home .. forth .. colorforth mail list archive ..

Re: [colorforth] merging edit time and run time


Albert van der Horst wrote:
On Mon, May 23, 2005 at 02:14:02PM -0400, Mark Slicker wrote:

In some of the cases above, I don't think there is a user illusion in the
light switch if it is made clear to the user that the effect of pushing
the button is to send a message to a computer to requiest that the light
be turned on or off. If they are instead trying to pass it off as a
coventional light switch, I think there is an illusion going on here. In
most cases it might apear to be a switch, if something goes wrong, the
problem may have nothing to do with operation of the light, the button,
and the connections between them.

Lots of people have the concept of electric circuits, where a switch opens or closes the circuit. Because they have that concept, some of them will assume that every switch opens or closes an electric circuit.

This is an assumption that comes from what they already know. Maybe the amish or people from mali wouldn't have that assumption.

But in the above case the car manufacturer would not change the
manual of the car. Lever in this position: light on.
Lever in that position: light off. This would not change after
introducing the CAN bus. Most car owners would hardly understand
if there is a remark there about computers and not conventional
anyway. The section about trouble shooting in the manual probably
changes. (No more: look at fuse X if the light doesn't come up.)
I can't see much wrong with this state of affairs.
There are a few things each car user should know. Use the right
fuel. Keeps the tires inflated. Etc. Rightly these are minimized.
Rightly these are stressed. But why tell him more? Unless he is
a Forther, I guess ...

I'm certainly interested in your opinion whether this is bad?
And if so,  what can be done about it?

What can be done about the assumptions that random users have based on what they already know? What should be done?

A couple months ago I got stopped by the police because both of my rear brakelights were out, leaving only what the policeman called the "idiot light" at the top to tell people when I put on my brake. I'd checked them about a week before. So I assumed they were both out because of some other problem, wiring or something. I checked the fuses which were of course all OK. I took the car into the shop and they immediately replaced the little lightbulbs, both of which were burned out. I asked, "What's the chance they'd both burn out the same day?". They said, "It happens all the time.". If I had thought it out I would have tried to replace the bulbs myself first, just in case that was the problem. But I assumed they wouldn't go out at the same time because I grew up in the days before really great quality control.

People go by what they know, whether it applies or not. Once I read about an artist who didn't understand anything about electricity. His ceiling light had some problem, and he reasoned about it with the assumption that it takes a noticeable time for the signal to get from the switch to the light. If he had been using the new fluorescent lights there *would* have been a noticeable time between flipping the switch and seeing the light turn on.

Ideally people learn how to use each device. Where a "user illusion" helps is when the new device mimics something they already know how to use, so they don't actually have to learn how to use it. The drawback is that when they get outside the usual envelope of experience they'll use what they know they'll assume the mimicry goes farther than it really does.

It would be nice if people learned all about every device they use. But most of them don't want to. So for example the usual idea of how the phone network works is that they used to have switchboard operators who plugged in wires to make phone connections. And now they have computers to replace the manual operators. And they don't need to know more so long as it works when they put in the phone number. They learn a little about reception for their cell phones....

---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com