Virtual extension methods -- a strawman design

Brian Goetz brian.goetz at oracle.com
Tue May 25 10:42:12 PDT 2010


> Alex, replies inline.
>
> --Reinier Zwitserloot

Reinier is spot-on here, I just want to amplify one point:

> The key point is that errors _do_ occur, which is a lot
> better than silently picking one of the multiple available default
> implementations (better because by picking one, order becomes relevant, and
> that's bad).

Even if we wanted to do this (and it was considered), we couldn't: the 
relationship between the order of interfaces in a source file and the order in 
which they appear in the class file is unspecified.


More information about the lambda-dev mailing list