On 9/17/07, Geoffrey Spear <[EMAIL PROTECTED]> wrote: > On 9/17/07, Ed Murphy <[EMAIL PROTECTED]> wrote: > > Zefram wrote: > > > > > It seems to me that switches are a special case of your variable concept, > > > where the restricting type is an enumeration. > > > > It doesn't have to be. A rule could define a switch's valid values > > as "all real numbers from 0 (default) to 1, inclusive", for instance. > > It could but that would break the switch metaphor IMO. I'd suggest > calling such a thing a "dial". Then again personally I think a > "switch" should have exactly two possible values.
Interesting point. This suggests to me that what we really ought to do is to split switches into two disjoint components: a variable to hold values, and a switch (or dial, or whatever we want to call it) that exists purely as a mechanism for changing its variable's value. -root
