Gregory's Server@ocdtrekkie @bagder Yeah, I know, the backwards compatibility is on the one hand awesome, on the other hand difficult.
|
3 comments
@ocdtrekkie @autinerd @bagder Imo software should be light, neat, minimal, and fit for purpose. To cover all needs it rarely ends up being minimal, but for os’s there should be one that most use as their daily driver and focuses on fast development good stability and performance. A second that is for compatibility for ancient commercial software, weird hardware, devices, interfaces etc. Windows 10 doesn’t need the bloat those who need it are on xp. I like nixos as it can be anything you want.  @halikular @ocdtrekkie @autinerd @bagder That is not commercially viable. You're essentially asking to split a development team in two. And the team that does new things would be starved of resources because the team focusing on compatibility has much more work to do. Ironically, having it in one codebase and one team means you don't do *too much* of the compat work and you balance how much of that you should actually do. |
@autinerd @bagder I really enjoy the details about like, how Windows devs would special case stuff in Windows 95 to do something legacy when it detected SimCity was running so not to break compatibility with the Windows 3.1 version.
It feels truly alien today to take backwards compatibility to the level Windows traditionally has. Not sure if that's a good or bad thing though honestly.