AUTHORITY

I’m stupid and got involved in another argument over Wikipedia, again. And I’m also vain, so I’m documenting the discussion below. Comments are welcome.

D&D: Wikilolocaust
(Note: it is a predictably shitty thread, and my comments don’t appear for several pages. Spare the trouble of actually reading it)

danno posted:

The English-language Wikipedia has 1.67 million articles. For the sake of argument, let’s say that on average an hour of work has gone into each article. Let’s also say that the type of labor that goes into Wikipedia is worth $15 an hour. Multiply and you get $25 million worth of free labor the project has been able to harness.

Now I know I pulled those numbers out of my ass, but my point is that Wikipedia isn’t some infant project trying to make its way on the Web. It’s had six years with thousands of people contributing millions of dollars worth of resources. I don’t think it’s unreasonable to start holding expectations as to its quality.

I’m hesitant to get involved in this shitstorm of a thread, but-

Look, Wikipedia is already a success, end of story. That success isn’t judged on the quality or quantity of the articles, because both those attributes are in constant flux. Individual articles can be criticized in any number of ways, and in any number of dimensions, according to the protocols set forth by the Wikipedia community. And, perhaps most importantly, those protocols are also in flux and can be changed by the community.

It is true that these rules are regulated by an insulated community that is strongly resistant to outside influence, and who devalue traditional forms of authority. But it is wrong to say that it does not accept any form of authority; the authority is just nonstandard. As such, it is entirely appropriate to raise serious epistemological questions about the reliability and accuracy of the Wikipedia model. And, if these endless Wikipedia threads have shown us anything, it is that we have no established and agreed upon ways to evaluate this model that people on both sides of the debate can agree upon that don’t reduce to simple question-begging, anecdotal or non-representative evidence, or pure dogmatic foot stamping. Every argument in this thread so far has amounted to one or the other of these kind of arguments, and so legitimate discussion has stalled.

But Wikipedia’s success doesn’t lie in the quality of the articles. Its success lies in the clear, obvious fact that people can come to rely on and to a certain extent trust a community driven, decentralized resource like Wikipedia. Such resources are entirely unprecedented in all of human history; before Wikipedia it was an open question whether such a model could ever sustain itself. It turns out that people are willing to devote massive amounts of time and energy to maintain these resources, and to passionately defend its legitimacy, at significant personal cost and with no promise of compensation. I’m willing to bet that danno’s estimates significantly undervalues the work put in to Wikipedia. If you would have suggested such a thing even 10 years ago, people would have thought you were crazy. But there it is, right before us.

Accuracy and quality are at best ideals that should definitely be sought, and are rare to achieve. Whatever you think about the Nature study, it at least shows that these ideals are difficult to achieve even in supposedly ideal settings. There is every indication that the Wikipedia community does care about these ideals, even if it doesn’t always live up to them. But these occasional (even if egregious) failures don’t undermine Wikipedia’s success; Wikipedia has succeeded in spite of the tremendous difficulties involved in sustaining this kind of site.

Thinkmeats posted:

The “Everyone sucks” platform has never been a very popular one. Are you planning to stick around to defend it? I’m not really sure what you included this bit in your post; it seems rather secondary to your main points.

It’s not an ‘everyone sucks’ argument. My argument is that we really don’t know how to evaluate the epistemic status of community driven resources like wikipedia, because it is unprecedented and radically diverges from traditional epistemic models. That doesn’t mean that everyone is wrong; it means that we are treading in unfamiliar territory.

No one in this thread has offered a useful model to judge Wikipedia except by way of comparison to paper encyclopedias, but as has already been noted, the medium constraints are so different in the two cases that it is not clear the comparison is appropriate.

So, as we see in this thread, the arguments boil down to ‘I like Wikipedia’ or ‘I don’t like wikipedia’. The only reasonable position offered in this thread is that people shouldn’t take it so seriously, but that is unsatisfying since it entirely avoids the pressing issue of developing new epistemic standards of judgment. I admit that I fall on the former side of the debate, but that doesn’t get us anywhere in terms of understanding how to treat Wikipedia; m discordia has been particularly dogmatic on the other side of the debate.

But what we really need is a new theoretical model for dealing with non-standard, decentralized epistemological resources like wikipedia. I should note that this problem is really one of the big problems with the internet generally; Wikipedia just happens to be a more visible example.

danno posted:

I don’t know if I agree as to the extent people can rely or trust Wikipedia; that’s really the argument going back to the OP, how “essjay” violated every traditional standard of trust and authority to gain those things from the Wikipedia community.

I was being somewhat unclear. I don’t mean to assert that people should trust such sources. That’s a normative claim that requires an epistemic basis for evaluation. My whole point is that we don’t have any such basis, so I can’t argue that we should trust the resource.

I’m making the descriptive claim that we can, and in fact do, trust wikipedia, in overwhelmingly large numbers. Whether or not it is warranted, Wikipedia has become the de facto point of entry in most cases of inquiry, from the serious to the mundane. And Wikipedia has become such a crucial source of information despite the lack of a basis for normative evaluation.

Again, it is easy to take Wikipedia (and the internet generally) for granted, but as Gibson says, we have no maps for these territories. If you had sat someone down 10 years ago and explained the basic operating principles of Wikipedia, I am sure that person would have seriously doubted whether people would actually bother to engage in a project like Wikipedia. But we now know that people will engage in the project, passionately and in great numbers. I think that is the real success of Wikipedia, and is why so many people are fascinated by it, whether they ‘like’ it or not.

danno posted:

I don’t agree that we can’t evaluate Wikipedia by traditional epistemological standards, but you’re right about this being the heart of the debate.

Well, you aren’t giving me an argument, so I don’t know how to take this. But I’ll give you the benefit of the doubt and try to clarify my claim.

Traditional epistemic standards rely on a definition of authority that is basically inherited: authorities have to be ordained by the right people, attached to the right institutions, etc to be considered legitimate. Whether or not that is fair, it begs the question against the Wikipedia model.

To be clear, I think that judging a particular claim to be true does rely on traditional epistemic standards; the truth of the matter doesn’t depend at all on the medium in which it is presented.

But, as Brittanica’s response to the Nature study makes clear, the argument isn’t over truth; the argument is over authority. Brittanica thinks it has authority that trumps Wikipedia. The challenge posed by Wikipedia is a challenge to our standard models of authority; and it is on the basis of authority that we are warranted in trusting the truth-claims from any source. Wikipedia is attempting to claim that community consensus, contributions from respected peers, and vigilant watchdogging can also establish an alternative trustworthy source that doesn’t conform to the standard model of authority.

I think Wikipedia has a real argument here, but it is an open question as to how it will play out. The biggest problem is that on the standard model, an established authority can be held responsible for his or her claims; on the community model, there is no locus of responsibility. The worry is that without responsibility, there is no reason to trust the source. Wikipedia is revolutionary in part because it abandons the need for responsibility- if someone makes an error, we don’t need to hold someone accountable, we just need to correct the error. But the worry again arises- without accountability there is no systematic way of identifying and correcting errors, which leaves a permanent liability in Wikipedia’s model and renders it a fatally untrustworthy source. On the other hand, Wikipedia holds that this liability is merely temporary, and that over the long run the community driven model can correct these errors in non-systematic ways and without any genuine responsibility. I think the Nature study for the most part confirmed this theory, and though people have raised doubts in this thread I haven’t seen any good refutations.

I hope my reconstruction above was fair to people who don’t like wikipedia; if not, I’d appreciate any correction.