Am 30.08.2013 11:13, schrieb Aleksandar Kurtakov:
That's exactly where we disagree. The success of the platform is also it's 
curse. The long term backwards compatibility is the usual excuse to not reject 
new contributions and old time contributors stop contributing one by one so we 
pile bugs and Eclipse platform starts to lose the battle for new projects when 
compared to others as it looks ancient to many developers. So yeah let's rely 
on backwards compatibility till there active committers are so many that can 
barely do the builds not to mention fixing bugs/enhancements. Is that what we 
look for? There is time when long term backwards compatibility must be traded 
for having active contributors.

Absolutely not. And I didn't think that these most fundamental policies are at any disposition now. I think over the past ten years the platform has proven the excellent ability to integrate new functionality while not breaking backward compatibility. That the result (e.g. extension interfaces, etc) looks ancient is certainly not the common perception.

The project is it's contributors not it's API.

That sounds a little as if Eclipse projects are only playgrounds for "the cool kids". I think a project is successful if what it produces (including the APIs) is successful, i.e. widely adopted. The adopters have to be pleased, not the contributors.

And when I say break the API - what's wrong with deprecating something today

That's okay because it's within the policies.

and drop it in 3 years?

That's certainly not okay without a major release. Adopters appreciate major 
releases if they benefit from them.

There is clearly no one interested in doing only maintenance, as can be seen 
from the codebase.

Clearly no adopter is interested in adjusting his code *without* any benefits 
for *him*.

Cheers
/Eike

----
http://www.esc-net.de
http://thegordian.blogspot.com
http://twitter.com/eikestepper


_______________________________________________
cross-project-issues-dev mailing list
[email protected]
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Reply via email to