« On the hot seat | Main | The zen of Web 2.0 »

Web 2.0's numbskull factor

April 27, 2006

On his blog, Harvard's Andrew McAfee mulls over some of the barriers to the successful adoption of Web 2.0 collaboration technologies by businesses. He points out that even the most prominent examples of Web 2.0 platforms, such as Wikipedia, are actually produced by a relatively small number of contributors. Many people may consume the product, but only a fraction of those contribute to its production. In addition to the famous "long tail" in demand for information goods, in other words, there's a long tail of production. As McAfee puts it:

I think there’s also a long tail among people, and it relates not to willingness to consume (i.e. demand) but rather to willingness to produce. In November of 2005, the most recent month for which comprehensive stats are available, Wikipedia had over 850,000 articles in English, and 2.9 million across all languages ... This content was generated by fewer than 50,000 contributors in English, and 103,000 total ... And even this population is skewed: active English wikipedians (more than 5 contributions in a month) numbered 15,600 last November, and very active (100 or more) numbered only 2,081 ...

If companies only get the same fraction of Intranet users to use [Web 2.0] tools, these tools will be roundly and rightly acclaimed as failures. Business leaders have to find ways to increase the ‘ambient percentage’ of internal wikipedians, bloggers, taggers, etc. well beyond what we’ve observed so far on the public Internet. Demonstrating that these tools will increase productivity, decrease workload, and put hours back in the week will certainly help, but I wonder if such demonstrations will be enough.

McAfee makes a critically important point. But I'd go even further. Although wikis and other Web 2.0 platforms for the creation of content are often described in purely egalitarian terms - as the products of communities of equals - that's just a utopian fantasy. In fact, the quality of the product hinges not just, or even primarily, on the number of contributors. It also hinges on the talent of the contributors - or, more accurately, on the talent of every individual contributor. No matter how vast, a community of mediocrities will never be able to produce anything better than mediocre work. Indeed, I would argue that the talent of the contributors is in the end far more important to quality than is the number of contributors. Put 5,000 smart people to work on a wiki, and they'll come up with something better than a wiki created by a million numbskulls.

The quality of any entry in Wikipedia, for instance, is ultimately determined not by how many people work on it but by how many talented people work on it. An entry written by a single expert will be better than an entry written by a hundred fools. When you look deeply into Wikipedia, beyond the shiny surface of "community," you see that the encyclopedia is actually as much, or more, a product of conflict than of collaboration: It's an endless struggle by a few talented contributors to clean up the mess left by the numbskull horde.

In a business context, the talent of the contributor pool is a particularly critical factor in the success of a Web 2.0 platform - far more important than the raw number of contributors or the ratio of contributors to total employees. As studies of organizational dynamics show, useful knowledge is not disseminated evenly throughout a business. The distribution is clumpy. A relatively few people hold a relatively large portion of the smarts, the expertise, the contacts, the political savvy and so on. Getting those people - the meritocratic elite - to contribute to a collaboration platform is a big challenge facing Web 2.0 in the enterprise. Those people, to speak generally, tend to be the busiest (the most in demand, anyway) and the least likely to have either the time or the interest to suffer the contributions of fools. They also have some very good economic and social reasons not to want to share their knowledge broadly without suitable compensation. As earlier knowledge-management failures have shown, the elite often have the least incentive to get involved, and without them, the project's doomed.

Moreover, if the elite hold back at the start, the product is likely to be defined at the outset by the contributions of the mediocrities. That's the kiss of death, because it turns off everyone else immediately. I wish I could say that companies should seek to find ways to both maximize the contributions of the organizational elite and minimize the contributions of the numbskulls, but if the former's difficult, the latter's impossible. When it comes to Web 2.0, if you build it, the numbskulls will come.

UPDATE: You knew it had to happen: A blogger has written a response to this post titled, without irony, "The Wisdom of Numbskulls."


Hi Nick,

So much for the wisdom of crowds I take it? :-)

Actually, I would venture that the real success factor for Web 2.0 in the enterprise lies not in whether the most elite or talented contribute at all.

