Corante

Authors

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

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.

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

mobile games on My book. Let me show you it.

http://www.gunforums.com/forums/showtopic.php?fid/30/tid/15192/pid/111828/post/last/#LAST on My book. Let me show you it.

Site Search
Monthly Archives
Syndication
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

Many-to-Many

« Fukuyama's Penguin | Main | Coates' new shorthand definition of social software »

January 6, 2005

Wikipedia: The nature of authority, and a LazyWeb request...

Email This Entry

Posted by Clay Shirky

There was another point of danah’s I wanted to respond to, but yesterday’s post had gotten quite long enough, and in any case, I had a slightly different take in mind for this, including a LazyWeb request at the end, relating to this image:

danah says “Wikipedia appears to be a legitimate authority on a vast array of topics for which only one individual has contributed material. This is not the utopian collection of mass intelligence that Clay values.” This misconstrues a dynamic system as a static one. The appropriate phrase is “…for which only one individual has contributed material so far.”

Wikipedia is not a product, it is a system. The collection of mass intelligence that I value unfolds over time, necessarily. Like democracy, it is messier than planned systems at any given point in time, but it is not just self-healing, it is self-improving. Any given version of Britannica gets worse over time, as it gets stale. The Wikipedia, by contrast, whose version is always the Wiki Now, gets better over time as it gets refreshed. This improvement is not monotonic, but it is steady.

Picking up on yesterday’s theme of authority, the authority of, say, Coleridge’s encyclopedia was the original one: authority derived from the identity of the author. This is like trusting Mom’s Diner, or the neighborhood tailor — personal reputation is worth preserving, and helps assure quality.

The authority of Britannica, by contrast, is the authority of a commercial brand. Their sales are intimately tied into their reputation for quality, so we trust them to maintain those standards, in order to preserve an income stream. This is like trusting Levis or McDonald’s — you don’t know the individuals who made your jeans or your french fries, but the commercial incentive the company has in preserving its brand makes the level of quality predictable and stable.

So, is Wikipedia authoritative? No, or at least not yet, because it has neither the authority of the individual merchant or the commercial brand. However, it does have something that neither mechanism offers, which is a kind of market, where the investment is time and effort rather than dollars and cents. This is like the eBay model, where people you don’t know (no Mom’s Diner effect) can sell unbranded things like art or one-of-a-kind clothes (no Levis effect). EBay can do this because the syndication of user attention and the possibility of recourse for bad behavior keeps people generally honest.

Now when eBay launched, people were skeptical, because the site wasn’t trustworthy. The curious thing about trust, though, is that it is a social fact, a fact that is only true when people think it is true. Social facts are real facts, and have considerable weight in the world. The fact that someone is a judge, for example, is a social fact — the authority that attaches to judgeship is attached by everyone agreeing that a certain person has the right to make certain statements — “Court is adjourned”, “I sentence you to 5 years in prison” — that have real force in the world. Those statements are not magic; their force comes from the social apparatus backing them up.

Ebay has become trustworthy over time because the social fact of its trustworthiness grew with the number of successful transactions and with its ability to find and rectify bad actors. Indeed, the roughest periods in eBay’s short life have been when it has seemed in danger of being a platform for fraud.

Like trustworthiness, authority is a social fact, though authorities often want to obscure this. A PhD is an authority figure because we all agree that the work that goes into getting a doctorate (itself a social fact) is a legitimate source of authority.

So, under what conditions might the Wikipedia become a kind of authority, based on something other than authorship or brand? And the answer to that question, I think, is when enough people regard it as trustworthy, where the trust is derived from the fact that many eyes have viewed a particular article.

And here danah points to something interesting — she believes, and I believe with her, that a Wikipedia page created by a single user isn’t as valuable as a page that has been edited by many users. Seeing this, I wrote a little script that fetches both a Wikipedia page and grabs 4 relevant facts from that page’s history: number of edits, number of editors, and the first edited and most recently edited dates, and put that info just below the page title, like so.

It’s damn slow, and should in any case be either a bookmarklet (O LazyWeb, hear my plea…) or a setting on the Wikipedia itself, but here are some others: Vinyl [Edited 25 times by 18 users, between 20 May 2001 and 3 Jan 2005], England [Edited more than 500 times, most recently by 209 users, between 2 Jun 2004 and 6 Jan 2005], Orca [Edited 319 times by 155 users, between 17 Oct 2001 and 6 Jan 2005], and so on.

