Log in

No account? Create an account

Sneezes while Sneaking

I recently came across Kat Walsh (a prominent-and-awesome Wikipedian I used to know)'s Women on Wikipedia essaylet. I don't have a lot to add on that topic — I'm sure that the regularly inculturated interaction styles differ between males and females (probably differing regionally and by subculture too), but not everyone shares that inculturation. There is a natural high tension on Wikipedia because of the nature of the project; by being a global context/setting/group where people with a lot of different cultural norms, speech habits, and notions of what an encyclopedia should be must collaborate, there is difficulty in convincing swathes of existing or future contributors to adhere to a set of norms. The mechanisms of consensus, broadly understood to be necessary and healthy for societal self-regulation, are called despotic by some former members of the community (particularly those driven out for their inability to compromise). This complaint is in the extreme minority, but differences of opinion (such as this) are important object lessons in community interaction, illustrative of ideas one could get from a study of political philosophy.

With Wikipedia, I could say that I was there before the start — I was on the nupedia mailing lists and watched that community flounder for a while before turning my attentions elsewhere, and once Larry Sanger started Wikipedia, I lurked and edited anonymously for a time before establishing an identity there. On one hand, Wikipedia has gathered a lot of content, some of it high-quality. It has been interesting to watch power structures evolve within the community, and I believe a very nice encyclopedia-ish thing could be carved out of what the project is. On the other, I think there's been a governance gap between the highest levels and the content-creators, and while it's been interesting to see the community more-or-less fill that gap, I suspect (not sure) that something more structured would've given better results. The problems could either have been dealt with culturally or through the more formal structures (there are formal structures, and I was part of some of them) — I believe that either the existing formal structures are not comprehensive enough or that the informal mechanisms have failed to generate a culture that could adequately handle big decisions and goal-defining on its own. I don't think Jimbo *or* Sanger were the right leaders for the Wikipedia project, and so I ask, with the benefit of hindsight, 「How would I have run Wikipedia differently?」

  1. I believe that there should've been a stronger central discussion of standards and rules for the project, and that very few disputes should've been discussed on a per-page basis. Content-based straw polls should've been rare, and much more comprehensive/tight standards for articles should've emerged. Some of this is for consistency, some of it is to prevent the same issues from coming up in many different places, and some of it is good for the community in that when decisions are reasonably final, people will accept that decision or leave
  2. I believe the initial goals of the project (and cultural guidelines) should've been more clear. Debates over deletionism versus inclusionism were (are?) due to an unclear vision, and these create ill-will. Were I to have managed this, I would've defined an encyclopedia as an effort to codify academic knowledge, and provided the metrics of
    1. The goal of an Encyclopedia is to present academic knowledge to the general population
    2. Any content that is not likely to be relevant in one hundred years should not be included. Any content that is included should take that 100-year perspective.
    3. There are other wikis; when in doubt, information that does not fit with traditions of Encyclopedia can and should be hosted elsewhere.
    4. Verifiability is necessary but not sufficient for coverage.
    5. People with excessive interests in pop culture or OCD can easily derail the project
  3. The product is the goal, not the culture that produces it
  4. There should be formal respect for academic expertise which takes precedence over popular reporting. On matters of truth, we should imagine the following hierarchy (on matters of phrasing/prose/presentation, no deference to expertise is necessary/desirable — experts do not own articles):
    1. Professors with known names speaking within broad academic consensus on their topic of expertise, supported by publications
    2. Other people summarising broad academic consensus, supported by publications
    3. Other experts speaking within topical consensus on their field, supported by nonacademic publications (only when the field is of academic/broad interest)
    4. Other people summarising topical consensus on a field, supported by nonacademic publications (only when the field is of academic/broad interest)
  5. Precedence should figure more strongly into disputes
  6. Leadership should not tiptoe around larger disputes on the project. When a decision is needed, it should be willing to step in impartially, firmly, and with relative finality. This should generally be reserved for matters that relate to project vision or to make sure that topics of contention don't remain damaging in the long-term.
  7. Flagged revisions should've been there from the start, and it should be present on all articles now.
  8. In general, discussions on the project should more resemble juristic reasoning, not democracy. Voting is less valuable than a reasoned discussion that references accepted principles of the project. Policy interpretation should be based on qualified sets of people reading the arguments, not a single person counting the votes.
  9. People new to the project should not be given weight in decisionmaking or disputes — time and an understanding/acceptance of policy should be required for someone to become vested. Likewise, admin rights should not be a standard rite of passage for each editor, nor should it be seen as "no big deal".
In sum, I believe that an excessively egalitarian spirit, ill-defined goals, and hands-off approach by Jimbo have hurt the project.

It is important to note that I would not expect conflict to go away or for the endless stream of malcontents who have left the project to become happy — to do that would be to set expectations far too high. There is no way for reasonable organisations to suffer trolls without causing great unhappiness (are trolls a symptom of undersocialised, excessively individualistic people? How do other cultures deal with them? Could they be "corrected" with "a good talking-to" in-person? What kinds of jobs do they tend to have off-the-network where their tendencies are not a barrier?)

