With our numbering scheme for 1.3.0-snapshot we are driving ourselves and
our users crazy: how can we still have 1.3.0-snapshots when 1.3.0 is final?
1.3-snapshot is clear: it is the current status for 1.3-head, it is also the
default numbering scheme used for maven version ranges. When we now update
our poms to reflect 1.3.1-snapshot status, *everybody* that wants to use
snapshots needs to update their poms, whereas it used to be 1.3-snapshot and
automagically worked.
Therefore I propose to from now and onto eternity to never use 1.x.0 as a
version number and use the common, sane version numbering strategy which
reads:
1.3-SNAPSHOT > 1.3.x > 1.3 > 1.2-SNAPSHOT > 1.2.x > 1.2 > 1.1-SNAPSHOT >
1.1.x > 1.1
So for 2.0, or 1.4, or whatever version we are going to use for our next
release (let's use 2.0 as an example) I propose the following version in our
trunk/pom.xml (not applying it *now*, but when we have 1.3.1 or 1.3.2 out
the door):
<version>2.0-SNAPSHOT</version>
I also propose to re-version current head to:
<version>1.3-SNAPSHOT</version>
Which will remove any and all confusion whether someone needs
1.3.0-SNAPSHOTto get snapshots,
1.3.1-SNAPSHOT, 1.3.2-SNAPSHOT, 1.3.3-SNAPSHOT, 1.3.4-SNAPSHOT,
1.3.5-SNAPSHOT, 1.3.6-SNAPSHOT, 1.3.7-SNAPSHOT, 1.3.8-SNAPSHOT,
1.3.9-SNAPSHOT, 1.3.10-SNAPSHOT, ad infinitum.
[ ] Yes, this numbering scheme seems reasonable
[ ] No, I like what we have now because....
Martijn
--
Buy Wicket in Action: http://manning.com/dashorst
Apache Wicket 1.3.0 is released
Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.0
