On Mon, Oct 21, 2013 at 12:28 PM, Henri Yandell <flame...@gmail.com> wrote:
> > > > On Mon, Oct 21, 2013 at 7:29 AM, sebb <seb...@gmail.com> wrote: > >> On 21 October 2013 11:52, Benedikt Ritter <benerit...@gmail.com> wrote: >> > >> > >> > Send from my mobile device >> > >> >> Am 21.10.2013 um 03:46 schrieb sebb <seb...@gmail.com>: >> >> >> >>> On 20 October 2013 15:03, Benedikt Ritter <brit...@apache.org> wrote: >> >>> I agree. If we don't deprecate it now, and agree to release the next >> major >> >>> version targeting Java 7, we would remove those methods without ever >> >>> mentioning it before. >> >> >> >> That's not how I see it working. >> >> >> >> I think the deprecations should be added once the code requires a >> >> minimum of Java 7. >> >> Later on, the deprecated methods are removed if required (they could >> be left). >> >> >> >> In any case, removal of the deprecated methods is not binary >> >> compatible, so new package/Maven coords are needed. >> >> In which case, it's not really a problem that the methods are not >> >> deprecated first. >> >> It would be sufficient to note the replacements in the release notes. >> >> >> >> Deprecation is only useful to users of a library if there is a >> >> replacement they can use. >> > >> > There is a replacement as Hen has pointed out. What you're saying is >> that the replacement has to be part of the library, right? >> >> Not necessarily, the replacement could be part of standard Java classes. >> >> But I don't think it's right to require users to migrate to a later >> version of Java than is required by the library itself in order to >> avoid the deprecation warning. >> >> And as I already wrote, it's important that deprecation warnings are >> removed (not suppressed) in the library itself. >> That is necessary to show that the deprecation makes sense. > > > What's your solution, Sebb, to indicate that we plan to remove this code > in 4.0? > Or I could need sleep. I was on the subject of deprecating the time package. :( Hen
