> > I am very disappointed in this outcome of the HACK proposal. I 
> > thought, it had a small change to be added to 1.7, but that 
> it didn't 
> > even make it into HEAD really astonishes me.
> 
> The main problem here is that the cotes were too varied - 
> some people demanded a project, some gave reasons but even if 
> you change code, you need to resubmit it again. That is the 
> (nasty) nature of hacks.
> This is a reason why we want people to discuss most votes first.
> 
> If you add the changes, you can resubmit the hack, and people 
> can bring out a vote on that hack proposal. Making changes 
> mid-vote can get too confusing, and I cannot guarantee 
> everyone will agree still agree.
> 
> I may have added to the confusion by adding the packaging 
> vote, though I don't think that really would have changed the outcome.
> 
> The easiest is really to adapt and recast the vote, and hope 
> the critics can agree with the changes, or to make it a small 
> project, which allows you to change the code as you progress 
> (depending on input).

I am just suprised that it was very easy to put the code here in production
and that it didn't pass the vote to put it into the development branch of
MMBase. I have no idea how to change the code to let it pass. I offered a
two-line code change for the raised issue without success. It seems silly to
me to propose the exact same code with only 2 lines changed for an
underconstruction codebase. The code in the 'speeltuin' has more lines
changed just to let the deployer decide what to do with the issue instead of
the developer.

I have no issues with the nature of a hack. IMO, committers are allowed to
narrow down the scope of the affected code (bases) and add requirements to
the checked in code. This is why I am suprised that it didn't make it into
head.
If it turns out in the vote that the offered code requires some re-thinking
it is better to resubmit the code, but I don't think my offered code does.

Just to be clear. I agree that the packaging vote only generated some more
messages on the list which didn't relate to my offered code. It was just one
step more in the right direction.

Nico


Reply via email to