You can always get to any given page’s history on the Wikipedia, of course, but it is a click away from the content page, and is so detail-filled that it is essentially inside baseball — it is mostly for other editors, not for readers. But the facts danah wants are there, under the surface — how many people have worked on this page? 1? 50? 200?, as well as the facts I’m interested in: how many times has it been edited? (More is better, and a low ratio of contributors to edits suggests a serious conversation went into the pages creation.) How long has it been around for review? (Older is better.) How recently was it updated? (How live is the work on the page?)

I am guessing that this kind of dashboard info, embedded in any given content page, would have two good effects — first, it would give the users a trust profile per item, and second it would serve notice to users about how often Wikipedia articles are edited — the article on England was edited over 500 times in the last 6 months alone. This would help viewers see that they are, at any given moment, seeing a static snapshot of a dynamic system, rather than a finished product.

And the more macro point is that Wikipedia is still in the early days of experimenting with models of governance, editing, or, as here, presentation to the users. That’s one of the remarkable things about the fluidity of web technologies — new things can be tried, modified, and retained or thrown out easily and continuously (it took me two dozen lines of python to add the header I wanted, and I am a lousy programmer,) giving Wikipedia a lot more headroom for earning the users’ trust.

And once that social fact is established, authority, albeit of a more diffuse and systems-oriented sort, won’t be far behind.

Comments (8) + TrackBacks (0) | Category: social software


COMMENTS

1. Jay Fienberg on January 6, 2005 8:26 PM writes...

Rather than "commerical brand" (or along with it), I'd use the phrase "recognized organization".

The brand is a nice way to summarize how the organization is recognized, but it's ultimately the organization behind the activity that one is trusting and/or seeing value in.

(In general, we're talking about groups of people, whether mob-like or hierarchically organized. But, one could also look at "recognized individuals" who produce authoritative works more single-handedly.)

So, I think another way to state your question is: what does it take for the mob-like group of people who make Wikipedia to become a recognized organization?

I think Wikipedia is recognized colloquially as a "wisdom of the crowd" kind-of organization. But, as a paradigm, governance by crowd / wisdom by crowd isn't formally recognized as something as valid as governance / wisdom managed by strict hierarchical organization (i.e., recognized by the existing paradigm of hierarchical organizations, e.g., universities and newspapers).

If this were different, one way to look at Britanica would be that it's like a broken Wikipedia in that (within the Britanica organization) there are too few people who are too tightly organized to capture the breath and depth of subjects that should be in an good encyclopedia.

Permalink to Comment

2. jake on January 6, 2005 11:18 PM writes...

A few comments (actually, quite a few ). Apologies, but I don't know how to embed quotes. So I'll just use quote marks:

Clay Writes:

"Picking up on yesterday’s theme of authority, the authority of, say, Coleridge’s encyclopedia was the original one: authority derived from the identity of the author. This is like trusting Mom’s Diner, or the neighborhood tailor — personal reputation is worth preserving, and helps assure quality.

The authority of Britannica, by contrast, is the authority of a commercial brand. Their sales are intimately tied into their reputation for quality, so we trust them to maintain those standards, in order to preserve an income stream. This is like trusting Levis or McDonald’s — you don’t know the individuals who made your jeans or your french fries, but the commercial incentive the company has in preserving its brand makes the level of quality predictable and stable."

Jake comments:

Yes, but a brand is some sort of an ethereal thing. It is a symbolic representation of a product or the underlying institution that created it. Trademark rights are common law in nature and they attach through use.

So while it is true that at some level this sort of structure is a reification, in practical terms its role in building trust must be considered. To me this is really the core of the concerns about Wpedia. Is the structure adequate to the task of establishing something authoritative enough to be useful.

Clay continues:

"So, is Wikipedia authoritative? No, or at least not yet, because it has neither the authority of the individual merchant or the commercial brand. However, it does have something that neither mechanism offers, which is a kind of market, where the investment is time and effort rather than dollars and cents. This is like the eBay model, where people you don’t know (no Mom’s Diner effect) can sell unbranded things like art or one-of-a-kind clothes (no Levis effect). EBay can do this because the syndication of user attention and the possibility of recourse for bad behavior keeps people generally honest."

Jake comments:

True, but the transaction costs involved in making something like the Wpedia effective are very different than those on eBay, because they address very different sorts of transactions.

This doesn't mean that Wpedia couldn't be evolved to address these differences, but I think it's useful to acknowledge that existing approaches to processing this sort information may be as they are because they meet these exigencies better than the Wpedia structure can (at least as presently constituted.

