DLJ 2.0 Re: [Fwd: DLJ 6u10 bundles have been posted on jdk-distros.dev.java.net]

[David Herron]

Mark Wielaard wrote:
> Hi David,
>
> On Fri, 2008-10-17 at 11:09 -0700, David Herron wrote:
>   
>> I suppose you're requesting that I stop sending the DLJ announcements to 
>> this list?
>>     
> Lets just try to figure out some way that the DLJ and OpenJDK
> communities could work together. The problem as I see it currently is
> that you are publishing these "bundles" under terms which explicitly
> deny anybody from combining, configuring, distribute or even running the
> DLJ bundles together with OpenJDK derived projects. Is there a way to
> get around that limitation?
>
> And what parts/ideas of the DLJ bundles do you think could be used (as
> inspiration) to improve the OpenJDK work? Would it be possible to
> publish those parts so others could extend and improve on the work in
> conjunction with IcedTea/OpenJDk?
>
> Thanks,
>
> Mark
>
>   

I changed the subject since you're kind of describing a DLJ2.0 (or 
DLJ-the-next-generation).  And I appreciate your willingness to work 
with us on this.

The part which would be easy to reuse is packaging scripts.

While the bundles themselves are under a non-free not-quite-icky closed 
license we chose the BSD license for packaging scripts inside the 
jdk-distros project.   This means jdk-distros could be used for shared 
development of packaging or other meta scripts some of whom I suppose 
could be applied to openjdk builds.  That could be done today no changes 
required.

As for the bundle license... I could imagine some changes but am 
reluctant to take many steps down that road.  It would cost some 
internal effort to work through changes, and I see little reason to 
pursue it because I believe there is limited usefulness to the DLJ 
bundles.  e.g. I think that when we get to the end of 
oh-what-the-heck-lets-call-it-java-7 and openjdk7 approximately equals 
closedjdk7 there will be little reason to have DLJ bundles because 
openjdk7 will be close enough to closedjdk7 that it won't make much 
difference.  The big question mark there is what will be the status of 
the plugin, and as we said last month that decision is up in the air.  
In any case if we're likely to drop DLJ next year because it's no longer 
needed then it doesn't seem useful to me to spend the resources to 
change it now.

One potentially useful DLJ change would be to change it so a non-distro 
(like JPackage) could participate and redistribute DLJ derived bundles.  
Right now the license is only open to distros.

- David Herron

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20081021/ede5ff8f/attachment.html