Slashdot Mirror
Do You Remember Bob?
GdoL writes: "Do you remember Bob? Byte's editor starts his monthly column talking about Bob the OS Interface from Microsoft in the middle 1990s. And he didn't forget either Bob the programming language from a former technical editor of Dr. Dobbs Journal, David Betz. This OO language is widely use on 'DVD players and set-top boxes produced by the likes of Toshiba, Samsung, and Motorola.' Do you remember any other language long forgotten that is still used in the real world?"
Prototype-based languages have an interesting minimalistic feel that allows language designers to get right to the heart of some issues, and I think they are very valuable for that reason. However, I think ultimately classes will win out.
One of the common mistakes in OO programming is to confuse kinds-of-things with instances-of-things. A program for a clothing store might have "shirt" objects with properties such as the manufacturer, size, colour, and quantity in stock. Seems ok so far. Then, when looking for a place to store information on a specific shirt, such as the customer who bought it, the programmer notices that a "shirt" has two meanings: a kind of shirt versus a specific individual shirt. The quantity-in-stock belongs to a shirt type, while the purchasing-customer belongs to a specific shirt instance. The design has to be juggled a bit to accomodate this new insight.
Combining classes and objects the way prototype-based languages do seems to encourage this kind of confusion on a massive scale. Programmers would undoubtedly add comments to indicate which objects are actually "meta-objects" (ie. classes). The Bob language examples do exactly this, as a matter of fact. I suspect that the degree to which they don't differentiate between classes and objects is exactly the degree to which they will have confusion between instances and kinds of things.
Incidentally, I think there is a great deal of promise in the idea that everything (including classes) is an object, but I don't see prototype-based languages as being a successor to class-based ones.
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
I had to read this book for a graduate Mass Comm class two years ago, and it is without a doubt the most awful excuse for experimental science that I have ever seen. Unfortunately for the authors, I had taken a class in research methods before I encountered their book (they should consider doing the same).
It's shoddy science, through and through. They ignore intervening variables, operate every experiment without controls, provide no accounting for intercoder reliability, the samples are always too small to be statistically significant (only one had more than 30 participants, many had less than ten), and comprised of forced participants (Reeves' and Nass's freshmen psych students at Stanford, to be precise. Even without a grade on the line, that's a bad sample). Usually, they rely on reported rather than observed behavior, and the only operating hypothesis ever examined is their goofy "equation" (you want me to spoil the beginning of the book for you? Here is The Media Equation: Media Equals Real Life. That's it. Word for word).
As if that wasn't enough, they make constant generalizations of their results (which with forced, nonrandom sampling is the first thing thrown out the window). They grandstand on every turn -- everything supports the Media Equation, and there is nothing it doesn't affect. You should always be suspicious when "scientists" do thirty experiments and always find their hypothesis supported exactly how they predicted. It's usually bunk, and in this case, it's a pantload. In one instance, they even admit to writing the hypothesis AFTER the experiment was performed. They make repeated references to other "research" in this area, but if you read through the bibliogrpahy, they are merely citing *themselves* from previous experiments. Many of these experiments, if you were interested, have still not been accepted for publication, many years after they were done. Most are not even available at the authors own Web sites. If it weren't for the fact that The Media Equation was published by Reeves & Nass's employer, I doubt whether they could've goten it published at all.
It's bad science, and it's only an afterthought: they plainly thought up their "equation" first, and then set out to prove it. That's the Scientific Method in reverse, people.
Let me give you one quick example in reference to the poster above: in the larger/smaller pictures experiment mentioned above, they show participants photographs of people's faces, some in close up and others standing 10 yards away. And the "test" is showing the subject another photo of the same people, and seeing which person the recognize most often. Guess what: it's the person who's face they saw in close-up. Surprised? You shouldn't be. No thinking adult would be; you see the face close up so you see more detail, and see it better. Plain and simple. But that's not the conclusion Reeves and Nass come up with; they decide instead that this turn of events means that you are having a psychological reaction to the face, and the biger face makes you happier because it seems more like a person. So if you think it's a person, you will remember it better.
That's the media equation, you see? The more person-like an electronic communication is, the better it works. The only time this has been tried under real-world circumstances, of course, is their grand experiment: Microsoft Bob. Funny how well that went over, huh.
Man, I hate that book.
Nate
PS - you might try the following link to Amazon, I submitted the review under "n8willis": here.
PPS - if anybody cares, I'll follow up what I say by emailing them the paper I had to write about this ridiculously bad book; it goes into more detail. Or perhaps I'll submit it as a /. book review. I meant to at the time, but it was just way too long....
-- Watch the REAL Jon Katz.