[Tm4j-developers] TM4J2 / TMDM Support Re: [TM4J-users] TopicMapObject.equalsByID - is it really needed?

[Kal A. <tec...@gm...>]

Hi all,

I've been reading the thread on the recent CVS changes with some interest
and after some discussion with Conal and Benjamin at the Topicmaps 2008
conference (face to face is *so* much better), I think I'm beginning to see
a strategy that could work for us in moving ahead. As you probably know, due
to other commitments I have not done much (read 'any') coding on TM4J for a
long while, but I would still like to see the project move forwards, make
released packages that can be used to generate more outside interest, and
move back to the forefront of supporting the newest and the forthcoming
topic maps standards.

So I've started another thread to discuss this.

I would like to propose the following :

1) XTM 1.0 support should be the primary focus of TM4J 1.x. I can see this
as being three main tasks
   1a) Bug fixes and API tidying. I believe this includes removing the
equalsByID as an interface method and fixing TopicMapWalker.
   1b) Ditch the Ozone backend. Ozone hasn't made a release for 3 years
(almost as long as TM4J ;-) and limited development resource is spread too
thinly with multiple backends.
   1c) Upgrade the Hibernate backend

2) Start TM4J 2.0, not as a branch of TM4J 0.9.x or TM4J 1.x, but as a
completely new codebase. Throw out everything and start with a discussion
about what should be in TM4J 2.0 - my feeling is that keeping stuff built on
TM4J 1.x will slow down progress rather than enhance it. At this stage it
should be about getting a plan and an agreement about what to code rather
than about getting any one individual creating a lot of code and *then*
having an argument about what to do with it :-)

I'm really glad to see discussion on this list again - and I hope that we
can turn it into some momentum to get code written again. I might even dust
off my Eclipse installation and get my hands dirty again...no promises ;-)

Cheers

Kal