Rather, the meaningful factor for success is if a socially constructed set of information is better than not having the information at all.

For example, I would wager that even a mediocre wiki with useful enough information is generally much better than not having one at all. Just look at Wikipedia if you need a good example.

In the end, the result of putting useful tools into people's hands, whether they are talented or not-so-talented, should be judged on whether value is generated, not if the information is always of the highest quality.



Posted by: Dion Hinchcliffe at April 27, 2006 01:45 PM

I originally read a phrase above as "these tools will decrease productivity, increase workload, and put hours backed-up in the week".

Nobody (practically) is going to hear me say this, but wiki's are basically a well-known programming tool called "Revision Control Systems". These are important for programming teams, to allow people to collectively edit the "documents" (code files), without stepping on each other and keeping track of changes. But nobody has ever suggested that a bunch of bad programmers plus a revision control system is equal to good programmers.

I wish more attention would be paid to how Wikipedia seems to be a loss-leader to selling businesses on useless fads.


"2006-03-30 10:29:07
Wikia wins $4 million series A venture capital
Bessemer Venture Partners, Omidyar Network led round

ST. PETERSBURG, FL.-Wikia, Inc., formerly known as Wikicities, has raised a $4 million Series A round from Bessemer Venture Partners, Omidyar Network and its angel investors, including Marc Andreessen, Dan Gillmor, Reid Hoffman, Josh Kopelman, Joichi Ito, and Mitch Kapor.

Wikia is an advertising-supported platform for developing and hosting community-based wikis. Specifically, Wikia enables groups to share information, news, stories, media and opinions that fall outside the scope of an encyclopedia. Jimmy Wales and Angela Beesley launched Wikia in 2004 to provide community-based wikis inspired by the model of Wikipedia--the free, open source encyclopedia founded by Jimmy Wales and operated by the Wikimedia Foundation, where Wales and Beesley serve as board members."

Posted by: Seth Finkelstein [TypeKey Profile Page] at April 27, 2006 01:49 PM

Synical but right to focus on the people aspects.
I believe it will take a few heroes within each enterprise to make any kind of web 2.0 impact.If the heroes start the movement and get enough traction and support it will happen organically. The problem would come if those heroes happen to be the same as the 'Smarts' you refer to. I would hope that at least some of them would have the 'making it better' mentality as well as the 'smarts' mentality.

It has to come from the inside see my SOA vs Web


Posted by: Al at April 27, 2006 02:15 PM

What's new here Nick? Business has always had sheep and shepherds - it's why there's only 1 CEO at MSFT and not 61K of them.

Posted by: Dennis Howlett at April 27, 2006 02:25 PM

but who are that "elite" in the company - often not the ones on the top/middle of the org charts. Another fascinating application of social networking is to see who in the information and influence flow. The X-Rays below the org chart...

Posted by: vinnie mirchandani at April 27, 2006 02:30 PM

Vinnie - That's why I said "meritocratic" elite. You're absolutely right that it's not necessarily the same group as the organizational elite. (Though if there's not at least a little overlap, you should probably start looking for a job with a different company.) Nick

Posted by: Nick Carr at April 27, 2006 02:38 PM

I have been meaning to tell my boss (me) to go pound salt -)

Posted by: vinnie mirchandani at April 27, 2006 02:45 PM

Are you suggesting that you, Andrew McAfee, Tony Karrer (definitely glad that I finally saw my name linked to the two of you) who are adopters of Web 2.0 approaches (blogs) are part of the numbskulls?

I think you will get a more even distribution than you imagine especially given the fact that some of the "numbskulls" will hold back because they realize they don't have the knowledge.

I do think there's an open question of whether we'll get massive adoption, but its more likely with the current toolsets than in the past:

Enterprise 2.0 - What's the PU?

