New release schedule proposal

[Caio Tiago Oliveira]

On 08/10/2013 02:29 PM, Antonio Terceiro wrote:
> The proposal we are making is based on models that already work pretty
> well in projects as Linux, GNOME, etc. I believe we have some experience
> with free software development and software engineering in general to
> know that completely upfront planning does not work. YMMV.

FWIW, one well known IT project with a design by committee is the Java
Specification, but they take ***years*** to include some feature and
even with lots of arguing, time and resources, a lot of the planned
features just miss one or ***two*** releases.
Another examples is the HTML working group (managed by W3), which takes
ages to change something and another working group was created to
discuss the proposals and then send them to the working group. Even
though, we have no official HTML5 today, for instance.

Designs by committee with discussion about what to include are present
usually when you have huge conflicts among the users of a technology and
the result is a slowdown in the whole process.

Oracle tries to alleviate that making deals and lobbying for the changes
they want to include.

For such a small project, most problems would be easily solved without
requiring a design by committee.

With this new release schedule, the user base would have some time to
test a *quite stable* version and find anything which breaks their app
or decrease the quality in some way.

Regarding huge structural changes, I agree that they should be widely
and openly debated before being implemented (as for the solr
pluginization). Although in this specific case one could install the
plugin to use the solr features. There is no reason to enforce someone
to use solr (run a JVM, eating memory and CPU cycles) if they want a
simpler search.
If there is documentation lacking on the subject for a newcomer user, it
would be nice to know.

Regards,