Hello all,

Now that we have release 0.11.0 done, I'd like us to have a discussion
about our plans for upcoming releases, and to decide when we're going to
make the jump to version 1.0.

Looking back, the core of Brooklyn is now very strong, and has been tried
and tested by many people for many months, some using for production
workloads. The operations side has matured with the support for persistence
and high availability. Pure YAML blueprints are now extremely capable, and
the Java-based blueprints in brooklyn-library have in many cases been
overtaken by pure YAML blueprints written by Brooklyn users. The catalog
provides a way to manage and compose blueprints, with OSGI providing a
means to manage and version. Finishing touches, such as supporting ZIP
files for catalog uploads (removing the requirement that a user must create
their own OSGI bundles) mean we will have a very compelling story for
blueprint developers.

So, what's left to do? Are there any major (or even moderate) features that
we would want in a 1.0 release?

My thoughts are that we release 0.12.0 next. This will put the Karaf-based
distribution as the primary artifact, with accompanying .deb and .rpms. The
classic distribution will remain but be deprecated. At this point, we
should make an effort to mark as "deprecated" as much as we possibly can,
so we don't have to take unnecessary baggage into the 1.x series.

After that, we aim for 1.0.0 - the classic distribution is removed, and so
is everything marked as deprecated.

In parallel with this, an effort to overhaul the documentation, so that our
1.0 docs are comprehensive and easy to use.

Any other thoughts?

Reply via email to