[sc34wg3] Re: Reducing redundant constructs

[Lars Marius Garshol]

* Patrick Durusau
|=20
| One of the issues where we did not follow the T+ model was on
| "identical" subject proxies.

Let me try to be a bit clearer on this.  In the editorial comments you
say that

  id({ < =D8, 0 > }) =3D identifier

(Using =D8 instead of "bottom", since that doesn't exist on my keyboard,
and they mean the same thing anyway.)

Well, {< =D8, 0, >} is a particular proxy, and any proxy that has a
single property where the key is =D8 and the value is 0 is the same
proxy as that proxy, because they will be the same set. That's implied
by the use of set theory, and if you don't like it you have to stop
using set theory and find some formalism that does what you want.
=20
| On the whole the notion of "identical" outside of a definition of
| "identical" just seems real problematic to me.

Set theory has a definition of this, and you're bound by it, since
proxies are defined to be sets.

--=20
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >