On 4 January 2015 at 20:07, Matt Oliveri <[email protected]> wrote: > Sorry about how late this reply is. Once I realized how carefully I'd > need to think to try and say something helpful, I kind of went into > procrastination mode.
Thanks for replying. I understand that the attributes we are discussing are conceptually different things. But it is important to note that these interact, and that the type system must ensure that they interact in such a way as to preserve the meaning of these things or they have no value. > Actually, a possible problem with thinking of immutability as a > property of an object (well, at least as an erased attribute assigned > once and for all at creation) is that it's conservative. An object can > be safely considered to be immutable once all read/write aliases > become unreachable. But how would you statically track that? Wait, > maybe that's one of William's tricks with immutable regions. Could we > be so lucky that readonly references fit well into the immutable > region picture? This is something I attempted to do with pypy effects; if you have fairly descriptive regions and effects it can be done. Especially if what you have is an actual type system and not a complicated offline multilevel cfa2 may-alias analysis (: -- William Leslie Notice: Likely much of this email is, by the nature of copyright, covered under copyright law. You absolutely MAY reproduce any part of it in accordance with the copyright law of the nation you are reading this in. Any attempt to DENY YOU THOSE RIGHTS would be illegal without prior contractual agreement. _______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
