On Feb 4, 2005, at 9:52 AM, Nico Klasens wrote:
[zapped my remarks]
What is the reason to have more committors then we have now? There are a lot
of people who have the right to commit, but only a few have used it in the
last months. This last year does contradict with the remark you make. Some
people who were willing to become a committor were denied, because they did
not contribute enough as a known mmbase developer.
The above is not very pragmatic and realistic. I did the changes in the
present where only a few commitors are active on the repository. I don't
understand why I should take possible future situations into account in my
current actions. I will deal with it when it is there.
You have 2 valid points there lets look at them on why the mmc and me (as ex mmc) was worried and pushed
for more developers and changes of some the rules.
Very few of the current commitors really commit into the core, So why do we need more of them ? Well the reverse
is that if for some reason these 4 or 5 people for whatever reason stop core development will stop. Most of these people
don't have 'work on the core' as a job title. imho we are living on borrowed time. The idea is with more commitors the changes
are we will find more very active developers who could and will be ready to replace that small group if needed. The only other way
i can see is just hope for the best or slowly move to the state where the about 30 sponsers seeing this and create 4 or 5 job
titles that say 'work on mmbase core' (removing them from any companies project cycle and making mmbase their only focus).
Your second point on but you (and others have voted down new commitors), Personally i voted down 1 not because of him but because
of the timing of the call (it was done when i left the mmc and we had a vote running for my replacement). I since then asked the person in
question (since he works at my company) if he wanted me to create a new call and he declined. The other negative votes seem to be 'ive not
seen any code from him or i don't know him' they are right in a way since we put that as demands on becomming a commitor !!
I know the mmc has been trying to find a way out, my personal view is work on project level, allow the projectleader to add developers on his own. This also makes more and more sense since we more and more work in layers and applications. But this seems to be hard to enforce in cvs so if this is done it will probably be a model
based on trust where every assigned developer can access the whole cvs and we frown and ask for comment from the projectleader if a developer commits outside of the project(s) the developer is assigned. I personally see your action in this light i feel it would have been better if you had a highlevel ok from the project leader on that project even if it was a small email or phone call.
Again this is my view but i would be for lowering the bar on allowing people in cvs to near zero (a project leader who wants you in his team would be enough) but i also feel that this can only work when we have some basic rules like stay within the scope of the project and problems and questions will be aimed at the projectleader.
The timimg of mine was okay I think. Not a lot of new functionality projects
are running and one big step in the cleanup project was just finished, And
to be very pragrmatic, most of the changes (if not all) would easily be
merged by CVS into changed code.
You still feel you did something wrong you didn't, My point is that the work should (and is imho) part of the cleanup project and it would have been nice
and made more sense if he knew you where about to do it. One of the reasons is that if someone else does it and f***up i would have asked him what happend?
[zapped some]
Maybe my perception on what a committor is allowed to do on the HEAD is much
more loose then it should be. I will keep that in mind next time.
Well its not a bad idea to talk about these issues and make them more clear for all. The end result
for all is that we allow developers to add useful things to mmbase and be happy.
Why do we need more developers? For what do we need them? What issue is the
community trying to solve with mmbase at the moment? Mmbase already solved
the issue to store content. It is a content toolkit which can be used to
build many types of systems.
So mmbase is done ? Well we have a high commit count compared to most opensource projects so something is
not done i guess. Ive talked about this before i feel we as developers need to rethink what MMBase is. Since there
is more than just the core. Most of the work seems to be done above the core these days in either frameworks (finished
cms systems) or tools and services. I think its time we who build the product !!! had a much better idea of what MMBase is
and what we want it to be in the future. My guess is its more than just the core and if that is the case we need a better think
about these.
I think it is good to talk about these topics, but not (only) on a
developers meeting. This list is viewed by more people then there will be
people at meetings. Another advantage to discuss it on the list is that
people can respond when they have the time to respond and not when the
meeting is held.
I am more than happy to talk about it this way but i know im not the best in english and making my ideas clear this way.
Nico
Greetings,
Daniel.
_______________________________________________ Developers mailing list [email protected] http://lists.mmbase.org/mailman/listinfo/developers
