First post, by lowe0
Before I begin: I'm carefully writing this to not make it a "is DOSBox dead?" or "where's 0.75?" post. For the sake of not getting it locked, please consider the same when replying.
Is a traditional release cycle still the best use of the DOSBox developers' time and effort?
Right now, think of the many ways you can get DOSBox. There's the original download from the website, of course, but there's also:
- bundled as part of a GOG, Steam, or EA release
- on Linux, via your distribution's package repository
- as a core for pluggable emulators (libretro or similar)
- from your preferred fork (DOSBox-X, etc.)
- built from source
It looks to me like there are already plenty of ways for SVN commits newer than 0.74.x to get into end users' hands. GOG has always had the option of cherry-picking whatever's best for each individual game. Package managers can pull in from SVN whenever they like, in accordance with their distribution's philosophy (less often for Debian, more often for Arch, etc.). And forks are gonna do whatever they want within the terms of the license.
Would it make more sense to develop DOSBox as a reference implementation, and let others deal with the release process?