User:Teratornis/Should editors be logged-in users?

Wikipedia allows unregistered users to edit most pages. Many people find this policy puzzling, as it would seem to invite vandalism. Studies demonstrate pretty conclusively that it does. Thus Wikipedia has a steady parade of relatively new users who question the wisdom of the current policy.

Justification for the current policy

The page: Wikipedia:Editors should be logged in users offers the brief rebuttal:

From Wikipedia:Perennial proposals#Prohibit anonymous users from editing: According to Jimbo Wales, "what is commonly called 'anonymous' editing is not particularly anonymous ... and there are good reasons to want vandals on ip numbers instead of accounts." While about 97% of vandalism comes from anonymous users, about 76% or 82% of anonymous edits are intended to improve the encyclopedia. (Prohibiting IP edits would not eliminate 97% of all vandalism, because if they have to, those inclined to vandalism could easily take the 10 seconds to register.) The ability of anyone to edit articles without registering is a Foundation issue.

Analysis of the justification

It's interesting to examine the fallacies, unsupported claims, or unraised issues in the rebuttal:

Prohibiting IP edits would not deter vandalism

• "Prohibiting IP edits would not eliminate 97% of all vandalism, because if they have to, those inclined to vandalism could easily take the 10 seconds to register."
  • Certainly they could, but would they? Having to register an account forces a vandal to leave the current page and go look at something else. It might take more than ten seconds to register, for example if the vandal's first N guesses at a username are already taken. The deterrent effect may be similar to locking a door versus leaving the door wide open. Many more people might be inclined to enter through an open door, which seems to invite entry, than would break through a locked door, even if breaking through a door would only take a few seconds for a skilled thief. If breaking in required going away to get some tools first, that would deter a fraction of people who weren't strongly motivated to break in.
  • This claim has never been adequately tested. The definitive way to test the claim would be to block anonymous editors for several months, and study the impact on Wikipedia. Would the presumed loss in constructive edits outweigh the presumed reduction in vandalism? Anyone who claims to know the outcome before performing the experiment remains mired in prescientific thinking.
    • There has been a partial test: Wikipedia originally allowed unregistered users to create new pages. After several years, Wikipedia changed the policy to so only registered users can create new pages now. How has this policy change impacted the rates of constructive and unconstructive new page creation? Do lots of vandals create accounts so they can create new nonsensical pages? One thing is certain: the rate of new article creation remains robust.
    • Another test is the semi-protection of heavily-vandalized pages. When a popular target for vandals gains semi-protection, do lots of vandals create accounts so they can continue vandalizing it? Does the loss of constructive anonymous edits substantially slow the progress of semi-protected pages toward featured status?

Many anonymous edits are constructive

• "About 76% or 82% of anonymous edits are intended to improve the encyclopedia."
  • No one knows how many of those anonymous editors would be deterred by the requirement to register an account.
  • No one knows how many of those anonymous editors actually do have accounts, but merely forget to log in, since Wikipedia does not require logging in to edit, and it's easy to forget. Some registered users might not bother to log in before editing, since they don't need to do so for routine edits.
    • For some users, it is actually bothersome that Wikipedia does not require them to log in, because any edits they make anonymously do not accumulate under the contributions of their username. This can be an issue for users who seek administrator status, for example, as it may artificially depress their apparent contributions.

Focusing on the edits that occur

The rebuttal compares constructive and unconstructive anonymous edits. But these are the edits which occur. What about edits which do not occur, under the current policy? Surely there are some constructive editors who tire of un-doing vandalism, and leave the project, or some potential constructive editors who decide the project is not credible because the current policy facilitates so much silly vandalism. Unfortunately, the potential constructive editors who do not edit because of the current policy leave no trace of their potential contributions, and thus this hidden cost of the current policy is at the moment unmeasurable.

However, given the number of users who question the current policy, it seems plausible that a substantial number of potential users decided not to stay here because of that policy. If some people question the current policy, yet continue to edit, then presumably there are others who object to the policy a bit more strongly, and simply leave. While Wikipedia cannot hope to be all things to all people, it would be useful to at least acknowledge this possible hidden cost of the current policy, when evaluating the policy.

