On 06/20/2012 03:30 PM, Joachim Schrod wrote:
Leonardo Uribe wrote:
Trinidad is considered "rock solid". Its API is stable, well tested
and it offers outstanding browser compatibility (now you get why it
doesn't look too fancy).
I have used Trinidad now for roughly two years in two smaller
projects; and yes, it is solid. I wouldn't say "rock solid",
though. In addition, I have no qualms at all about its look, the
skinning facility and component simple mode made it possible to
realize all layouts that our designers prototyped.
My question was triggered by the following observations:
-- In the last half year we had roughly 16 threads about Trinidad.
5 of them got some reaction (one was the long thread about a
different JS implementations that popped up here again), and 11
postings got no answer as all. Alone 3 of them in the last
couple of weeks. I have to admit that I didn't know the answer
either, otherwise I would have answered; but I'm still quite
a newbie.
I.e., it's not only the developer community, the user community
is very slim as well. There is not much help available if
problems in using Trinidad raise their head.
Very true. I think with the lack of increased functionality in
Trinidad, there has not been as much enthusiasm as there has in the past.
-- After long prodding, Scott has agreed to make new releases, and
he put out 2.0.1. But 1.2.15 is tagged since ages in the SVN
repository, and the actual bundling for a new package doesn't
happen. More Trinidad JIRA tickets seem to be created than
solved; at least I got the impression over the past few months,
maybe that's wrong.
That last part is actually wrong. The first part is right. :D To let
people understand, the company I work for donates a lot of time to
Trinidad. We hired Matthias who was full time on this project. I took
over for him and I'm not. And while much of my work and the work of
many of the Trinidad core developers has gone into getting stuff to work
with new releases of JSF and extending the infrastucture to support
ADFFaces, you're correct in that there hasn't been a lot of new
functionality added.
As for the JIRA tickets, we are still closing more then what we are
opening, and that is even with many tickets getting duplicated. I
usually do massive cleanups around release time. Admittedly though,
some have fallen through the cracks.
As for 1.2, the 1.2.15 tag needs to go away and get regenerated. After
taking over for Matthias it became apparent that the 1.2 branch was
never sunsetted like it should have been. Still, no fixes were being
added to it and I know a lot of the core developers don't have the
cycles to dedicate to a 1.2 release, we're all about 2.0 and 2.1 for
now. So yes, it's been a while. What we need is one final release and
let the thing get sunseted. We, as a project, typically only keep open
the two latest branches.
So I deduce that current developer community is too small or
that its interests are different. And you support that opinion:
Most of the interest of MyFaces developers these years has been around
JSF spec, MyFaces Core and MyFaces CODI (future top level project
Apache DeltaSpike), which have primary importance for JSF and are
critical parts in the day-to-day work.
Please note: My problem with Trinidad is not its missing further
development. My problem are bug fixes and missing releases that
incorporate those bug fixes. Since many months I use patched
versions, and self-compiled JARs. And since I don't see that this
will change I start to evaluate my options. No bad feelings from my
side, really.
Thanks for the input. Which branches are you using?
Scott
