Re: Beyond GCC 3.0: Summing Up

[kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)]

    > But a key point is who are the "users" here?  I'd argue that the
    > "users" are in fact the set of developers since they are the people
    > who mainly use the top-of-tree sources.  So what we're doing with the
    > proposed approach is frustrating one group of users in favor of another.

    Yes, but if, say, IRIX 6.5 is broken, and a developer who works on the
    Java front-end on an IRIX box cannot work without manually reverting a
    bunch of patches, and who may not even known what has gone wrong, then
    you've frustruated GCC developers too.

It sounds like you're agreeing with me.  The tradeoff is between
frustrating potential developers who want to be able to contribute 
without having to have a very large amount of knowlege of the compiler
and frustrating developers who are working on the specific areas where
things ended up broken.

That's why I said it's critical to distinguish the degree of breakage
rather than making a uniform and arbitrary rule.  That says how many
people are "frustrated" by the change and can be traded off against
the frustration of the new developer.  For example, even if IRIX 6.5
won't bootstrap, but basically works, the Java front-end developer can
most likely get a lot, if not all, of his work done.  If the compiler
bootstraps but fails some regression tests, some people indeed may be
hampered, but even fewer.  If the compiler won't even compile (as was the
case on Alpha this morning!), then yet more people are frustrated.