One Lucene issue I'd like to have in 9.0 is https://issues.apache.org/jira/browse/LUCENE-9047, which is about making our directory abstractions little-endian instead of big-endian. It would be breaking enough that it should be done in a major.
On Sun, Feb 2, 2020 at 3:35 PM Erick Erickson <erickerick...@gmail.com> wrote: > What are people’s thoughts about Solr 9.0? It’d be a little faster than > our recent cadence, the last two major releases were about 18 months after > the one before and we released Solr 8.0 last March. > > Besides the usual reasons, there are two drivers I can think of: > > 1> We can drop Java 8 compatibility. > 2> We can migrate to exclusively use Gradle as the build system. > > The Gradle build isn’t complete yet, although it’s maturing pretty quickly > (and Dawid has yeoman’s duty!). My question isn’t so much “should we start > the release process for Solr 9.0 next week” as it is “What milestones > should we reach before starting the 9.0 release process?” > > Maybe something as simple as “Start the 9.0 process a month after removing > the ant components of the build system”. > > And a reasonable answer at this point is “it’s way too early to even talk > about it”. > > Erick > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org > For additional commands, e-mail: dev-h...@lucene.apache.org > > -- Adrien
