Clay Shirky
( Archive | Home )

Liz Lawley
( Archive | Home )

Ross Mayfield
( Archive | Home )

Sébastien Paquet
( Archive | Home )

David Weinberger
( Archive | Home )

danah boyd
( Archive | Home )

Guest Authors
Recent Comments

pet rescue saga cheats level 42 on My book. Let me show you it.

Affenspiele on My book. Let me show you it.

Affenspiele on My book. Let me Amazon show you it.

Donte on My book. Let me show you it.

telecharger subway surfers on My book. Let me show you it.

Ask Fm Anonymous Finder on My book. Let me show you it.

Site Search
Monthly Archives
RSS 1.0
RSS 2.0
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline


« If Six Apart acquires Live Journal.... | Main | Year of the Enterprise Wiki »

January 5, 2005

Wikipedia: Me on boyd on Sanger on Wales

Email This Entry

Posted by Clay Shirky

My response to danah’s response about the Wikipedia/anti-elitism debate:

First, some background. I have the same “yes, but” reaction over and over to Wikipedia detractors — much of what both Sanger and boyd say is wrong with the Wikipedia is wrong with it, but then there’s this incredible leap from “The site as it stands has faults” to “…and so it must be ignored or radically altered.”

Reading pieces like Sanger’s, I feel like I’m being told that bi-planes fly better than F-16s because F-16’s are so heavy. You cannot understand how well things fly without understanding both weight and thrust.

It’s a similar leap to assume that, since the Wikipedia has disadvantages relative to the Encyclopedia Britannica, Britannica must therefore be better. The real question is “Weighing the advantages and disadvantages of the Wikipedia against the advantages and disadvantages of Britannica, under what conditions is Britannica better, and under what conditions is Wikipedia better?”

And of course, sometimes Wikipedia is better, since, as with the Indian Ocean tsunami example, Britannica simply has no offering. So, at the margin, a casual user who wants free access to a Web site that offers a communally-compiled and non-authoritative overview of a recent event will prefer the Wikipedia to nothing, which is what Britannica offers. In this case, Wikipedia comes out on top, and walking along several of those axes like cost, availability, topicality, and breadth of coverage, Wikipedia has the advantage, and in many cases, that advantage is increasing with time

Now Britannica doesn’t want this to be true (god, do they not want this to be true) and so they try to create litmus tests around authoritativeness — “WARNING: Do not read anything that does not come from an institutional source!” But this is as silly as audiophiles dismissing the MP3 format because it wasn’t an improvement in audio quality, missing entirely that the package of “moderate quality+improved cost and distribution” was what made the format great. Considering MP3 as nothing more than a lossy compression scheme missed the bundle of services that it enabled.

So with Wikipedia — it’s possible to make Britannica look superior to the Wikipedia by simply excluding characteristics Britannica cannot compete on, like cost, accessibility, and timeliness. And with that as background, I found this statement of danah’s, “It will never be an encyclopedia” (emphasis hers), to be both right and wrong in illuminating ways.

I agree with her if you add the unspoken bit to that sentence: “It will never be an encyclopedia (of the sort Britannica is.)” Stated this way, the assertion is true, and I don’t think anyone from Wikipedia would dispute that — not only is it radically different, it’s principle value is in that radical difference.

However, Britannica is not the only model for an encyclopedia. Before Britannica, encyclopedias had authors — Pliny, Bacon, Coleridge. Even Diderot and D’Alembert’s version, the bridge between the ancient and modern encyclopedia, had its entries written by the great and the good of the era. Britannica made it possible to extend the project of gathering and expressing encyclopedic knowledge, by substituting institutional brand for authorship. This allowed them to have more people working on more subjects, over a wider range of both inquiry and time, than was previously possible.

This was a huge leap, and one that required a shift in authority, away from the literal (authority derived from a particular author) to authority by proxy (though the entries have authors, the authority is derived from the values of the institution that employs those authors.)

So the idea that the Wikipedia will never be an encyclopedia is in part an ahistorical assertion that the definition and nature of encyclopediahood is fixed for all time, and that works like Britannica are avatars of the pattern. Contra boyd, I think Wikipedia will be an encyclopedia when the definition of the word expands to include peer production of shared knowledge, not just Britannica’s institutional production.

Comments (11) + TrackBacks (0) | Category: social software


1. Alex on January 5, 2005 3:07 PM writes...

Very good rebuttal... this is a very interesting debate :)

Permalink to Comment

2. Tony on January 5, 2005 3:27 PM writes...

Yes, good points. I suspect that Wikipedia is, on average, less accurate and filled with more errors than Brittanica, but Brittanica certainly has errors as well. I would never trust either one 100%.

Wikipedia is a useful resource, despite its shortcomings.

Permalink to Comment

3. Jay Fienberg on January 5, 2005 4:02 PM writes...

I think the historical perspective (Clay mentions) is useful, because it illustrates how any collection of information is reflective of its source(s). While, different styles of human organization gather/produce information in different ways, there is no absolute "good" way to gather/produce information.

