Are they eagerly awaiting Doxia 1.1?
What I'm wondering is, what are we standing in the way of by not putting
the latest, greatest Doxia into Maven 2.1.0?
Back in July/August of last year, we talked about pulling back from the
Doxia upgrade so it could be released and soak into the community for
awhile, as a substitute for in-depth testing. If Doxia had been released
as 1.1 back in October, and had been used (or even tried) in the
interim, a lot of this hand-wringing could be dispensed with. We could
have bumped the Doxia version first thing in the 2.1.0 release process,
and gone on our ways...and tested it all along while we fixed other bugs.
If Doxia doesn't really have other consumers, then that means Doxia
effectively has a slave release cycle, locked to Maven's. In this case,
releasing early doesn't matter, and the risks are unavoidable when we do
an upgrade. The risks can be minimized with a large set of tests, but
they're unavoidable.
I'm just trying to understand sort of why and how we got to this point.
I'm not as convinced that the work we're doing in 2.1.x is pointless
because 3.0 is right around the corner. So for me, it still makes sense
to plan out our migration path WRT Doxia and other libraries in a way
that will absolutely minimize the possible impact on our users.
-john
Vincent Siveton wrote:
Hi John,
2009/3/3 John Casey <[email protected]>:
Is Maven the only consumer of Doxia? When you say "everything else moving
AFAIK, Xwiki and m2eclipse are also using Doxia
Cheers,
Vincent
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
