| Sunday, July 18, 2004 |  |
| Connected Information |
 I occasionally wake up sweating, thinking I've forgotten to do my paper route, and undelivered newspapers are stacking up somewhere. When awake, the feeling gets replaced with a vague feeling that there's something somewhere that I've forgotten, and it is important. Of course I haven't had a paper route for, oh, close to 30 years, so it probably isn't that. Except for that I'm not totally sure if I handed in my resignation, so maybe there really is 20 tons of bundles of old newspapers standing there, and a lot of angry customers waiting for their morning paper. How do I know?
But the feeling is quite close to how I relate to many kinds of information. You know, there's something I should know about, and that I should be doing something about, but currently it is disconnected from what I actually do, and I don't quite remember. But it might come back and bite me at any moment, at which time the *information* will clearly show that I'm an idiot who's not doing what I'm supposed to. Until then, the information might well be hiding somewhere where I'm not looking. So, where should I look to find all the information I should be aware of? Paying bills is a bit like that for me. Relations with public authorities certainly is. I have the sneaking suspicion there's probably some big things I should be doing that I'm not, like paying taxes or having the proper licences, permits and immigration papers. I have some vague ideas. But, I look around me, and there's nothing in my environment that tells me much about any of those things. It takes some hard work to go and track down the information I'm supposed to have, and then figuring out what to do with it. And I never know if I got the whole picture. Yet, there's no excuse for not knowing the law. Meaning, there are 10s of thousands of pieces of information I really should know about, many of which could put me in jail if I didn't act correctly in accordance with them. Oh, most other people don't know them either, but they somehow manage to get the most vital pieces, and then they look around themselves and see what other people do, and they figure they can't go all wrong if they do roughly the same. My point is that information is typically something disconnected and abstract. Data certainly is. It is some kind of extract representation of something, which possibly might be real. Information is when we put some of that data together so it seems to say something coherent. I get a bill in the mail. That's a datum, or maybe several. If I study it and discover that, aha, I'm supposed to pay this before the 10th of next month, and this is the amount, then that is information. At that point it is still unconnected to everything else. Who says I'll remember that next month, or that I have any money at that time? Where do I put this thing so I find it again? Well, if I really think it is terribly important, I make some kind of system to remind me. I make a folder for that creditor, I write it in the calendar, I try to estimate my budget over the next month. Of course, you say, everybody does that. But it is not that *of course*. It is just that you're used to living in an information world where things like bills and deadlines appear to be real. Even though they're disjoined and often inconsistent abstractions, several levels removed from anything real. If you call up Microsoft for tech support, based on some odd problem you have with Windows, they'll usually end up telling you that the answer is there, plain and clear, in technical note number 17536, and you're a bit of an idiot because you didn't know that. They have their information asses covered, just like the law. It is your fault you don't know. To be fair, most open source mailing lists work like that too. You ask a question, and there will always be somebody quite willing to tell you how stupid you are because there's some file somewhere that has the answer on line 514. And you didn't know. It is a bit of crime to not have processed and tabulated all the information in the world. In a society like the french, often things aren't said explicitly, but it is assumed that you'll somehow know. The school doesn't necessarily give you a piece of paper that says when vacation is over, because it is assumed that everybody knows. If you apply for being admitted into a new school, they might not even bother to tell you if it was accepted or not, because you somehow ought to figure that out by yourself. The business world works like that too in many ways. You're sort of supposed to guess what is going on, even though the information isn't readily available, and your boss isn't talking. You've gotta be well networked, drawing on an extensive intelligence network when you do anything. You're supposed to know things, but there might not be anybody who tells you exactly how. And the available data itself might not help much. There's an implicitly idea somewhere that data adds up to information. Which potentially might be structured into knowledge. And if you then really internalize it, it might become wisdom. It is a questionable model. It assumes that the direction is from disjoined snippets of data towards something more integrated and useful. Who says it works like that? Nature doesn't seem to me to work like that. There nothing quite equivalent to data out there. Nature includes lots of systems that have partipating elements that send messages to each other. A plant or animal that is trying to procreate often sends out millions of little seeds or pollen or eggs. And there are millions of ways they might get activated. DNA is certainly information, but it is replicated billions of times. There are billions of ways that DNA might hook up and produce the next generation. Billions of signals with certain receptors might be met with billions of possible counter-parts that have matching receptors. It certainly doesn't depend on one little piece of information hidden once somewhere, which