Tom Davenport is a smart guy.  I knew him slightly in the days when I produced a corporate magazine at Ernst & Young and have a lot of respect for his writing and his understanding of knowledge management strategies and technologies. 

In one of those gotcha responses that editors love so much, Optimize magazine had Davenport respond to a podcast interview with Andrew McAfee and J.P. Rangaswami about Enterprise 2.0.   Tom is skeptical of the organizational impact of Enterprise 2.0 and while most of what he says is true, it is also incomplete and misleading. 

Blogs, he writes, ”…have a tragic flaw: No one has the time to read them. As of the end of July, there were something like 70 million blogs in existence. Only a very small fraction is read by anyone other than their creators. When someone comments to my own blog, it’s almost always a spammer. And maybe it’s a good thing we don’t read all these blogs; if we did, they might bring down our economy.”

True, but irrelevant to the question of whether they are useful to enterprises.  Yes, there are too many blogs and not enough time to read them.  It is also true, as he implies elsewhere in his response, that some Wikipedia articles are factually suspect.  And as I’ve written here and here, there are plenty of reasons most CEOs shouldn’t blog and shouldn’t allow just anybody who wants to blog about the company to the outside world do so.  The chances of democracy breaking out in large corporations is roughly the same as they are for the same thing happening in Iraq.

Yet we know that in the enterprise context blogs are enormously valuable ways for teams of dispersed experts to stay in touch with what their colleagues are working on and thinking about.  We even know that the CIA has a private blogging network.  I can’t think of a large organization that wouldn’t benefit from a secure exchange of regular peer-to-peer blog-style information at virtually every level of the enterprise. Throw in the capacity for comments and you have instant interactive dialogue and a valuable learning experience.  It won’t be an everybody-talks-to-everybody thing; the key words are “peer-to-peer.” 

One of the misunderstandings that Davenport and other skeptics seem to have is that social media in enterprises will be just like social media on the public web.  For example, he says of Wikipedia that he doesn’t use it that much because of “a nagging doubt about who contributed the knowledge and what their agendas might be.”  Fair enough.  But an enterprise wiki-based collaboration space devoted to a specific information-gathering project wouldn’t be accessible to everybody; just the people who are working on the project, which could also include suppliers, customers and other “trusted” individuals. 

There is a huge difference between the trustworthiness of an article or links assembled by “sleepyhead28″ for Wikipedia and an article contributed by the chief researcher of an enterprise’s own Princeton lab.  Enterprise wikis will clearly have much clearer links to real-world identities and reputations.

Enterprise wikis offer a painless (because the process is “social”) and efficient way for work teams and other communities of practice within organizations to gather and share information for all kinds of collaborative projects and to leave behind not only valuable and easily stored digial repositories of knowledge, but a complete record of the processes that created it.  

Tom’s major point appears to be that nobody has time to read all this stuff so why bother to collect it.  At his own blog (last updated on June 6, which should have been a clue to the Optimize editors that he wasn’t really into social media), Tom writes: 

We’re finally getting a handle on how to get value from structured information. But most organizations don’t have a clue about how to mine blogs, emails, instant messages, presentations, and so forth. We don’t have time to look at all of this stuff to see if it’s interesting and relevant to us, so we will have to have systems that find the good content and serve it up or summarize it for us. It is individuals that read and take action on unstructured information, so we need to address this issue at the individual level. 

Off the top of my head I can think of two companies that offer software that can do this right now and do it very well.  KnowNow can provide information that users identify as important to them in real-time from virtually any source to which they have access.  And System One refines your search in real-time as you add new information to your project.  

Come to think of it, Google Desktop is a great little tool for searching unstructured information and it has the attractive advantage of being free.  

Tom is dead wrong when he writes that social media will “…live largely on the margins of organizations because they don’t fit their mainstream needs.”

Social media will start at the grassroots level–marketing, communications, research–and gradually insinuate themselves into the fabric of large organizations as they prove their usefulness and top executives learn that they are valuable and can be controlled.  Sure, there will be resistance from the top but it will fade over time. There is a whole new generation of executives coming up who grew up on the internet.  Social media will be as familiar to them as spreadsheets are to today’s generation. 

The downstream result will be far more–not less–control over an organization’s collective knowledge.  Think of the enormous value that preserving a digital record of particular projects would have in the future as new people come into an organization and pick up where their predecessors left off.  I have a vague recollection of talking to Tom about 15 years ago about the knowledge that is lost when experienced people leave an organization.  That was the problem that knowledge management was supposed to solve.  But, it didn’t because the tools for capturing and retrieving that valuable information didn’t exist then.

They do now, and it surprises me a bit to find that Tom apparently doesn’t see a big future for the best and most compelling tools for knowledge management yet devised.