I speculate that all(?) of the semantics of finalization
given in JLS are moot, to be replaced by new semantics
that will be defined solely in terms of weak references.
That is, the weak reference will be the only JVM
primitive w.r.t. this issue.
(There may also be little issues, like access control to the
finalize method.)
The semantics for weak references do not exist yet, AFAIK.
~~
http://www.sni.net/~cargill
-------------------------------
JavaMemoryModel mailing list - http://www.cs.umd.edu/~pugh/java/memoryModel
This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:23 EDT