On 12/15/12 4:07 AM, Gilles Sadowski wrote: > On Fri, Dec 14, 2012 at 10:03:14PM -0800, Phil Steitz wrote: >> The classes in .optimization.fitting seem to have been duplicated / >> moved to .fitting, but the originals are not deprecated (though some >> methods in them are). Shouldn't we deprecate the original classes? > Yes. In fact, all the classes under "o.a.c.m.optimization" are deprecated; > I intended to mark them as such, but somehow got distracted by other > things... ;-) > Thanks for the reminder. > >> Also, @since for the moved versions should be 3.1, correct? > It depends what is meant by "@since"... Is it the appearance of a new > feature (algorithm or data structure), or the actual naming or renaming > of a class or method? [There are arguments for each, but both cannot be > accomodated with a single tag.]
Good point. I would say leave the moved version @since tags as is. > > In the former case, the answer to your question would be no since it is a > only a refactoring of existing features. > > In the latter, we'd have a problem: many "@since" tags are 1.2, 2.0, ... > whereas they should _all_ be 3.0 because of the package name change (from > "o.a.c.math" to "o.a.c.math3"). > Since that had not been a blocker for a major release, I propose that we > postpone the resolution of this situation to the next release (opening a > JIRA report) or until we are sure how to best use "@since" must be used in > CM. I need a couple more hours to review and fill in the rest of the missing ones. Phil > > > Regards, > Gilles > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
