, the deletionists would become a much weaker force on WP.
By and large, deletionists don't write articles--their primary activity is to delete the work of other people, and they do this not out of principle, but simply because this is the fastest way they can increase their edit count. Deletion is not in itself bad, but it is bad when done without sufficient thought. In most cases, thinking over a bad piece of text will enable an editor to rewrite it. WP's improvement requires thoughtful rewriting, not frantic deletion.
It would be nice to see a change in attitude toward edit counts--if there were a bit of a stigma to generating hundreds of edits a day, the deletionists would become a much weaker force on WP.
Friday, October 2, 2009
It belongs to everyone
Wikipedia is a commons, the property of us all. It's like the city park that, as Jane Jacobs observed, because it belongs to everyone, belongs to no one. People who have nothing better to do drift into the park: the unemployed, the drunk, the homeless. Respectable people scurry quickly through, and because they don't feel a sense of ownership, they don't speak up and reprimand those who behave in antisocial ways.
Wikipedia is like that city park, in that anyone can show up, but most people have something better to do. Editors who spend all day long at Wikipedia, accumulating hundreds of edits a day, are fairly common. These cannot have jobs, cannot have families, cannot even have many dates. Try to visualize these editors and an image appears of a middle-aged guy living in his parents' garage.
Not a problem, except victory in Wikipedia always belongs to the most persistent. In a dispute, people with real responsibilities will state their case and then give up if it looks like a long struggle is ahead. People with a lot of time on their hands can thus often get their way, and can easily manage to become a real pest without actually falling afoul of the rules.
Wikipedia is like that city park, in that anyone can show up, but most people have something better to do. Editors who spend all day long at Wikipedia, accumulating hundreds of edits a day, are fairly common. These cannot have jobs, cannot have families, cannot even have many dates. Try to visualize these editors and an image appears of a middle-aged guy living in his parents' garage.
Not a problem, except victory in Wikipedia always belongs to the most persistent. In a dispute, people with real responsibilities will state their case and then give up if it looks like a long struggle is ahead. People with a lot of time on their hands can thus often get their way, and can easily manage to become a real pest without actually falling afoul of the rules.
Sunday, August 23, 2009
Academic Freedom
Most Wikipedia editors are clueless about academic freedom. Like the common run of folks, they believe that an academic should be busy producing insights that confirm what all of us already know. Their reaction to an academic whose work has unveiled uncomfortable truths is one of distaste, repulsion, condemnation. Perhaps the best example would be that of historians specializing in the late Ottoman Empire, historians such as Justin McCarthy. McCarthy, a demographer able to read the Ottoman archives, has examined changes in local populations during the last years of the empire, and has supplied substantive numbers to the claims and counterclaims of ethnic cleansing and massacres during that period. His conclusions are not to the taste of those who maintain that there was an "Armenian Genocide." How is McCarthy treated on Wikipedia? By being formally categorized as an "Armenian Genocide Denier." I'm not making this up, take a look.
The Ottoman historians are not alone. Hostile editors edit articles on other academics working on controversial issues. A good example is Richard Lynn, an IQ researcher who has become a leading expert on between-group IQ differences. His work is extremely politically incorrect, but its quality compares favorably with that of his more politically correct peers. A decent article on Lynn has eventually emerged on Wikipedia, but it took a long time, and lots of struggle.
A final example would be that of Ian Stevenson, a psychiatrist at the University of Virginia who spent his career examining cases of children who appear to remember details of past lives. His findings give mild support to the existence of reincarnation, a result not at all to the liking of skeptics, who are imbued with the unshakable faith that reincarnation can not possibly exist. Stevenson's article, as well as the related article on "reincarnation research", have been turned into pages where obscure journalists are cited to buttress claims that this research is "pseudo-science". Problem is, Stevenson's team has produced the only scientific work on reincarnation, so it boggles the imagination how those defenders of science--the skeptics--are able to announce that science disagrees with his findings.