Clay continues:

"Now when eBay launched, people were skeptical, because the site wasn’t trustworthy. The curious thing about trust, though, is that it is a social fact, a fact that is only true when people think it is true. Social facts are real facts, and have considerable weight in the world. The fact that someone is a judge, for example, is a social fact — the authority that attaches to judgeship is attached by everyone agreeing that a certain person has the right to make certain statements — “Court is adjourned”, “I sentence you to 5 years in prison” — that have real force in the world. Those statements are not magic; their force comes from the social apparatus backing them up."

Jake comments:

I think trust is more than simply a social fact. In many cases, it is an emperical fact. This is what separates eBay from something like Wpedia (at least in many cases). I have been playing the guitar for 26 years. If I buy a guitar on eBay, I have the knowledge necessary to determine whether I got what I paid for. I'm buying a tangible good. I can play it. I can see if it works. I know enough to see if there are buzzes, a warped neck, etc.

The only reason I would look at eBay feedback is for assurance that the seller isn't going to take my money and not send me what I think I am buying (and if returning the item is an option that the seller will take it back).

But ultimately, if I didn't feel confident that I could assess emperically the good I was purchasing, I would not buy it on eBay, even if the seller had great feedback. At least for me, the risk is simply too high. That's why there are many items I would never buy on eBay.

Nevertheless, there are enough people in the world who have knowledge about at least a few items of merchandise available on eBay, that a system like eBay can function and trust can be built. For at the end of the day, mechandise is a tangible thing. It usually has a zone of purpose for which it was designed, so at that level there already exists a reasonably clear consensus about what it means for that item to "work."

But when the market relates to intangibles like ideas, analysis and information, it becomes much more difficult for the average person to assess the trustworthiness of the information emperically.

This is why it's much more difficult for most people to hire a Doctor, Lawyer, Mechanic, etc. The thing being purchased is much harder to pin down and assess than something like a guitar or computer, because one needs a certain amount of specialized expertise to truly determine whether what you are buying is good.

In this context, people are much more likely to look to testimonials of other people they trust, or look to external indicators like price, a fancy office, or whether the person has on a nice suit. Even though these things ultimately have no direct connection to quality, they are the best we can do. So we use them, and if nothing goes wrong, they validate that we have made the right choice on the right basis.

I suppose to some extent one might argue that things like a fancy degree are kind of the same thing. But I think credentialing structures do have more of an empirical conection to quality than do things like a fancy office and testimonials of satisfied customers.

Clay continues:

"Ebay has become trustworthy over time because the social fact of its trustworthiness grew with the number of successful transactions and with its ability to find and rectify bad actors. Indeed, the roughest periods in eBay’s short life have been when it has seemed in danger of being a platform for fraud.

Like trustworthiness, authority is a social fact, though authorities often want to obscure this. A PhD is an authority figure because we all agree that the work that goes into getting a doctorate (itself a social fact) is a legitimate source of authority."

Jake comments:

I don't disagree that authority is a social fact. Everything is socially constructed. But this is beside the point. The issue isn't that this is so. The issue is what sort of institutional structures yield authority that furthers a social good. Obviously this is a political questino and a political process. It's contested terrain, and I think the people who have concerns about Wpedia worry that at least as presently constituted it will be bad for society if it acquires too much authority (or in the alternative that the existing structure is incapable of acquiring enough authority to make it truly useful to society).

Clay continues:

"So, under what conditions might the Wikipedia become a kind of authority, based on something other than authorship or brand? And the answer to that question, I think, is when enough people regard it as trustworthy, where the trust is derived from the fact that many eyes have viewed a particular article."

Jake comments:

I can't agree with you on this. In this context I'm not sure there can be any meaningful authority that isn't based on authorship or brand (to the extent that brand is the outward symbol of an underlying institutional structure that insures a minimum level of quality)

To my mind, the quality of the eyes looking at an article is much more important than the quantity. Five hundred people reading/editing an article who know little or nothing about the topic do little or nothing to improve its authoritativeness. Conversely, one person who knows what they are talking about can produce something that's very useful.

Ultimately, this is one of the flaws with any sort of system like eBay or for that matter customer reviews of a product on Amazon or whatever. It's useful as far as it goes. But it's only as good as the people reviewing the product.

That's why authorship and brand/underlying institutional structure are such important and imbedded concepts in our culture.

It's also why the open source idea makes more sense for software than it may for something like the Wpedia. If you are building software, eventually it will have some stated purpose, and there will be at least some provisional emperical standards against which it can be assessed. To use a term from the comment above, if the mob decides to build a word processor, eventually the success or failure of this project will hinge on how well the word processor works. So if an incompentent programmer contributes to the code base, it will be relatively easier to see that their code either doesn't work or is inefficient to the task.

