For about the nine squillionth time, I’ve been forwarded the link to the video “Did You Know?” (a.ka. “Shift Happens”). You’ve probably seen it. It sure does impress people. And it has a lot of impressive numbers in it.
I have to admit it’s at a bit of a disadvantage from the beginning with me due to the fact that “Did you know?” in any forward is almost a guarantee that I’m about to be told something that’s inaccurate. Also, I’m getting tired of the music, which has been used on another video since, and is now for me, at least, the leitmotif for hype and hand-waving. It is rather apposite, though, in that the words start with strong-sounding repetition of a straightforward phrase without actually specifying what is right here, right now, and then go into something that sounds like detailed content but is actually indecipherable.
The video does present a fair bit of interesting food for thought, and some of the numbers are quite impressive (if wanting more context), but it can be remarkably fluzzy and often not especially thoughtful in itself. (Nor are the data referenced, so they could just be making it up, or misquoting, getting it badly wrong…)
So I just wanted to pick on a few things in it, as a bit of an editorial fact-querying exercise:
It does not follow that just because India has more than four times as many people as the US it has more honors kids than the US has kids. Honors kids, first of all, are not just the kids with IQs in the top 25% (and seriously, did 5 out of every 20 kids in your school get honors?). There are many other factors, including an environment that encourages learning, and, for that matter, schools at all – the education rate is much lower in India.
What are the top 10 in-demand jobs in 2010? (This was posted in 2008, so how do they even know?) What does “in demand” mean? By whose scale? Is it jobs that people want to have, or jobs they want to hire people for? If, for instance, it’s jobs that need proportionately the most people to fill them, then it’s hardly surprising that they would be new positions; if you have an existing kind of job with, say, 100 people in a company, and in any year they have a turnover of 5 people, and a new position created to match a new need, and it has 10 positions, in the first year you’ll have to fill all 10, well, the new position is more in demand no matter how you look at it, eh? Which is not to say that’s what they mean by “in-demand.” But they don’t say, so how do we know what they mean?
So Bermuda is (was, at least) the #1 country in broadband internet penetration (no reason for caps on that phrase, by the way). So what? Bermuda is a small country. It’s easier to give it 100% penetration than it is to give any major North American metropolitan area the same.
Counting words in the English language is one of the biggest mug’s games in existence. Consider: make, makes, made, making… how many words? They’re all inflected forms of make. If dog and dogs are two words, are fish and fish? How about cleave and cleave, which mean opposite things? Is low-life one word or two (some people would write it as two: “He’s a low life”; others would write it as one: “He’s a lowlife”; different dictionaries include or don’t include different words different ways). And do you count technical terms? All of them? Even that chemical name that takes several pages to write? Where do you draw the line?
And then there’s their vague quantifying of information. Who has estimated that a week’s worth of the NYT contains more information than a person was likely to come across in a lifetime in the 18th century? What are they counting as information? A person in the 18th century would have had their eyes and ears open as many hours a day as we do, continuously receiving signals. They would have heard things from their friends about people they know and happenings of the day. Is that not information? Information is any received input that modifies existing awareness, not just lexicalized denotative statements about a set of topics considered sufficiently important. If you sit and read a week’s worth of the NYT from cover to cover, do you really think you have taken in more information than a person who lived and breathed for 60 years did in the totality of his life in, say, 1750? More details, more things that register in your cognition, more awareness of the world? More things you can act on? Just because it names more places, say, or more people, or has thousands of little ads? Please, guys, at least give some basis for your measure of “information.”
I mean, say I say, “There is a man with a blue dog at the door.” How many pieces of information is that? One? Three (man, dog, door)? Five (blue, happening now)? Six (you know that I know it)? More (you note the tone of my voice as I say it and that tells you something vague about my state of mind regarding the fact; you suspect the man is here to kill you, perhaps – oh, how do we count suspicions)? What if there’s more than one door? How do you compare the amount of information in “at the door” when you don’t know which door with “at the door” when there’s only one door? What if it turns out it’s a woman with a red dog? Was there less information in what I said first because it was incorrect? Don’t forget to count the fact that you now know that I couldn’t tell a woman from a man and that I may be colour blind. Or is that only information if you in fact deduce it? (Does Sherlock Holmes get more information from the same input than Watson does? If so, if a computer calculates something but no one ever looks at the calculation, is it information?)
And when they’re talking about exabytes of unique information, what is “unique information”? Do we mean to say that a 10-megabyte photograph of a moment at a kid’s birthday party is 1,000 times as much information as a ten-kilobyte text file with a description of the party? Who knows, it may be – but if it is, then looking out your window for a couple of minutes will give you much more information than reading the whole New York Times of the day will. But apparently they’re only counting information that is generated through, um, computers and publishing or something. I mean, I’m not denying that we have a hell of a lot of computing power now, and can process operations in seconds that would have taken months in previous times. But this kind of measure makes a couple of minutes on Facebook much more information than a half hour chatting with friends. Please, guys, information is a very poorly defined term here, especially if you’re trying to quantify it. Clarity!
I’m amused by the statement that for students starting four-year technical degrees, half the information they learn in their first year will be outdated by their third year. I’m amused because it’s my experience that in many fields, more than half the information you learn in intro classes is outdated at the time you learn it. A lot of what they teach you in intro linguistics is no longer thought to be completely true, but it forms a good introductory basis, for instance. Of course if you’re in a field where new things are being developed all the time, you’ll have to keep your knowledge fresh. Some fields – medicine, for instance – require continuing education to maintain your licence.
And what the heck are the computation capacities of the human brain? Computers can already do many operations far faster than your brain can. They also fail utterly to do some other operations that your brain can. The brain is not a binary-based computer. Computers are not brains; they are computational prostheses for the human brain. They do some things much better and are simply incapable of others. So what is the implication in real-world terms of the statement that a computer will be built by 2013 that exceeds the computational capacity of the human brain?
And what is the computational capacity of the entire human species? Are they talking about doing floating-point operations in your head? Computers can already do that far faster. Are they talking about parsing grammar? What are they talking about? And should I point out again that the human species has created computers as mental prostheses, and so every computer is an addition to human computational capacity, and therefore it is impossible for a computer to exceed the computational capacity of the species because its computational capacity is a subset of the capacity of the species?
I wonder if, when they count the number of songs downloaded illegally during the course of the presentation, they’re counting the presentation viewer’s downloading of Fatboy Slim’s “Right Here, Right Now,” which has been playing during the presentation without the viewer’s paying for it.
So what does it all mean? Well, one thing it means is that with enough flash and bang and big statements and cool graphics you can get people all impressed and they might not notice that you’re not giving them a whole lot in the way of usable information. (See, the video was five minutes of information, in one way, but at the end what more do you really know that has a real-world correspondence or use? Do we count it as information if it’s too vague to be viable? If I say “There are a million croopahs on the vargnal,” is that information? if I explain that a croopah is a moment and the vargnal is a plate, does that help? Et cetera.)
But do you feel all prepared for the future now? Oh goody. As William Gibson has said, “the future is already here. It’s just not very evenly distributed.”
Sesquiotica 