In the last century or two, big bureaucratic insititutions were looked upon as the highest authorities, and knowledge-work was often organized in this institutional model. But, that model hasn't always been so popular, and it seems all good that this new "wiki" model is getting such a thorough testing in wikipedia.

Permalink to Comment

4. Eric Vaughan on January 5, 2005 4:29 PM writes...

I'm glad that you included criteria on which to compare Wikipedia and Brittanica. It's important that people understand that differences as well as advantages and disadvantages. Using Wikipedia for formal references (scientific papers, published books) may not be as credible as the proven Brittanica. However, commonly referenced articles and general knowledge can be accessed more quickly with more information available by using Wikipedia. In all, it's important to realize when to use each knowledge base when comparing the two. It seems you've done a good job in explaining this.

Permalink to Comment

5. jake on January 5, 2005 4:57 PM writes...

Wikepedia is good and all, but I think the whole Tsunami coverage argument is a little weak in defending it.

To me, the point of an Encylopedia is to give a synthetic overview of a topic. Obviously, there will be an implicit bias in this account. There's always a politics involved in this sort of account.

But the thing that gives this account any legitimacy at all is some shared communal/cultural sense that it represents more or less an uncontroversial conventional wisdom account of the subject.

It's the very filteredness of it that makes it a worthwhile reference. Obviously, meaning is always contested terrain. And conventional wisdom shifts over time. Historically, Encyclopedias have reflected this.

Nevertheless, I think there is a zone of reasonableness with regard to accuracy at any given time. And for better or for worse, it takes some time for information to get processed to the point that it can reasonably be considered to reside in this zone.

Part of what gives us assurance that the quality of the information processing is up to snuff, is that the people doing it have been trained in the appropriate methodologies and ethical standards.

I know it's perhaps a bit of a stretch, but would you let your teenage son or daugher perform surgery on you? Maybe they'd do a great job, because they have a natural talent for it. Maybe if more laypeople like them did surgery there'd be a lot more and quicker access to much needed health care. But maybe it's not such a good idea, because it seems like the risks outweight the benefits.

Is it elitist to want someone with some training cutting you open?

If the answer to the question above is no, then why should the processing of information be any different? Why is it all of a sudden elitist to take a bit of comfort in the knowledge that the people filtering the info also have some skills and training in this regard. It may be a small comfort. In some cases, it may also be misplaced (just as our trust in doctors in routinely misplaced). But it does provide a bit of a framework and some stability.

So I guess it's great that there is a Tsunami entry in the Wpedia. But on another level, if I'm going to place my trust in contemporaneous accounts of this event, I'm probably more likely to just look in the newspaper. At least there it's clear what I'm getting. It's also clear to me that there is an institutional and ethical framework in place to guide journalistic practitioners. This is certainly no guarantee of accuracy or ethical behavior. But it does provide some sort of base line.

Or I guess I can go look in some Blogs. This can be useful too. But there again, it's clearer what I'm getting. These are also generally one person affairs. So it's easier over time to determine whether a given blogger can be trusted. Indeed, one of the really valuable things about many Blogs is the way in which these folks serve as one person filters. If you come to trust them, they can really be useful resources.

If the Wpedia account of the Tsunami is simply an aggregation of journalistic accounts, that could be useful. But to the extent it porports to be a synthetic account unto itself, there is a risk involved, and I don't think it's completely off base or elitist for people to raise concerns about the "as is" nature of the info offered.

I'm all about freedom, individuality, etc. But everyone does not bring equal skills and training to the table. Sometimes, these things don't matter. Obviously, there are plenty of incompetent and unethical people out there who have the right credentials. Sometimes the amateur turns out to be the person who really does the best and most thorough job.

So there is no certainty. We're talking more about an actuarial sort of thing. On average, who is likely to provide us with the more accurate account. There is a certain appeal to the notion that more accounts of something automatically aggregates into a better and richer understanding of the subject. I like that idea in theory. It's certainly congruent with my personal ideals about democracy, etc. I'm just not sure it turns out to be true in practice, because often it seems like it just creates a lot more noise and not that much more signal.

Maybe I'm just stuck in some old elitist paradigm. But we've got a pretty long history of processing information along the lines I've described. Obviously, there is a power/cultural politics in that. It's a form of battle, just like everything else. Sure it has served certain interests during its history. But that doesn't mean it's without pragmatic utility.

So I guess I'm sympathetic with the concerns being raised here. It's scary enough how many people seem to believe that if it's on the internet it must be true. Democratizing information is a great thing. But it has pitfalls too. Those need to be acknowledged, and I for one don't fault people who have concerns about quality control.

Permalink to Comment

6. Jimbo on January 5, 2005 8:54 PM writes...

You're right about Encyclopedia Britannica not having anything on the Tsunami. It also doesn't have an entry on the GNAA,, or tubgirl. Therefore, Wikipedia is better.

