xc/programs considered harmful

[Egbert Eich]

Keith Packard writes:
 > 
 > Around 15 o'clock on Dec 17, Owen Taylor wrote:
 > 
 > > What really disappoints me here is that modularization has been 
 > > discussed, experimented with, tried out for several years, and nobody
 > > has sat down and wrote down a concrete plan for:
 > > 
 > >  - How will the code be structured in CVS
 > >  - What will be the released tarballs
 > >  - What are the stages for moving code to match 
 > 
 > This is putting the cart before the horse.  Until we have an actual 
 > commitment from X.org (however that is done) that release 'foo' will be 
 > modular, any planning for modularization is strictly navel gazing.

I may be mistaken but I have seen a general consensus that modularization
is the way to go for the future.
But what else do you expect if there is no plan, no roadmap, no concrete
proposal? 
We must not expect people to commit themselves to a less than certain
future. Instead let us overcome this 'hen and egg' problem and 
establish confidence by:

a. aiding people to transition to the modular bits that already exist.
b. being open to issues that people raise with these bits.
c. providing a (preliminary) plan and roadmap as a basis for a discussion.

 > 
 > We have to get help from everyone involved, even those vehemently 
 > opposed to modularization, to make sure the system will work for them.  
 > Right now, we just hear about how 'environment foo' or 'system bar' will 
 > break irrepairibly as a result of modularization instead of patches to 
 > make it work again.

I agree with you here. 

 > 
 > There's no way we can get rough concensus about how modularization should 
 > work until we have rough concensus that it should happen at all.
 > 

And this consensus can be achieved much easier following Owen's advice
to propose a plan which can be discussed. 

Egbert.