The Ottoman historians are not alone. Hostile editors edit articles on other academics working on controversial issues. A good example is Richard Lynn, an IQ researcher who has become a leading expert on between-group IQ differences. His work is extremely politically incorrect, but its quality compares favorably with that of his more politically correct peers. A decent article on Lynn has eventually emerged on Wikipedia, but it took a long time, and lots of struggle.
A final example would be that of Ian Stevenson, a psychiatrist at the University of Virginia who spent his career examining cases of children who appear to remember details of past lives. His findings give mild support to the existence of reincarnation, a result not at all to the liking of skeptics, who are imbued with the unshakable faith that reincarnation can not possibly exist. Stevenson's article, as well as the related article on "reincarnation research", have been turned into pages where obscure journalists are cited to buttress claims that this research is "pseudo-science". Problem is, Stevenson's team has produced the only scientific work on reincarnation, so it boggles the imagination how those defenders of science--the skeptics--are able to announce that science disagrees with his findings.
Friday, May 8, 2009
In the long run, we are all dead
The Classical school of political economy holds the view that recessions are short-lived and self-correcting. When a recession occurs, some labor and capital falls idle. According to the Classicals, these idle factors of production react by accepting employment at a lower price; factor prices continue to fall until firms have willingly hired all "surplus" factors. Thus, unemployment (of capital and labor) automatically ends, through the mechanism of falling factor prices--a mechanism requiring no government intervention.
John Maynard Keynes famously disagreed with the Classical view, arguing that factor prices are "downwardly rigid"--they do fall when factors are idle, but reluctantly and slowly. "In the long run," Keynes acknowledged, falling factor prices would eliminate the recession, "but in the long run, we are all dead." In other words, the self-correcting mechanism works too slowly to be useful.
This is also the problem with the eventualist view of Wikipedia. Sure, given enough editors, articles will eventually evolve toward something fair and balanced, but the problem is that eventually can be a very long time. Too long to be useful.
Some parts of Wikipedia move towards excellence much faster than others. A sophisticated user of Wikipedia understands this, and has learned which parts are worthy of trust and which parts should best be avoided. The problem is that the average user of Wikipedia is not sophisticated--she is a high school student or college student, looking up a topic about which she knows next to nothing. That the topic may eventually have a good article is not much consolation to the student who is imbibing misinformation today. And it is certainly no consolation to her teacher, who will continue to ban Wikipedia as a legitimate source for research papers.
John Maynard Keynes famously disagreed with the Classical view, arguing that factor prices are "downwardly rigid"--they do fall when factors are idle, but reluctantly and slowly. "In the long run," Keynes acknowledged, falling factor prices would eliminate the recession, "but in the long run, we are all dead." In other words, the self-correcting mechanism works too slowly to be useful.
This is also the problem with the eventualist view of Wikipedia. Sure, given enough editors, articles will eventually evolve toward something fair and balanced, but the problem is that eventually can be a very long time. Too long to be useful.
Some parts of Wikipedia move towards excellence much faster than others. A sophisticated user of Wikipedia understands this, and has learned which parts are worthy of trust and which parts should best be avoided. The problem is that the average user of Wikipedia is not sophisticated--she is a high school student or college student, looking up a topic about which she knows next to nothing. That the topic may eventually have a good article is not much consolation to the student who is imbibing misinformation today. And it is certainly no consolation to her teacher, who will continue to ban Wikipedia as a legitimate source for research papers.
Tuesday, June 24, 2008
Clique cohesion requires an enemy
Wikipedia editors are by and large too individualistic to stick together in cliques. Even in situations where several editors work together on an article, there is not likely to be full agreement and editors must learn to accept that others approach the topic differently. It's not much fun to make big compromises, and most editors prefer to work on articles where they can make edits unhindered by the need to accommodate others.
