Re: Beyond GCC 3.0: Summing Up

[Carlo Wood <carlo at alinoe dot com>]

On Mon, Jul 09, 2001 at 06:25:31PM -0700, Mark Mitchell wrote:
> We cannot go writing buggy code and then hope that we will have the energy 
> to fix it later and somehow expect to get unbuggy code. If we are willing 
> to check in buggy code, and not fix it in short order, we have to be 
> prepared to have constantly buggy code, and expect to have buggy releases.

Isn't the real problem that, when multiple developers work on a source tree,
large patches can collide when, after a long time, they are committed.

The reasons why developers want to commit changes on a short notice instead
the needed time to test things before committing it are imho the following:

1) As long as only one person is working on a large patch, he can 'update'
every day and as such feel confident about 'being up to date' with the tree.
But when two or more people are working on a large patch, the first one who
commits his work can sit back and relax, while the other has to fix a lot
of collisions - or worse - faces changes that makes his approach of the
problem invalid.

2) Another reason is that a patch at some time needs to be tested by others,
and the CVS tree is a nice medium to distribute an experimental patch and
have it tested.

The above is essential the cause for the problem that is being discussed.
If there can be found a solution, or compromise, then it is related to this
I think.

-- 
Carlo Wood <carlo@alinoe.com>