From: Carter Cheng Date: 2011-10-26T01:45:29+09:00 Subject: [ruby-core:40362] Re: 2.0 feature questionnaire --20cf307cff3a4e705804b0224405 Content-Type: text/plain; charset=ISO-8859-1 I am not sure if I am in a position to comment here but it does seem that some proposed features in MRI may in essence make it more "static" and thus a harder platform to evolve for the future. AST inspection- I am not sure what the exact motivation is for this but it does expose portions of the frontend of the compiler system which basically adds to the "spec" of the language. YARV bytecode- I think the same comment applies to this- and I am uncertain if the YARV byte code system effective as it has been as an internal representation for the 1.9 releases would be as an exposed portion of the spec. My "gut" tells me that in a JIT type situation certain opcodes could probably be removed and be inferred by the system. At an extreme would be something like the high level byte codes used by Craig Chambers. Just my .2 cents, Carter. --20cf307cff3a4e705804b0224405 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I am not sure if I am in a position to comment here but it does seem that s= ome proposed features in MRI may in essence make it more "static"= and thus a harder platform to evolve for the future.

AST inspection- I am not sure what the exact motivation is for this but it = does expose portions of the frontend of the compiler system which basically= adds to the "spec" of the language.=A0

YARV bytecode- I think the same comment applies to this- and I am uncertain= if the YARV byte code system effective as it has been as an internal repre= sentation for the 1.9 releases would be as an exposed portion of the spec. = My "gut" tells me that in a JIT type situation certain opcodes co= uld probably be removed and be inferred by the system. At an extreme would = be something like the high level byte codes used by Craig Chambers.

Just my .2 cents,

Carter. --20cf307cff3a4e705804b0224405--