Posted by: Tony Karrer at April 27, 2006 04:13 PM

Tony, I don't actually see individual blogs as being examples of platforms for participative content production. I see them as the fulfillment of the promise of the vanity press. Nick

Posted by: Nick Carr at April 27, 2006 04:21 PM

Nick, Nick, Nick!
What you have said is so true!
Now we can only sit and wait for the intelligent elite to start blogging so that we no longer will have to peruse the blathering of the numbskulls currently filling the blogsphere.

Posted by: Mike Drips [TypeKey Profile Page] at April 27, 2006 04:44 PM

It's unfair to compare the goals of your average internal company website with those of the Encyclopedia Britanica, or of Stephen King. The right metric is whether your average company, with some stars and some average people will do a better job of sharing information with a wiki or a bunch of blogs than with a paper-based newsletter, or with email as businesses have been using it for a decade or so. It seems to me that there are quite a few applications for such tools within big business that will make it easier for people to find other workers inside or outside the enterprise with whom they share interests, and with whom they could collaborate effectively.

Posted by: ChrisHibbert at April 27, 2006 04:48 PM

Collaboration tools can capture information value from people who have less meeting-resistant communication abilities.

Posted by: Don Marti at April 27, 2006 07:48 PM

This mirrors what we've been seeing in the open source community for decades (before it was even called open source).

The majority of contributions are made by a relatively small number of people, and in successful projects these people are invariably highly talented (either technically, or they may simply have good organizational, leadership and motivational skills) and committed to making the project a success. This is the profile of the original BSD project in the 70s, and every successful major open source project both before and since then (Linux, FreeBSD, X, Python, Perl, Nethack, Apache, GIMP, just to name a few).

The idea of a truly egalitarian project, where all contributions are equal, is something of a pipedream. In reality everyone has differing amounts of skill and time to commit. The successful projects will be the ones which manage to attract a critical mass of both.

Perhaps Wikipedia’s position that all contributions/opinions are equal dooms it to failure. After all, Wikipedia is relativly young and the barrier to entry is pretty low, given that a competitor can harvest Wikipedia itself for content. All this requires is a few talented and driven people to start a competitor with superior policies (in the eyes of the open source market) which enable it to attract a larger mass of talent and time.

Posted by: Peter Evans-Greenwood at April 27, 2006 07:50 PM

I agree that "An entry written by a single expert will be better than an entry written by a hundred fools", etc.. In a corporate context, if you carefully choose the right contributors for the right type of content - content lending itself to wiki type content management; you will obtain a very useful corporate wiki even if the number of contributors is relatively low.

Here is an example in a governmental legal office context (see the first bullet): http://tinyurl.com/rwoeh

Posted by: Patrick Cormier at April 27, 2006 08:24 PM

Half-right as usual, Nick :-)

It's true that skills and knowledge and numbskulls are not evenly distributed throughout an organization, or the Internet for that matter -- and that more people like to consume rather than interact or produce.

But you still seem to be leaning toward the model where some elite has to do the selecting of who is the smartest or the best at a particular thing, or has some particular knowledge. The best thing about a model like Wikipedia is that it is essentially self-selecting, or at least almost self-correcting. That's a lot better than the management or elite-chosen models I've had experience with.

Posted by: Mathew Ingram at April 27, 2006 08:39 PM


But you still seem to be leaning toward the model where some elite has to do the selecting of who is the smartest or the best at a particular thing

Sorry, but I don't believe I said or implied anything of the sort.


Posted by: Nick Carr at April 27, 2006 09:34 PM

It's an endless struggle by a few talented contributors to clean up the mess left by the numbskull horde

This is only true if you assume that someone can't be both a talented contributor and a numbskull at the same time. Wiki's, and other social tools, capitalize on a variety of expertise.

Although the World War II naval battle expert may not have the best grammar he could contribute technical details that the talented copy writer might not know. Some 'talented contributors' do editing and cleanup, and some work on technical details.

It's this diversity and breadth of knowledge that makes the claim 5,000 smart people will create a better wiki than a million numbskulls near fantasy. In reality, a million numbskulls will have expertise in some domain.