Cohesion in a clique requires a force to counteract the friction of working with others. That force is usually ideological, though personal ties reinforce the ideological. A clique will form with the intent of protecting territory from an enemy. A sense of in-group and out-group develops. The out-group is seen to have reprehensible beliefs or to belong to a reprehensible segment of the population. The clique requires an enemy, and will assign that role to almost any editor who enters their territory, imputing to this stray editor the worst of motives.
In such cliques, there is not likely to be much agreement beyond the agreement that the enemy must be kept from the territory. The pages protected by a clique will be badly written and full of erroneous information. The reason for the low quality is not the constant attacks of the enemy, but rather that the clique controlling the territory is unable to organize itself for the task of article writing. In fact, the members of the clique might not know much about the subject of the pages they are defending, since ideological fervor is often a sign that one does not understand the full complexity of an issue (an example of the Dunning-Kruger effect).
The ability of the clique to control territory is based upon the willingness of clique members to back each other up in an edit dispute. In Wikipedia, of course, edit disputes are over content, so there is always an argument about content at the surface of a dispute. Clique members have the illusion that they have won a dispute because they were correct about content, when in fact they have only won because of their numbers and their cohesiveness. Arguments over content are likely to be repeated whenever a new editor wanders into their territory, and the clique finds its arguments more compelling each time they are repeated (an "availability cascade"). Trust simultaneously strengthens among the members, as each sees that the others, again and again, back them up. Cliques thus can become stronger over time.
Cliques can usually stay within the boundaries of the rules. A numerical advantage allows it to out-revert the solitary editors who oppose them, and the ease with which the clique wins content disputes helps its members stay calm and civil. Solitary editors in conflict with the clique are more likely to run afoul of the rules, reverting too often or exploding with frustration and saying uncivil things. As the rules currently stand, cliques are not often threatened by administrators.
Cliques are a special form of the "ownership" problem on Wikipedia. One proposal I've seen is that all of the editors active on an article or its talk pages can be asked to leave (for several months), opening up the article to a new set of editors. For this to work, a rule must determine when an article hits a state that all of the active editors must go elsewhere. It must be a simple rule, requiring little research on the part of the administrator, so that it cannot be contested--like the 3RR rule.
Cohesion in a clique requires a force to counteract the friction of working with others. That force is usually ideological, though personal ties reinforce the ideological. A clique will form with the intent of protecting territory from an enemy. A sense of in-group and out-group develops. The out-group is seen to have reprehensible beliefs or to belong to a reprehensible segment of the population. The clique requires an enemy, and will assign that role to almost any editor who enters their territory, imputing to this stray editor the worst of motives.
In such cliques, there is not likely to be much agreement beyond the agreement that the enemy must be kept from the territory. The pages protected by a clique will be badly written and full of erroneous information. The reason for the low quality is not the constant attacks of the enemy, but rather that the clique controlling the territory is unable to organize itself for the task of article writing. In fact, the members of the clique might not know much about the subject of the pages they are defending, since ideological fervor is often a sign that one does not understand the full complexity of an issue (an example of the Dunning-Kruger effect).
The ability of the clique to control territory is based upon the willingness of clique members to back each other up in an edit dispute. In Wikipedia, of course, edit disputes are over content, so there is always an argument about content at the surface of a dispute. Clique members have the illusion that they have won a dispute because they were correct about content, when in fact they have only won because of their numbers and their cohesiveness. Arguments over content are likely to be repeated whenever a new editor wanders into their territory, and the clique finds its arguments more compelling each time they are repeated (an "availability cascade"). Trust simultaneously strengthens among the members, as each sees that the others, again and again, back them up. Cliques thus can become stronger over time.
Cliques can usually stay within the boundaries of the rules. A numerical advantage allows it to out-revert the solitary editors who oppose them, and the ease with which the clique wins content disputes helps its members stay calm and civil. Solitary editors in conflict with the clique are more likely to run afoul of the rules, reverting too often or exploding with frustration and saying uncivil things. As the rules currently stand, cliques are not often threatened by administrators.
