[cap-talk] "Same" key

Mark S. Miller markm at cs.jhu.edu
Sat Feb 3 16:29:53 CST 2007


 >> Can you give an example where two not-EQ caps designate the same object?

Dean Tribble wrote:
> 2 examples:
> - Transparent forwarders
> - Caps with capabiity bits used for logging data but not behavior difference.

In both cases, I would describe these as separate objects which, EQ aside, are 
not observably distinguishable. Similarly for my previous answer to Charlie:

Mark S. Miller wrote:
 > Charles Landau wrote:
 >> By "distinguished by their identity", do you mean keys that can be
 >> distinguished by EQ but not by any other means? I don't think there
 >> are any such keys in KeyKOS.
 >>
 >> Since you've used that distinction in KeyKOS, can you give examples?
 >
 > Two start keys served by the same domain, but with different facet-ids,
 > where the serving domain ignores this difference.

Another example along these lines: start keys to two distinct 
separately-created prompt domains which happen to behave identically. Likewise 
in E:

     def f(x :int) { return x+x }
     def g(x :int) { return x*2 }

EQ aside, f and g are not observably distinguishable. But in the semantics, 
they are distinct objects. We often make distinctions for descriptive purposes 
that are not observable to the entities within the described system. In 
physics we invent notions like "quark" as descriptive conveniences, even 
though physics seems to prohibit observation of individual quarks. 
<http://en.wikipedia.org/wiki/Colour_confinement>

-- 
Text by me above is hereby placed in the public domain

     Cheers,
     --MarkM


More information about the cap-talk mailing list