Gianugo Rabellino wrote:
<snip/>
Well, I think I do know at least something that blocks will buy us.
But again, what I'm questioning is whether the problems that blocks
solve are self-inflicted pains (love that wording, Berin) or real
problems that have a direct impact on users. So far what I'm seeing is
the need to learn yet another tool (Maven2) to manage block's
complexity: does that *really* go in the direction of helping users?
I'm not that sure anymore.
sure it does :-)
We need to be able to release our blocks independent of the core. Being
able to produce a snapshot of a block and make it *instantly* available
to all users without even doing an official release is extremely
valuable to our end users.
Example 1 : 2.1.8 was delayed for one week because we thought the jsf
block had files in it with a non-compatible license (which turned out
not to be an issue in the end)
Example 2: since maven 2.0 was released, issue [1] is blocking our
ability to use archetypes. I spoke to Jason (vanzyl) again about this
yesterday evening, and he offered to do a quick fix. One hour later i
had a new snapshot of the archetype plugin with a workaround. There was
no previous big discussion on how to best implement it, no new release,
no code freeze, no vote - nada.
Yes we can manage this all ourselves, but maven makes it soooo much
easier. And ofcourse we will hide the perceived complexity for our users
through tooling.
Regards
Jorg
[1] http://jira.codehaus.org/browse/ARCHETYPE-1