somebody has to remember to go look for. That is where our informational systems tend to go wrong. We put something in some suitable place, and then one is just supposed to know where to find it. And, oh, one can make all sorts of reminders that makes it easier. Like, if on the web a piece of information is stored in some place, other sites can link to it, and people can make bookmarks, and they can write a little note for themselves to remember where it was. And you can go search in a search engine. And that helps, and somehow most things work out. But it still seems vastly inferior in some way to the relatively effortless manner information is used in the natural world. Our systems depend on somebody remembering what to look for, at the right time, and discovering the right context. It is very fragile. Just like our technology, which is built on similar models. One little wire is disconnected or one little comma in the wrong place in a program, and the car doesn't start or the space probe misses Saturn. Our constructions often have many single points of failure. Nature generally doesn't. We can learn, I'm sure, from nature, how to help things link up more often, and more reliably. Of course what I really want is for the information I need to be available exactly when I need it. I want the information to come to me at the right time, and I want to not have to analyze large amounts of data to find out what is going on. I want to be reminded of my bill at a reasonable time to pay it, and I want the expectation of having to pay it to be wired to my budget somehow. I would want to know of course, the moment I try to buy something else frivolous that it would mean that I couldn't pay that bill next week. Well, really I'd rather not have a bill. I'd rather that my economic metabolism took care of the needed energy exchanges, continuously. A lot of our information use carries with it the phenomenon of "Now the cake should have been in the oven for 1/2 hour". You know, you get to a certain point in the recipe and it tells you that there's something you should have done at an earlier stage, but you weren't aware of it at that stage. I might put a reminder in my calendar to pay a bill, and when it rings, I will remember that I should have made sure I had money to pay it. Which I might not have done. Disjoined information doesn't necessarily add up. There's no guarantee it adds up to meaningful information and useful knowledge. Just because you spread a lot of screws and other mechanical parts around on the floor there's no guarantee you can construct a refrigerator from it. If I started with a working refrigerator and dismantled it, then I likely could. But independently aggregated bits and pieces, no. And most of our information is like that. It doesn't come from a dismantled whole, but rather from unrelated bits and pieces, which you're then supposed to put together into a whole. Which is next to impossible. We are pretty good at managing anyway. Despite that we live in a sea of disjoined information, we're pretty good at creating some kind of network between them and making useful things happen. We do it in a rather primitive low-tech kind of way, maybe. Most countries have an enormous and ridiculously complicated system of law. Most people, including the politicians who wrote the laws, and the lawyers and judges who interpret them, do not at all have an overview of what it adds up to, and many laws contradict each other. So, for normal people the answer is usually to look around you and guess at what the laws seem to be. Stop for red lights, don't steal other people's stuff, etc. And the lawyers will just search for laws that support a particular point when the need arises. We find a pragmatic way of dealing with it, in the face of the impossibility of knowing what the laws actually say in total. So what do we really know, if the pieces are scattered all over, and we're just barely coping? "In what sense is a thing known if five hundred people each know one constituent of it and nobody knows the whole? Or again; what if this truth has a thousand constituents and half of them are not known to anyone, but only stored in libraries? What if all of them only exist in libraries? Is it enough that somebody knows how to look them up if they should ever be needed? Indeed is it enough that this person should have access to a system which will look them up? Does the enquirer even have to understand the questions which these truths answer?" - Mary Midgley (Wisdom, Information and Wonder: what is knowledge for? 1989)Is that really the best we can do? Can't it all connect better? Well, one possibility is a structured semantic web. If all information is meticulously categorized and related with all other information. Possibly in some huge all-encompassing hierarchy. I don't know how likely or possible or even desirable that is. Another possibility is making everything easier to find, and to constantly look for matches for everything. That's more like nature's way, I think. You put everything that needs to be remembered out in loads of redundant copies. And then loads of little pieces are constantly looking for matches to what they're looking for. You know, a Synchronicity Engine of some kind. There's still some major key missing, though. We need a paradigm shift that takes us from the overwhelming complexity of scattered information to a world where things might again be simple, but at a new level. You know, you're hungry, there's an apple on the tree in front of you, so you eat the apple, and you feel good. That kind of simplicity. You're tired and you sleep. But while at the same time being globally connected with a vastly bigger network of people and information. Rediscover the simplicity in a higher order of complexity. I have no doubt that it is there. And if we don't find it, it is probably because we still address information complexity the wrong way. [ Knowledge | 2004-07-18 10:15 | 13 comments | PermaLink ] More > |