Your point is best taken for collaborative tools that require domain knowledge. A Victorian Literature wiki needs domain experts to get off the ground - in the case the elites aren't interested, hordes of people who haven't read any Victorian Literature couldn't help.

Posted by: Gordon at April 27, 2006 11:36 PM

It will be interesting to see whether Enterprise2.0 collaboration tools really take off. Just throwing them at the workforce probably won't work any more than throwing CRM or ERP tools did in the past. As the old saying goes, Old Organisation + New Technology = Expensive Old Organisation (OO+NT=EOO).

Successful adoption is likely to be driven by the usual three support cycles involved in effective change: achieving personal benefits from using them, seeing peers achieving the same benefits and continuous management support over the 24-36 months required to embed them in business as usual.

Seeding the change is more difficult. My experience at PwC with similar non-Enterprise2.0 collaboration tools was as you suggest, in that a minority of typically very busy people have potentially the most to contribute. Some of these people will do so of their own free will (in the same spirit as within open source communities) but the majority may initially need incentivising to do so. But we should not underestimate how collaboration tools support reaching out to others with useful information by enabling "the power of weak ties". This was a critical factor in making the sanctioned (but not supported) collaboration tools at PwC far more effective at providing real-time insights than the official, expert-driven KM tools.

Keep up the good work challenging sloppy thinking.

Posted by: Graham Hill at April 28, 2006 06:47 AM

"The best thing about a model like Wikipedia is that it is essentially self-selecting, or at least almost self-correcting."

Wikipedia doesn't select on quality of contribution. Wikipedia selects on quantity. Hence you too are only half-right: Yes it self-corrects, but it self-corrects towards mediocrity, away from being shit and away from being excellent too.

Don't believe me? See this article: http://www.zephoria.org/thoughts/archives/2006/04/15/on_being_notabl.html where people with expertise in a field were told their opinions were irrelevant because they weren't "Wikipedians".

Posted by: George at April 28, 2006 07:02 AM

I believe that one of the most interesting things with wikies is how they can bring together a large number of knowledgeable people in a specific field. It is probably not a technology that is meant to be "implement" in an organization (at least most of the ones I have worked in). I am freelancing today :)

Posted by: Klas at April 28, 2006 08:29 AM

Umair haque posted an interesting rebuttal to this on bubble generation Definitely a clear and concise argument worth looking at. Thanks for the original thoughts guys.

Posted by: Amar Rama at April 28, 2006 11:11 AM


If you weren't implying that someone has to do the sorting in order to weed out the numbskulls and get the people who know what they're talking about to contribute -- particularly in the enterprise -- then I guess I fail to see what your point was. That most people are numbskulls?


Posted by: Mathew Ingram at April 28, 2006 12:22 PM

How much of wikipedia's popularity is based on quality, and how much on ease of access? The "average person" learns he/she can do the following for any topic: [step 1] bookmark wikipedia [step 2] type in search statement, [step 3] click search button.
One could argue that this "ease of use" scenario outshadows trust or belief in the expertness of the creators; maybe the majority of wikipedia users generate search statements where wikipedia content is "good enough." If that's the case, this argument about expert's versus numbskulls in the creation of Web 2.0 content is somewhat academic. Haven't we all be in situations where the concept of "good enough" was a key factor in fundng -- or not funding -- some sort of business or IT initiative?

Posted by: Dennis D. McDonald at April 28, 2006 12:23 PM

It's actually even worse than all that... the work of the hypothetical 5000 smart people in this environment can be undone or crippled by even one numbskull in their midst. Look at all the trouble Wikipedia has had with a handful of people abusing it to their own ends. Any bets on how much productive time it takes away from the smart people to go back, weed out, and fix that? The fact that these things are self-correcting somehow glosses over the fact that it takes more time to correct than to simply get it right the first time.

But as a couple of other people pointed out, this isn't really terribly different from how things work now in most businesses, or the rest of the world, for that matter. And there may be some value in having the means of identifying the numbskulls more quickly in the Web 2.0 world than most businesses can now.