If Wikipedia were to test the current policy, by reversing it for a period of time (i.e., by requiring users to log in before editing), then presumably some of the constructive contributors who might be deterred under the current policy would then become more likely to contribute, perhaps offsetting those contributors who strongly object to registration for whatever reason.

It's a foundation issue

• "The ability of anyone to edit articles without registering is a Foundation issue."
  • While this claim is true, and is of historical interest, if someone uses it to justify the current policy, they are making an appeal to authority and an appeal to tradition, neither of which are logically valid justifications for the policy.
    • Appeal to tradition is logically meaningless, because the policy may or may not have been originally justifiable, and conditions have changed in any case, implying that the policy needs to be revisited periodically to make sure it has not outlived its usefulness. Certainly, Wikipedia has changed substantially over the years. Today Wikipedia is the world's most popular wiki, and one of the most popular Web sites of any type. The English Wikipedia has 47,389,967 registered users; even if that number was somehow insufficient to build an encyclopedia, it is growing by several thousand per day.
    • Appeal to authority is equally meaningless for determining whether something is a good idea. The backing of authority only indicates the difficulty of changing something that currently exists.

It's better for vandals to use IP numbers rather than accounts

• "There are good reasons to want vandals on ip numbers instead of accounts."
  • OK, what are these reasons? Do they outweigh the enforcement benefits of having vandals on accounts? Why don't these "good reasons" seem to apply to the current policies of requiring accounts for page creation, and to edit semi-protected pages? Have those policies encouraged lots of vandals to create accounts? If not, why would blocking all anonymous editing encourage vandals to create accounts? The current policy has the downside of confusing and potentially punishing constructive editors who share IP addresses with vandals. How many warnings and blocks directed at vandals have confused and perplexed innocent unregistered users? I can see a potential problem if vandals were to create large numbers of spurious accounts, thereby chewing up more of the available usernames, but we have 47,389,967 registered users in any case.

Anonymous editing capabilities are eroding

As I mentioned in my responses above, Wikipedia has gradually reduced the editing capability of unregistered users over the years; they can no longer, for example, create new articles, edit semi-protected pages, nor do they get an easy way to rename pages. The notion that we allow edits by unregistered users is gradually becoming less true, and as far as I can tell, all the arguments in favor of continuing to allow unregistered users to edit pages also apply to allowing unregistered users to do all the things we no longer allow them to do. I don't understand why we do not trust unregistered users to create new pages, while we trust them to edit the existing pages. The burden of explanation should be on the people who created the inconsistent policies and argue for them inconsistently.

Most advocates of the current policy do not seem to use it

I also find it odd that almost everyone who upholds the principle of anonymous editing doesn't seem to use it. The constituency in favor of the current policy consists almost entirely of registered users. It is strange that a group of people who promote a feature would not themselves use it.

Wikipedia is different now

When Wikipedia was very new, nobody knew whether it would work, and relatively few people had heard of the project. Thus there were strong arguments in favor of making Wikipedia as easy to edit as possible. There may have been an early tendency for news about Wikipedia to spread faster among potential positive contributors than among vandals.

Today Wikipedia is no longer a secret, and no longer requires promotion to attract contributors.

The ultimate goal of Wikipedia is to have every article attain featured status. If this should ever happen, would we still need to allow every adolescent to edit anonymously? As the quality of an article increases, the article becomes steadily more difficult to improve further. That means the pool of people qualified to edit the article steadily decreases.

Imagine an analogy with a large outdoor wall open to the public for artistic expression. Initially, most of the paintings on the wall probably would not be very good. But over time, better paintings might emerge. If that happens, at some point the wall requires protection to maintain the quality it has attained.

Imagine if an art museum issued paints and brushes to all its visitors and allowed them to touch up the paintings on display. This would make sense if the paintings were bad, or the visitors were improbably good, but if the paintings on display are masterpieces, only a handful of people in the entire world might be capable of improving them.

Thus Wikipedia's history of "creeping protectionism" is not surprising. As the project has grown larger and more established, and the rules have grown more complex and more explicit, the editing privileges available to unregistered users have gradually eroded.

It seems safe to predict that this erosion will continue, in lockstep with Wikipedia's growing success.

See also

• Wikipedia:Anonymity

External links

• a comment I left on WikiAngela about the need to test the effect of requiring logins to edit, rather than merely speculate about the effect