The Semantic Advantage

December 7, 2009

Resisting the hive mentality

Filed under: knowledge work — Phil Murray @ 6:53 pm

We certainly need better ways to find expertise in organizations, but we should carefully consider the implications of HiveMind and other technologies that look at the surface manifestations of behaviors rather than at the actors and activities themselves.

From the BinaryPlex website:

We’re building a product called HiveMind that helps you know what knowledge and expertise the people in your organization are demonstrating, without them needing to update a manual profiling system. Our philosophy is to manage information on behalf of people instead of adding to the flood. We call this “People Centric Software” [emphasis added].

NO, it’s not people-centric. It’s information-centric. It does what it does based on information, in the same way that you can learn things about bees and ants by converting their movements and interactions into information and interpreting that information.

But people aren’t bees or ants. The core problem, for both managers and individuals knowledge workers, is that the knowledge-based organization quite literally does not know what its members are doing. It does not know (or have a record of) who is engaged in what activities with what tools. It does not have an accounting of the inputs and outputs of those activities. It does not track — except in formally organized projects or processes — who or what processes are the beneficiaries of those activities. This is an astonishing reality accepted as par for the course, a level of ignorance that would be considered grounds for immediate dismissal in a manufacturing environment.

An information-driven tool is a poor solution for this problem. Consider the following:

  • A solution like HiveMind replaces analysis of work (what people actually do on a daily basis) with guessing games. That doesn’t seem like a great organizational policy, especially when it is possible to know what they actually do. A well-conceived analysis of work activities does not have to be intrusive, time-consuming, or static. It can be helpful to individuals themselves, to managers, and to the organization.
  • The HiveMind solution asserts a top-down association between language (vocabulary) and skills or job roles. I’m sorry, but pairing a language-based solution — even one supported by well-designed ontologies — with static and/or highly conventional descriptions of work activities is going to produce only a marginal advantage. What’s more, many (maybe most) productive work activities occur at a finer level of granularity or specificity than “skill/expertise” or “job description.”
  • Like other applications that skim large amounts of information for certain kinds of facts or for consumer sentiment, an algorithmic analysis of information about skills is a “derivative” instrument. As we have seen in the marketplace, derivatives are often accorded the highest value, even though the information on which they are based is farthest from reality. And we know what happens when derivatives drive mind share and management. This is not a stretched metaphor. Meaning connected to reality is the source of value in knowledge-based organizations.
  • The worst possible situation is one in which a solution seems to make sense, especially when it grabs the imagination but is actually deeply wrong-headed or distracting. I think this one falls into the latter category.
  • All practices and technologies that create, consume, or process information — especially the language we use to communicate meaning — ultimately have a deep impact on how we work. You have to work out the implications before you adopt those practices and technologies.

Do you really want to encourage a hive mentality? A hive perspective? Do you really believe we behave like bees? That our individual acts have value only when they are summed? [paranoia alert] I believe that just the opposite is true, but there are people who want us to believe that because they know how to steal some of that value from us.



  1. i’d be curious for your thoughts on this idea; any feedback would be appreciated!

    – @venessamiemis

    Comment by Venessa Miemis — March 12, 2010 @ 9:00 pm | Reply

    • Excellent thinking enhanced by elegant writing. I don’t have time today to read and respond in detail, but I will try to do that.

      I was, however, struck by one thought in particular:

      “I think the web, in it’s nowness, has tricked us into a constant state of reaction.”

      My most recent article — see…-really-…-Putting-meaning-to-work-61174.aspx — is a somewhat different interpretation of the impact of this state of reaction. Thanks for reading and commenting.

      — Phil

      Comment by Phil Murray — March 12, 2010 @ 10:01 pm | Reply

  2. smart piece, thanks for sharing.

    some more fodder for the grist mill: just came across this piece of research via twitter and pondering the implications for ‘ambient findability’ – You Are Who u Know: Inferring User Profiles in Online Social Networks

    Comment by Venessa Miemis — March 13, 2010 @ 3:33 am | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at

%d bloggers like this: