← Older revision

Revision as of 22:58, 1 October 2008

Line 185:

:--[[User:Peter gummer|Peter gummer]] 00:34, 19 March 2008 (PDT) Paul mentions the word "optimize" twice in the article: it's only a potential bonus of his proposal, not the principal motivation. And his proposal isn't for "static methods" (although it does provide a mechanism for them: a frozen objectless routine would be equivalent to a "static method", I suppose). Your "implicit class" idea is interesting, but it's not clear to me how it addresses the concerns that Paul raised.

:--[[User:Peter gummer|Peter gummer]] 00:34, 19 March 2008 (PDT) Paul mentions the word "optimize" twice in the article: it's only a potential bonus of his proposal, not the principal motivation. And his proposal isn't for "static methods" (although it does provide a mechanism for them: a frozen objectless routine would be equivalent to a "static method", I suppose). Your "implicit class" idea is interesting, but it's not clear to me how it addresses the concerns that Paul raised.

+

:-----[[User:Paulb|paulb]] 15:58, 1 October 2008 (PDT)

+

Actually any call made using the Eiffel static calling convention would be performed statically instead of dynamic dispatch. In that case objectless calls need not be marked <e>frozen</e> in order to benefit from performance gains. And, although performance is not the generalized issue here, I would like objectless calls to be implemented for performance reasons also. There is a dramatic increase in performance when using static calls as apposed to dynamic dispatch.

Talk:Objectless Calls - Revision history

Paulb: Added comment.