A Brief Continuation of Series 3
I gather that there is some reluctance towards the idea, but I feel that the current series could benefit from a brief continuation (not an indefinite one), hence this appeal to those in charge for a MuseScore 3.7 to be issued.
Some points I’d put forth in justification:
- It could minimise discrepancies and bridge gaps between Series 3 and 4
- It would become available to download and update from Windows Store (the current offering is from several versions ago)
- In terms of third party software (operating systems; Qt; etc) and hardware (e.g. MIDI keyboard), we can better ensure compliance (standards, etc) and take advantage of their improvements (e.g. for accessibility)
- Support for older operating systems that cannot run Series 4 (users not upgrading their machine - for whatever reason)
- Improved support (i.e. presentation; MusicXML import) for Official Scores (on .com)
- Improved synchronisation between desktop, .com, and mobile apps
- Conservative users may continue using Series 3 as 4 stabilises (and it may not be available until the first half of 2022)
- General refinements from within (with an emphasis on fixes over features)
- An opportunity to address unexpected issues that arose, partly a result of rushed releases and a lack of testing
- The forum and issue tracker will become less clogged (fewer complaints; fixed issues)
Comments
One suggestion has been to assemble a team of non-staff developers to oversee this endeavour (as to avoid diverting time and resources away from the upcoming Series 4), so I’d propose the following: after reaching an agreement on and implementing the pull request produced for this, we’d issue betas, release candidates, and then the stable item; each of these is tested privately prior to public release and within a reasonable time period.
A subsequent 3.7.1 is probable, but with a focus on oversights or anything unforeseen. Further development of Series 3 would be drawn to a close thereafter unless severe or frequent issues arise.
It was originally proposed to version an update as 3.6.3 (and continues to be suggested, albeit with fewer changes to minimise the risk of regressions), but the quantity and context of changes suggest 3.7 is more fitting.
I should say that I’m not in favour of an unofficial fork because the improvements would only be found by a few and not recognised in official MuseScore services (i.e. the website or apps), and problems could arise later if there are attempts to open it in them.
I think the Report a Bug function ought to be removed. Aside from my argument about bad reports, it’s not evergreen - and it especially becomes irrelevant for a soon-to-be-retired series.
As I've said numerous times and ever since 3.6.2 got released: I'm all for it!
See https://github.com/musescore/MuseScore/issues/7449 for a list of issues and potential fixes and https://github.com/musescore/MuseScore/pull/9000 for a huge PR combining most of those (and beyond).
One problem is to actually get a release build, for that we'd need a staff member, a 'mere mortal' can't build a release (or Beta or RC) and publish it (unless given approrpriate priviledges and some additional knowledge).