Month: October 2001

Mark Pilgrim (in response to my Radio as knowledge management tool post yesterday):  Most companies and employees don't work like that.  (I have added my notes in situ)

I agree with everything you say, in theory.  I'll bet the solution you outline works really well at Userland: a small, tight-knit, content-centric company.  I agree that a desktop-based solution is the only thing that can technically scale well to a larger enterprise.  But I'm afraid you're ignoring the biggest problem, which is that it won't socially scale well.  (The biggest behavior change is to get people to post to a weblog instead of giving status reports by e-mail.  Answer the questions:  what are you doing, why are you doing it, is it working, what do you think about it, where can you find it, what did I change in this document, etc.)

Problem #1: most people are not passionate about their work.  They won't voluntarily spend their own time writing a corporate weblog, and if the company forces them into it, then you'll just get a bunch of crap as people try to fill their weblog quotas.  (Most knowledge workers hide at work. Weblogs make it easy to see who is doing meaningful stuff.  Frankly, I think Jack Welch of GE was right = 5-10% of people should be let go every year in a big organization.  Knowledge management is a way to unearth the value and the fluff.)

Problem #2: most companies don't see the value of having people document anything, much less their daily thoughts.  Mostly this is an ROI problem (or a perceived one).  Writing good documentation is hard; writing a weblog that is worth the company time it takes to write it (remember, most people won't write on their own time to benefit the company) is also hard.  (Some people will be better at it than others.  Those are the people that will be at the nodes of the knowledge network = the most popular Weblogs.  It works the same today but isn't visible.  They are the ones that employees talk to.  Weblogs are a  a way to automate, route, and archive their wisdom.)

Problem #3: most people are not good writers.  They don't enjoy writing, it intimidates them, it's difficult and frustrating for a variety of reasons.  English may not be their first language.  This last problem is especially bad in development shops (at least the ones I've worked in), but it's a problem everywhere. (Writing is a skill.  It takes work and time.  However, a good technical blog may be code examples and little text.  There are ways to communicate value without being Norman Mailer.  Writing skills are useful in other places too…)

Problem #4: people only have so much writing in them every day.  Even the best writers have practical upper limits of how much good stuff they can write before they need to recharge.  If somebody is a good writer, they're probably not going to be using that energy for the benefit of the company; they probably have their own weblog out there that talks about stuff they really care about, or some other creative project outside the company's control.  (The more verbose a blog is, the less valuable it is.  Studies have shown that people typically don't read on the Web, rather, they scan.  Short two sentence posts work well.  For example:  I just talked to customer A and they are upset with feature B.  Here's why  x, y, z.  Better yet:  here is what I did to get them over it:  n, m, o)

So you're looking for people who are (a) passionate about their work, (b)good at writing, (c) not spending their daily creative efforts elsewhere, and (d) in a company that values those assets enough to let them do it on company time.  I'm sure that Userland hires only these kinds of people, and I'm sure Userland is the kind of company that encourages you.  Let me tell you: you're the exception, both at the individual and corporate level.

My company has nothing like what you're suggesting; they're just one of those companies that sits around bemoaning the fact that “all that knowledge just walked out the door” every time somebody quits.  We have a static intranet that nobody uses, except for installation instructions or new employee setup.  To our credit, one of my co-workers read Scoble's site and saw his article on whether intranets suck, and got inspired.  We were considering putting something simple in place, but it quickly got derailed when my boss insisted on using Zope.   (Yikes!  Talk about unnecessary complication.  Manila or Radio are the tools to use.  What most people don't realize is that open source almost always costs much more and is more difficult than inexpensive, high quality, commercial software.)