Working again on the Debian Java policy (Was: FREEZE RESCHEDULED
On Sunday 7 November 1999, at 13 h 19, the keyboard of Richard Braakman
<dark@xs4all.nl> wrote:
> I'm rescheduling the
> freeze for mid-January, the weekend of the 15th and 16th.
Therefore, we have more time for the Debian Java policy. I suspended any work on it because I thought potato was too close but it is no longer the case.
The goal: a nice Java environment for potato. Not only programs, we had already several Java things with slink but since they was no integration, it was very painful to work with them (when you only have one package, everything is fine...)
The method: a Debian Java policy, such as the one I proposed in <http://www.debian.org/~bortz/Java/policy.html>.
The release: before potato freezes.
The editor: me, unless someone else raises his hand.
Since slink was released, many things happened in the Java world:
- in good (gcj, jikes - which is now free),
- in bad (SCSL),
or stalled (the freeness of the JDK and of the Java language itself, the numbering scheme of Java - discussed but not solved in the current policy proposal, free core classes, Japhar).
And in the Debian world, we now have three working Java compilers (I do not count guavac, not maintained) including two free.
A specially interesting thing is gcj: since it can produce native code, we could, at least in theory (many programs still do not compile with gcj), become independent of a Java virtual machine. Programs like the two XSL processors we now have could be compiled in native code: their users are not Java developers, they don't care about the bytecode, they don't need the classes, and it would certainly be the best solution.
So, time to go back to work.
Reply to: