Closed GoogleCodeExporter closed 8 years ago
I had also thought about this and I think we should not upgrade unless there is
a particular bug or API in a new UIMA version that affects uimaFIT. If we
upgrade, we also upgrade all our users because Maven automatically selects the
highest version of a dependency, unless the user actively overrides this. If we
do not upgrade, we leave that choice of using 2.3.1 or 2.3.0 to our users.
Original comment by richard.eckart
on 14 Apr 2011 at 6:45
So you think it's likely that people will want to upgrade their UimaFIT from
1.1 to 1.2, where there are backwards incompatible changes (e.g. the renaming
of JCasAnnotatorAdapter) but they won't be willing to upgrade their UIMA from
2.3.0 to 2.3.1 where there are no backwards incompatible changes? That seems
unlikely to me...
Original comment by steven.b...@gmail.com
on 14 Apr 2011 at 8:58
Well, for me personally it doesn't matter. I have upgraded to 2.3.1 already. If
the consensus is to upgrade, I'm not opposing more actively than stating why I
personally did not suggest such an upgrade in uimaFIT.
Original comment by richard.eckart
on 14 Apr 2011 at 9:29
I lean towards upgrading to 2.3.1. It seems that a user could switch back to
2.3.0 if they had to - but that is not really our concern.
Original comment by phi...@ogren.info
on 15 Apr 2011 at 4:47
I guess the other reason I lean towards upgrading to 2.3.1 is that its the
first "real" release in the sense that 2.3.0 was an incubating release, but
2.3.1 isn't.
Original comment by steven.b...@gmail.com
on 15 Apr 2011 at 6:30
Well, it sounds like a +1 from me, a -0 from Richard and a +0 from Philip, so
I'm going to go ahead with it. Anyone who really still needs the
2.3.0-incubating release instead of 2.3.1 can achieve that by appropriate
exclusions in their pom.xml.
Original comment by steven.b...@gmail.com
on 15 Apr 2011 at 8:25
This issue was closed by revision r598.
Original comment by steven.b...@gmail.com
on 15 Apr 2011 at 8:26
Original comment by richard.eckart
on 8 May 2011 at 10:41
Original issue reported on code.google.com by
steven.b...@gmail.com
on 14 Apr 2011 at 1:44