Re: another gnome-libs .02

[Maciej Stachowiak <mjs eazel com>]

Havoc Pennington <hp redhat com> writes:

> 
> I think you're simply wrong about this. A giant lib is "less work"
> only if you don't consider that you've just forced dozens of unrelated
> modules that could have been done in parallel on different schedules
> to work together. Fred Brooks would have your head on a plate. ;-)
> 

One specific way in which George is right is that a system with more
independently changable and separately release components with
different maintainers will take longer to release engineer. If you
doubled the number of modules in GNOME 1.4 without increasing
fucntionality, just by splitting things, I expect we would be looking
at a 4/27 release date, not a 3/27 date.

So while in general factoring things more is better, we should not
assume that adding additional modules to the core release is free.

> Clear bounds on the libraries. Parallelization of maintenance. Minimal
> interdependencies.
> 

I agree we would have these benefits but we must realize the cost in
increased coordination complexity. 

Getting all the different packages together for GNOME 1.4 beta 1 was
already non-trivial, and that was a case where we could use an old
release for many modules, something that will not be true for
compat-breaking releases.

 - Maciej