Cliques are a special form of the "ownership" problem on Wikipedia. One proposal I've seen is that all of the editors active on an article or its talk pages can be asked to leave (for several months), opening up the article to a new set of editors. For this to work, a rule must determine when an article hits a state that all of the active editors must go elsewhere. It must be a simple rule, requiring little research on the part of the administrator, so that it cannot be contested--like the 3RR rule.
Wednesday, February 20, 2008
Free riding in Wikipedia
Axelrod and Hamilton (1981) showed that a winning strategy in prisoner's dilemma games is tit-for-tat; that is, one should start out behaving cooperatively, and then on every subsequent move simply match what the other agent did in her previous move. This strategy, however, describes only 2-person, repeated games. In games where multiple agents interact with each other, even a small number of defectors will prompt the conditional cooperators to also defect, so that the game moves to an equilibrium of no cooperation. Thus, the model cannot explain how cooperation persists in groups with multi-agent interactions.
Experimental games have shown that cooperation can be maintained when defectors are punished. But since punishment is costly to the punisher, it would be rational for a player to let others punish the defectors--these rational players are called second-order free riders (since they are free riding by letting others punish free riders). Without rewards for the punishers, or punishment for the second-order free riders, no punishment of defectors will occur.
Panchanathan and Boyd (2003) have shown that reputation works well in models as the reward sought by punishers, who gain in reputation by helping the deserving (those with good reputations) and punishing the undeserving (those with bad reputations--i.e., persistent defectors). The ultimate reward of reputation is that others willingly cooperate with them. The reputation model requires that agents know each other's reputations, which is only realistic in small groups.
Second-order free riding is a common problem on Wikipedia. One often encounters editors who appear to be doing something wrong: inserting a strange point-of-view; moving pages without discussion; deleting good work done by others. But it is costly to battle these people. Conflict itself is unpleasant, and one is likely to break some rules when in a conflict. If an administrator observes the broken rule she is unlikely to have the time to look back at the history of the conflicting parties and figure out who is the rogue and who is the guardian, and she will simply punish whoever broke the rules. What she lacks is ready access to information on the reputations of the parties in conflict.
A mechanism to record and update reputation would encourage good editors to oppose the actions of bad editors--it would reduce the problem of second-order free riding, and make it more difficult for rogue editors to get their way.
Experimental games have shown that cooperation can be maintained when defectors are punished. But since punishment is costly to the punisher, it would be rational for a player to let others punish the defectors--these rational players are called second-order free riders (since they are free riding by letting others punish free riders). Without rewards for the punishers, or punishment for the second-order free riders, no punishment of defectors will occur.
Panchanathan and Boyd (2003) have shown that reputation works well in models as the reward sought by punishers, who gain in reputation by helping the deserving (those with good reputations) and punishing the undeserving (those with bad reputations--i.e., persistent defectors). The ultimate reward of reputation is that others willingly cooperate with them. The reputation model requires that agents know each other's reputations, which is only realistic in small groups.
Second-order free riding is a common problem on Wikipedia. One often encounters editors who appear to be doing something wrong: inserting a strange point-of-view; moving pages without discussion; deleting good work done by others. But it is costly to battle these people. Conflict itself is unpleasant, and one is likely to break some rules when in a conflict. If an administrator observes the broken rule she is unlikely to have the time to look back at the history of the conflicting parties and figure out who is the rogue and who is the guardian, and she will simply punish whoever broke the rules. What she lacks is ready access to information on the reputations of the parties in conflict.
A mechanism to record and update reputation would encourage good editors to oppose the actions of bad editors--it would reduce the problem of second-order free riding, and make it more difficult for rogue editors to get their way.
- Axelrod, Robert, and William D. Hamilton. 1981. “The Evolution of Cooperation.” Science 211: 1390-1396.
- Panchanathan, K., and R. Boyd. 2003. “A tale of two defectors: the importance of standing for evolution of indirect reciprocity.” Journal of Theoretical Biology 224:115-126.
Subscribe to:
Posts (Atom)
