Joe Schaefer wrote:

it's not the multiple codebases that's the issue; it's that they both had the same name and both continued active development.

as a counterexample, both httpd 1.3 and 2.0 were in concurrent active development for a long time. 1.3 development slowed down, and when 2.0 was finally released, 1.3 went into a sort of maintenance mode -- meaning that bugfixes and minor backported features can show up in it, but nothing that doesn't also appear in 2.0 at the same time.

i'm not familiar with the data around the tomcat issue, just the metadata, but i'm getting the feeling that major divergent feature enhancement was occurring in both branches. (e.g., something might be added to tomcat 3 that wasn't (and won't be) in tomcat 4). is that correct, or a misunderstanding on my part?