Forests versus patches [was: IcedTea7 2.0 Branched for Release]

[Andrew Haley]

On 09/29/2011 01:05 PM, Matthias Klose wrote:

>  - The movement to the "forest" development model doesn't help many
>    people, only a few.  It's not communicated, only a few more equal
>    IcedTea developers have commit rights, the regression testers use
>    the icedtea7 branch, and iiuc, people like Xerxes still have
>    difficulties or issues contributing to the icedtea7 forest.

I'd like to take just one of these issues on its own.  It's not
completely clear, but I assume that the issue you're describing is the
use of an icedtea forest at java.net versus a patches directory at
icedtea.classpath.org.

The use of the patches directory never was a good feature of IcedTea.
It was perhaps good enough at the time, particularly as we did not
want to maintain a separate fork of the OpenJDK sources.  Today,
however, there isn't any really good reason for maintaining patches.
We have a version control system that tracks changesets correctly, and
in the long term it's the best way to handle changes.

Of course, this only works if people have access to the forests.  I
don't know why that isn't so, and that isn't the point if this mail.
I'm trying to address the issue of the icedtea7 forest.

Andrew.