Some of you may not have heard about the murder of David Kato, a gay activist in Uganda. The details of the murder are not clear, although he had recently been targeted by a newspaper that outed many gays and suggested people "hang them". As background, a number of American far-right missionaries have been working with the Ugandan parliament pushing a bill calling for the death penalty for homosexual acts. This murder is a sad act, a sadder one yet if it is in fact related to homophobia in Uganda. Until:

  • homosexuals
  • bisexuals
  • those who have changed their gender-identities or choose to ignore gender-norms
  • live in safety
  • marry their partners of whatever gender-role or sex they choose
there is a fundamental injustice to address in the world. Likewise, we can and should disrespect the portions of any religion, philosophy, or culture that disapprove of homosexuality/non-gender-normativity or would block legal rights for the same. This is our commitment as liberals, and in common with those of whatever other political identities that would join us on the issues.

I believe U.S. Foreign Policy should take an active stance in pushing other nations on this front. Should there be criminal liability for those evangelicans that are either pushing the death-penalty-for-homosexuality bill or contributed to the culture of hate? Probably not, although I believe:

  1. It is an option to ban missionary activity as a reason for travel. This (international travel) is within the realm of reasonable control that states may place on their people, just as they might reasonably not permit someone to leave the country to join a foreign war. People might lie about why they travel, and it's reasonable that the state might apply sanctions for that deception or investigate the reason for travel. (I believe it is also a reasonable option for states not to permit people to enter for missionary activity)
  2. It is desirable to consider any Ugandan lawmakers, judges, or police that supported or enforced anti-gay laws as persona non grata.
  3. It is reasonable to censor broad publication of works that are sufficiently racist, sexist, homophobic, or otherwise inspire hatred, provided a reasonable metric for doing so can be established.


I think in your discussion of Wikipedia, you have the benefit of a lot of hindsight; and while that is in part the point, I think we might disagree on what would have been reasonably forseeable at the start of the project.

I think more central discussion at the beginning could well have produced worse results for the project, because nobody at the beginning could forsee what wikipedia would turn into, and what things would work or not work. I tend to agree with your point 8 -- juristic reasoning works best. But if you examine how that's done in real judicial systems, it's never by pre-planning, but always by example and accumulated precedent from real cases and controversies. So ultimately I disagree that disputes should not be discussed on a per-page basis; but I do agree with the idea, also seen in judicial systems, that once a decision is made, turning it into a precedential rule is important so people don't keep arguing about it.

Again with vision -- who's to say, if there had been a clearer vision to begin with, that it would have been the right one? An unclear vision allowed the community to feel and fight its way to good answers, and it's not immediately clear to me that it would be possible to eliminate all the bad blood and conflict without also eliminating the results. (As you probably know, I am also a strong inclusionist, so I don't agree with the specific vision you have laid out; and even if I might concede that it's a good vision for the encyclopedia _now_, I am not convinced wikipedia would have gotten the critical mass it has without all the non-academic articles to draw in contributors.)

I do agree that the product is the goal, but it's important to respect the culture that produces that product; fail to do so and you get nupedia, or scholarpedia, or any of the other failed wiki projects.

Similarly with respecting academic expertise: I think wikipedia's culture of explicitly disrespecting academics is bad and should be changed, but I think if you give academics formal precedence over the rest of the community, you will drive away good contributors -- the ones who actually put time and energy into the project -- in favor of expert contributors, and if you look at scholarpedia you can see what a disaster that is.

I do absolutely agree about flagged revisions.

I also agree that established contributors should have more decisionmaking weight. To some degree that's already true. But I think it would be nice if there were some more formal way of enforcing it; I don't like the current system where people who spend a lot of time studying and gaming the rules can get their way over people who spend a lot of time editing articles.

I also happen to think it should be easier to ban jerks and assholes from the project, no matter how good their credentials or how much they have contributed. Time and again I think one finds that productive people who can't play well with others just aren't worth working with. I think this would go a long way towards making a more civil environment for discussion, which is something I think would materially improve wikipedia for contributors.
It is true that I both have hindsight and that I'm guessing as to what would've been better. I wouldn't claim any of this was really forseeable.

I think what might've worked for precedent is for people to have kept an eye on decisions made non-centrally and pulling them in (to the village pump) whenever they're likely to come up again. This happened a little bit when I was active on the project, but it had to be intiated by editors who already knew about it/wanted to do it. People often preferred decisions be made by topical cliques.

I don't think there are clearly right or wrong project visions (well, maybe there are a few clearly wrong ones, but there may be several right ones). For example, while I don't think an inclusionist Wikipedia is a good encyclopedia, it's still a worthy information project. It's one possible "right" design, just as a moderately deletionist one might be (my stance) or a hardliner one would've been. The same goes with other topics — the point is that a decision should've been made.

Citizendium and Scholarpedia have the advantage and disadvantage of going second. Filling an already-filled niche is very difficult; it's hard to pin their difficulties on their policies. Having a conceptual divide between the ability to state facts and an ability to present them hopefully would still leave room for informal contributors. I think that the point of an encyclopedia is to disseminate academic knowledge; nonacademes can help with presentation of that knowledge but having a significant mainstream academic input (either through papers or directly) directly contributes towards the goals I would like the project to have.

I think banning people who "do not play well with others" is a general difficulty on the internet. I wonder how much of this can be attributed to issues in geek culture versus intrinsic difficulties in semi-anonymous large-scale communities.