db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel John Debrunner <...@apache.org>
Subject Re: 10.4 Feature Freeze
Date Wed, 27 Feb 2008 17:16:26 GMT
Rick Hillegas wrote:
> Daniel John Debrunner wrote:

>> I'm trying to add Java security manager checks (DERBY-3462) to the JMX 
>> MBeans so that security is not compromised by the addition of JMX. 
>> While I'm not blocked by DERBY-2109, if I proceed ahead of DERBY-2019 
>> committing code JMX related permission code then any DERBY-2109 patch 
>> will have to re-worked again. However since there doesn't seem to be 
>> any activity on DERBY-2109 I may just go ahead anyway.
>> Dan.
> Please do not do this, at least not in a way which will make it 
> necessary to rework and retest DERBY-2109.

I'll avoid doing this, I think there is a temp workaround I can do. If 
we get to the branch cutoff though without any progress on DERBY-2109 
then changes may need to be made.

> Martin is actively working on 
> this feature and deserves our patience and respect.

That's great, but you have to look at it from the issue of activity on 
the list: Given no messages for over two weeks from any contributor 
there are a number of scenarios one can imagine from not being active on 
derby anymore to being stuck on a problem that the community could 
possibly help with. Without communication it's impossible to tell so one 
has to assume not active if one wants to fry a fish in the same or 
overlapping area.

> Each rework/retest cycle is turning out to be very expensive.

Is there a way the community could help, posting a newer version of the 
patch, helping solve issues, running the tests on various platforms?

I think this patch would have benefited from continuing the incremental 
development approach it started with. The current patch is trying to 
solve at least four issues. With an incremental approach my opinion is 
that most of the current patch's functionality could have already been 
committed, allowing focus on specific problems and a quicker turn-around 
on testing etc. I wonder if some folks are reluctant to perform 
incremental development because they think patches are too slow to be 
applied, thus they do a mega-patch which of course will take time to 
commit, thus we have a self-fulfilling prophecy?


View raw message