Information processing and synthesis doesn't work quite this way. An open source model has its merits at the level of style and grammar editing (which is akin to making existing computer code more efficient). But at the level of editing substance, I think it becomes far more complicated, and issues of governance, etc become paramount if the work product is to have much utility.

Permalink to Comment

3. Matt Jones on January 7, 2005 1:40 AM writes...

Re: Your lazyweb plea. It would be rather nice to have a small 'sparkline' type infographic (a graphic element that can display dense, multidimensional information inline with text, proposed by Edward Tufte) that displayed the shape and speed of alterations to the article, a la Martin Wattenberg's History Flow project for IBM.

This would enable you to see immediately whether the entry was smooth or spiky - contentious or not contentious etc...

Permalink to Comment

4. Zbigniew Lukasiak on January 7, 2005 5:22 AM writes...

I would add the number of views to the metric. A view with not edit confirms that the information is right just as an edit shows that it was corrected. With an initially good article you will not have many edits - so we need some more complex measures. I believe we should squeeze all available data and number of view is another one beside number of edits. Then there could be a system where some views and some edits would value more than others (a karma system).

Permalink to Comment

5. Zbigniew Lukasiak on January 7, 2005 5:34 AM writes...

Or perhaps there should be an 'Confirm' button just to express the fact that the information is correct in ones eyes.

Permalink to Comment

6. pedant on January 8, 2005 3:31 AM writes...

Your link to a 404 message at http://shirky.com/wiki/Wikipedia:Text_of_the_GNU_Free_Documentation_License does not quite satisfy the GFDL's requirements ;)

Permalink to Comment

7. Blake on January 11, 2005 11:55 AM writes...

That's pretty interesting. I never realized Wikipedia worked in that nature...edited by others. My trust in it is no less, just like my trust in eBay is no less if I get a bad auction. The system is good, but that particular individual didn't have my best interest in mind.

Permalink to Comment

8. soobrosa on January 20, 2005 4:55 AM writes...

getting back again. we need infographics, something like Metacritic has, but with CSS.

"Examples of possible prototypes

... Wikipedia. Considering the nature of the knowledge construction taking place in Wikipedia, one never knows of how much "truth" is the actual page contains, until she/he does not check some crucial information about the given page. [Worth noting the overall good "truth-factor" of the Wikipedia, by Lih, Andrew: Wikipedia as Participatory Journalism: Reliable Sources? Metrics for evaluating collaborative media as a news resource, 5th International Symposium on Online Journalism, 2004]

Every Wikipedia page has some metadata-like attributes those are not shown, or not easily recognizable, but stored in the database. These are the following
- was the content of the page discussed ever, by anyone,
- when was the page started,
- how many people contributed to the page,
- how many edits were made to the page,
- were there any major flame or vandalism regarding the content of the page,
- how many other pages link here.
I propose that if every Wikipedia page would have a graphical representation of these data or they relation to each other, that would help the user to have an immediate opinion of how much should she/he should trust that page. (Font size, coloring, shading could be easily done even within HTML using CSS - no need for special graphic generation methods.)

If we continue this idea in a way that considering the user is an active contributor, we should show her/him, how "close" or "far" is a given article to her/him. We should interpret "closeness" based on an Erdos number-like model. Closeness means trustability. And here we also come back to the geographically-sensitive sticker board that really works when you know can trust or mistrust information based on "closeness".

Trust is handled in quite different ways in collaborative knowledge communities, I personally think that the most interesting experiences came from slashdot (see: Rutigliano, Lou: When the Audience is the Producer: The Art of the Collaborative Weblog - http://journalism.utexas.edu/onlinejournalism/audienceproducer.pdf). Even technically sophisticated users are lazy and all feedback mechanism should be formed according to this easy phenomenon. Giving away a number of "trust points" they are able to give to articles I propose three easy buttons: "I like it" (you give trust), "I don't like it" (you lower trust) and "Alert" (you see something strange). If you are using either "I like it" or "I don't like it" too much, the system have to make you argue why do you think so. In case of "Alert" you have to describe why do you think that you do not trust at all what you see."

Permalink to Comment

TRACKBACKS

TrackBack URL:
http://www.corante.com/cgi-bin/mt/teriore.fcgi/1802.

Listed below are links to weblogs that reference Wikipedia: The nature of authority, and a LazyWeb request...:


EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
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"