hyc@math.lsa.umich.edu (Howard Chu) (10/19/90)
A lot of folks have been paying a lot of attention to Halio's article, even though the observations she makes were not made in a scientifically controlled setting. (Of course, it's really hard to even try to set up a good experiment for this. Get volunteers - where do you look? what media do you use to present the experiment? For an experiment that wants to make a broad statement about a general population, getting volunteers, and selecting which to employ, is probably 80% of the experiment. Statistically speaking, how many people from how many different working/living environments do you need to "test" to get a good representation of the population at large?) Even assuming you can somehow select the ideal mix of volunteers, what can you prove - that Mac users are poor writers, or that poor writers like to use Macs? It's all such hazy stuff.... And yet, people are spending real money to do research on just these questions. I admit to being curious about what the real answer may be, but I don't think it's the sort of question that can be answered to anyone's satisfaction.... -- -- Howard Chu @ University of Michigan one million data bits stored on a chip, one million bits per chip if one of those data bits happens to flip, one million data bits stored on the chip...