Hi Matt, All, Totally agree on starting 1.0 and the version naming scheme.
With regards to noise about 1.0: there's no such thing as too much noise. - I'll start working on a release announcement for everything that has happened since mid 2019. - I started working on a 60-90 second video with the 1.0 highlights. Of course all input and any other help is welcome. - A 1.0 version of HOP-3259 [1] could be another option. - in addition to announcing our releases on announce@, we could also check if we post to press@ - mailings, PDF versions of the 1.0 website page(s), tweets, linkedin posts, third party blog posts, ... The most important thing probably is to get the word about 1.0 out to and through as many channels as possible. Mentors, are there any tips or experiences with previous first major releases you can share? [1] https://issues.apache.org/jira/browse/HOP-3259 Regards, Bart On Fri, Sep 10, 2021 at 4:47 PM Matt Casters <[email protected]> wrote: > Hi Hops! > > The list of tickets > <https://issues.apache.org/jira/projects/HOP/versions/12350233> for 1.0 > seems to be getting shorter every day so I think it's time to release our > iconic 1.0 lest we want to end up in a Duke Nukem Forever scenario. > > Anyway, Hop is looking fantastic and I detect some pressure left and right > to dig in again with new and bigger features and improvements. Doing those > on 1.0 seems like a bad idea. Let's cut a 1.0 release instead somewhere > next week and move the master snapshot version to 1.1.0 > > I would propose to do the sensible and simple thing and release minor > versions going forward in a Major.Minor.Patch version naming scheme. I > know a lot of other projects like Beam only release patch level 0 but it > doesn't hurt to have it just in case we make a horrible mistake > <https://issues.apache.org/jira/browse/VFS-807> along the way. > > The major versions I would reserve for the larger architectural changes and > in fact I wouldn't wait too long to start working on 2.x with support for > Java 11. > > As far as compatibility is concerned I would first and foremost focus on > the Hop execution engines on existing metadata. The integration tests are > our best friends. Java API changes shouldn't concern us too much unless > they occur in the core of our stack and IMO are of lesser importance for > plugins. > > Ideas for making noise about 1.0 are also welcome! > > Cheers, > Matt >
