[Jsr308-statements] Are loop annotations within scope?

Michael Ernst mernst at csail.mit.edu
Sat Feb 3 09:40:15 EST 2007


Trevor-

> some seem quite opposed to the very idea of something like  
> loop annotations, and that's before any details have been proposed.  
> So there's a real disconnect between what I'm reading on the status  
> page and what I'm seeing on the jsr308 list.

Please do not be discouraged by loud complaining by some individuals.  At
the end of the day, volume will not determine the outcome of JSR 308.  (And
console yourself that there are people vocally opposed to every single
change, including all those currently in the JSR 308 document.)

For the record, I have not yet made up my mind:  I am neither in favor of,
nor opposed to, annotations on statements.  I like the motivation, and I
don't think the syntax will be a problem.  Some potential problems must be
clearly addressed, however.  One example is how to deal with compiler
optimizations, such as those that move code out of loops.

A major reason for the jsr308-statements mailing list is to give people who
are in favor of these annotations some quiet space to work without constant
sniping.  Once you come back with a complete proposal (with motivation,
rationale, etc.), then it is fair game for their criticism.  And when the
details have been proposed, the current, largely speculative, opposition
will either wither or be shown to have been valid all along.  So please do
pay close attention to the technical nuggets behind others' comments, even
if not necessarily to the tone.

                    -Mike



More information about the Jsr308-statements mailing list