How do we know stuff, and how do we know what we know? Theoretical questions like this have plagued philosophers for centuries. More recently, such questions have furrowed the brow of many a computer scientist (especially, I suppose, in the field of artificial intelligence). Computers are wonderful at combing through databases and extracting data that meets certain defined criteria, and they do it so fast that humans feel deficient and threatened. And yet, computers don't know what they know.
Classic example. Ask a person if they know who was the eighteenth president of the United States. Most people will think for a second and say "I don't know." They don't search their entire database of knowledge before answering. They don't have to. That's because they know what they don't know. Computers, obviously, are different. And so that raises all kinds of questions about how knowledge can be represented, or described, for computers.
So the tantalizing question of knowledge has moved from plaguing the ancient philosphers to bothering the AI people at MIT. Now it's bothering me, even though I am merely a lawyer.
As I have said, I think that a large part of the practice of law involves "processing information". But I'm most definitely not the only lawyer who is spouting such notions; Richard Susskind has written extensively about the idea that lawyers will inevitably embrace technology to process the vast amounts of information with which they are confronted. He focuses on so-called "knowledge management" as a critical component in the practice of law. Most lawyers right now have no idea what that concept means. I think I'm one of them.
I guess what I mean is that I don't know what the concept of knowledge management means in the abstract, but I have a longing sense of what it should mean--or more precisely--what sorts of practical problems it should address.
I work in a law firm that has offices in three cities, and there are about 40 lawyers in our firm. We have a network and we have a document managment system so that we can retrieve (theoretically) any document that anyone who works in any of the firm's offices has ever created. Sounds good. And mostly it is. But that's not a "knowledge management" system. Or, if it is, then it's a crude one. It only works efficiently to the extent that we have created "fields" in our document mangagement database that can be used to filter out information that the searcher doesn't want. You can search for documents based on whether a certain word is present in the document, but now that our database of documents has grown to a collossal size, it is rarely efficient to search that way. We could "add fields" to the document profile sheet, but then people would get bogged down indexing new documents. There's got to be an easier way.
"Knowledge Management" to me suggests an approach that not only creates a formal system for categorizing knowledge (i.e. setting up a database that indexes documents and forces the user to fill in certain fields), but also one that implements a system that is dynamic to catch the little bits of knowledge that are in the minds (and hard drives) of the people who work in an organization.
Since I have been using Radio and doing this blog, I have begun to wonder if this is the answer. I don't fully understand the Manila thing (vs. the Radio thing, or the Frontier thing). But I gather that Manila allows people within an organization to "publish to the web" (be it an extranet, or an intranet, or wherever) and thereby put the knowledge that they have out there where it can be accessed by others. Seemingly, the key is for the workers to be able to put the information out there in a central place. It's like an AP "news feed" that others in the organization can "subscribe to" (which I guess would be done via "the news aggregator"). And that's true. The centrality of the information, and its being accessible to all is a good thing. But that's not the real key.
The key to this concept (which I am doing a poor job of describing, especially to anyone that is not blogging through Radio and using the "news aggregator") is not centrality, but immediacy. If you could get workers to 'blog their ideas everyday then the information that is fed to the central place would always be updated. And because people could be confident that it was updated they would feel drawn to add their own knowledge. So, ideally, the information would be constantly refined. Clearly, this is possible with one of these programs (Radio or Manila)
For those of you at home who don't have these programs (and therefore have no idea what I'm talking about) here is a concrete example. I post stuff on this blog (so far mostly useless meanderings, and maybe this is another one). But let's say I work in an office, and me and all my co-workers write blogs, but only about work related stuff (okay, let's be realistic, mostly work-related stuff). After awhile, it is recognized that I know stuff about how to use computers in the law so my blogs tend to be about that stuff. Obviously, no one reads my blogs (nor should they).
But, an industrious paralegal named Connie goes to court everyday to file things, and she is always learning the newest filing requirements because she is there in court everyday and learns first hand all the lastest court gossip (i.e. they're getting picky about the font size requirement for pleadings). Her blog puts that information out there. And so when people are going to file something they check her blog first to make sure their pleading complies. Or Connie keeps an updated list of court phone numbers on her blog site. Then she adds the updated list of operating hours for the local courts. A couple of months go by and the office manager decides to have someone comb through Connie's blog and "data-mine" the information that is in there that can be formally set out and put on the firm's intratnet site and to assign Connie the task of keeping it updated.
Great! But I can hear a nay-sayer somewhere in the back (maybe his name is John C. Dvorak) saying but Connie could just E-mail everyone the same information. Well John, that's true, but she won't. And if she did, most people would delete the E-mail because the information wasn't useful to them at the time they received it. And when they did need it they wouldn't go in their E-mail trash to retrieve it because it was buried in their along with about five thousand other emails.
No, this is clearly something worth exploring. I'm sorry I can't describe it better. Maybe some of you can post some comments to help me sort this out.