Posted by: Scott Wilson at April 28, 2006 12:27 PM


My point is that the success of Web 2.0 platforms, like previous knowledge management platforms, hinges on getting the employees with the most valuable knowledge to participate. Companies can (and do) use various techniques to achieve this goal. They don't have to have some organizational elite do the choosing about who can and can't participate - and I certainly wouldn't recommend that approach.


Posted by: Nick Carr at April 28, 2006 12:39 PM

Thanks for the clarification, Nick.

Posted by: Mathew Ingram at April 28, 2006 01:51 PM

Hi Nick,

I think your last comment helped to clarify what you are saying tremendously, although I would maybe challenge the word "most" before "valuable" but I can live with it.

But your earlier comment about blogs surprised me. You said:

Tony, I don't actually see individual blogs as being examples of platforms for participative content production. I see them as the fulfillment of the promise of the vanity press. Nick

Posted by: Nick Carr at April 27, 2006 04:21 PM

Maybe you are hinging your statement on "participative", but I would say that exchanges such as we are participating in suggests that even as bad as blogs are in terms of supporting dialog, they do foster the exchange of ideas.

The "vanity press" thing surprises me. While certainly, there needs to be some level of ego (feeling of value to your content) associated with any blog posting, the fact that in niche discussions and inside of corporations you get such interesting discussions happening makes it seem like something other than the "vanity press." I think I must have missed your point on this.

Posted by: Tony Karrer at April 28, 2006 03:00 PM

The best real-world usage of wikis I've been a part of is documenting a repeatable process so that one individual wasn't the only guy who knew how to do 'that thing he does' in the entire company.

I'm sure we've all encountered knowledge hoarders who like to feel important by being the only person who can get a particular task done in an organization. These types would not thrive in a pro-wiki, knowledge sharing org.

I've also seen wikis used by company leaders to draft out strategic visions like business plans, marketing strategies, etc.

Posted by: Shanti Braford at April 28, 2006 03:44 PM

Nick, my only disagreement with you:

The enterprise has an advantage over the public web site - they have hiring / firing power.

I guarantee you - in an enterprise setting with a properly aligned incentive system (compensation, recognition, etc.) supporting their web2.0 software - the numbskull ratio will go way down.

And those who are truly numbskulls will have a big incentive to disguise that fact.

Posted by: lawrence coburn at April 28, 2006 08:50 PM

I'll take your hypothesis one further: beyond the "talent" of contributors is the inherent communal "agenda, POV, or politics" of the contributor pool.

If this small group is always creating the internet's content... then the value of user-generated content diminishes even further, creating The Bloggosphere Hillbilly Effect.

Posted by: Kevin Perkins [TypeKey Profile Page] at April 30, 2006 12:40 PM

Post a comment

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Remember me?

carrshot5.jpg Subscribe to Rough Type

Now in paperback:
shallowspbk2.jpg Pulitzer Prize Finalist

"Riveting" -San Francisco Chronicle

"Rewarding" -Financial Times

"Revelatory" -Booklist

Order from Amazon

Visit The Shallows site

The Cloud, demystified: bigswitchcover2thumb.jpg "Future Shock for the web-apps era" -Fast Company

"Ominously prescient" -Kirkus Reviews

"Riveting stuff" -New York Post

Order from Amazon

Visit Big Switch site

Greatest hits

The amorality of Web 2.0

Twitter dot dash

The engine of serendipity

The editor and the crowd

Avatars consume as much electricity as Brazilians

The great unread

The love song of J. Alfred Prufrock's avatar

Flight of the wingless coffin fly

Sharecropping the long tail

The social graft

Steve's devices

MySpace's vacancy

The dingo stole my avatar

Excuse me while I blog

Other writing

Is Google Making Us Stupid?

The ignorance of crowds

The recorded life

The end of corporate computing

IT doesn't matter

The parasitic blogger

The sixth force



The limits of computers: Order from Amazon

Visit book site

Rough Type is:

Written and published by
Nicholas Carr

Designed by

JavaScript must be enabled to display this email address.