UNCLE! (was Re: eq and hashing (was Re: [plt-scheme] How to make unit functors?))

On May 29, Matthew Flatt wrote:

> If any future implementors complain about the cost of `eq-hash-code'
> on symbols, I'll send them to you guys. :)

[Technically, this is the future...]

I (personally) find your previous statement:

| If you're willing to pay for that (and it's a fairly small cost for
| typical symbol sizes), then use `equal-hash-code'.

very reasonable.  I wouldn't like the extra cost approach not only
because of that cost, but because I take it as something that should
expose something which now it wouldn't expose anymore, and I wouldn't
like the system to fake it for me just for symbols...  I'd prefer if
`eq-hash-code' didn't exist over modifying the implementation because
it exposes some internals.

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                  http://www.barzilay.org/                 Maze is Life!