Teams

[Steve Gehlbach]

Troy Rollins wrote:

> Using libraries and "start using" and such is both more clumsy and a 
> step removed from the actual development process. Even CVS, with all of 
> its associated bull, comes closer to the team development intent, I think.

I have found the bull worth it, though at first it is confusing.  But 
for distributed development, which is done more and more these days 
across not just rooms, but countries (and maybe soon even planets and 
solar systems :-)), CVS is hard to beat.  Sourceforge, which uses CVS 
exclusively AFAIK, though sometimes slow, has been central to the 
collaborative methodology of the open source movement.  I have 
personally been involved in a couple of these projects with engineers in 
the US, UK, and Germany, and it is quite a thing to be part of.

But there in another reason for team development, and that is project 
stability.  The loss of one team member, though difficult, does not kill 
the entire project, as it can when it is all in one person's head.

IMHO, if a file format and method cannot be found to use RR in a team 
approach, I think all RR projects will self-limit to a smaller size and 
more limited problem set than otherwise would be attainable with a team 
approach.

-Steve