Permalink to Comment

7. John Mark Ockerbloom on January 5, 2005 10:35 PM writes...

I'm enjoying the comments, and you're quite right that Wikipedia and EB are different kinds of works, but both fall into the general encylopedic tradition in different ways.

This doesn't mean, though, that Wikipedia as it is run today is the only or even necessarily the best way to produce an globally edited, open access encyclopedia. One of the useful bits from the critique by Robert McHenry (the former EB editor) was that frequently edited WP articles tended to entropically decay over time, not so much from vandalism as in the fuzzy, inconsistent, and obtuse little changes made by random people over time. Of course, more knowledgeable and meticulous folks can go back and fix this periodically, but there's some evidence that a number of them get tired of doing this over and over again, and drop out, often leaving no one in their place to tend the article.

Designating article moderators can potentially help stem this entropy, by letting them manage "stable releases" that can be consulted by readers yet be freely built upon by contributors, with the best contributions being edited into subsequent "stable releases". (As with many OSS packages, you could have both a "stable release" version and a freewheeling "CVS-style" version of an article, with the former being the default and the latter one link away.) You needn't set up a Sanger-style elite, or impose an unscalable Nupedia-style review process for that. One way to avoid those paths is to allow as many moderators as volunteer for the job (usually 0, sometimes 1, occasionally 2 or more). Each moderator could work with the others, if any, on the same development fork, or branch off their own article forks if they didn't like what the other moderators were doing with an article. (There would probably need to be some ways to rate forks or moderators, cross-reference particular forks, and auto-deprecate long-untended forks, but those are implementation details.)

Would this work better than what WP does now? I can't say for sure-- it might well introduce too much complexity-- but I think it's an interesting idea to consider and possibly test. Just because WP works well in many respects as it's set up now doesn't mean it, or a related spinoff, couldn't work better.

Permalink to Comment

8. Phil on January 5, 2005 10:59 PM writes...

An interesting theory to apply to this debate is the one Cory Doctorow used about MP3: New technology doesn't get over by doing the things old technology did better. It does the new things better than the old technology ever could.

Permalink to Comment

9. Franz on January 6, 2005 5:45 AM writes...

I love Wikipedia for a quick overview.
Yet it is not something that can be quoted in a student's paper! especially since there's no guarantee that an article is backed up by accurate research.
e.g. you want to look up some philosopher, then there's a difference if the entry is found in some authorative source (Routledge Guides, etc.) or just a Wikipedia entry compiled by someone who is not established as a recognized expert of the field in question!!
Wikipedia cannot be the consulted source for academics!

Permalink to Comment

10. Benjamin Rosenbaum on January 6, 2005 10:31 AM writes...

Would I let my teenage son or daughter operate on me if the operation could be viewed by hundreds of experts, semi-experts, and other interested parties, and any false step could be quickly reversed with no damage to the patient?

Sure, why not?

Permalink to Comment

11. Benjamin Rosenbaum on January 6, 2005 11:40 AM writes...

Though of course, you can't be sure that *every* article in Wikipedia has the host of experts looking over every posters' shoulder.

Wikipedia is not analogous to an open source project. Rather, wikipedia is analogous to SourceForge -- a *collection* of open source projects. And there's an enormous difference in reliability between a hot open source project like Linux or eclipse (where the level of reliability is higher than any commercial software could dream of) and an open source project that was maintained
by two guys for a while until their courseload spiked.

Wikipedia needs (as a commenter on an earlier post suggested) something analogous to SourceForge's prominent listing of "activity level", which I find absolutely critical in deciding whether I should stake a project on a chunk of code I find in SourceForge. Similarly, if a Wikipedia page hasn't been updated since danah was taking Anthro 101, I should not entrust my head to its content.

And something the net in general needs is a scalable solution to verification of expertness. I expect this will come -- I expect in twenty years to be able to go to an expertise-trust server configuration page and say: "I trust people with PhDs from the following universities -- except for *these* people -- about their topics; people who are committers on open source projects on software topics related to those projects; I trust people trusted by the following people and *distrust* people trusted by these other people; and beyond that, please use a neural-net analysis of my clicking, buying, and posting for the past 25 years to guess at what I will trust. Ok, NOW show me the Wikipedia I deserve."

Permalink to Comment


TrackBack URL:

Listed below are links to weblogs that reference Wikipedia: Me on boyd on Sanger on Wales:


Email this entry to:

Your email address:

Message (optional):

Spolsky on Blog Comments: Scale matters
"The internet's output is data, but its product is freedom"
Andrew Keen: Rescuing 'Luddite' from the Luddites
knowledge access as a public good
viewing American class divisions through Facebook and MySpace
Gorman, redux: The Siren Song of the Internet
Mis-understanding Fred Wilson's 'Age and Entrepreneurship' argument
The Future Belongs to Those Who Take The Present For Granted: A return to Fred Wilson's "age question"