Honestly starting to feel like the past 4 years of OS X changes have just been too much burden for the few remaining OS X software developers across the board. I recall when the Mac App Store came about a lot of developers moved their software to it, then sandboxing happened and that seemed like an absolute nightmare for most to implement, then after months of that they started to leave the App Store and then there's Gatekeeper too.
Now I'm not saying these are bad things, I definitely see the benefits but there just isn't the mass of engineering talent or money in building OS X software to justify the work required.
This has been tangible to me, 6 years ago my Mac was running all Cocoa apps apart from Creative Suite. Today all the apps I'm running are Electron apart from Textmate and Creative Suite.
The neglect of the Mac platform and the divvying up of the small Mac userbase into Mac users, iPad users and iPhone users means the future of the OS X app experience is just going to be running Javascript apps in Chrome wrappers.
I agree that the basics are excellent (alignment, copy & paste, sizing, ...) and still best in class.
Think Figma took them completely by surprise, Sketch's success was built on the user experience of a cocoa app feeling much more solid and slick than antiquated Adobe software, don't think they ever expected webtech to be able to compete but obviously with the force behind it and OS X being stagnant, it did.
- Lack of macOS staffing meaning that it's very much in maintenance mode (it was "make sandboxing work" in 2013 and "make dark mode work" recently), which parallels general pushes from Apple on iOS and iPadOS.
- Lack of need for new features: I think Omni could have pivoted towards Sketch/Figma style "export to mobile" (which would have actually probably meshed really well with the rest of the mobile app dev business, since we did mocks in OmniGraffle), but I'm curious what else Graffle really needs to be useful. I think historically there has been enough net new customers to not need to provide all that much upgrade incentive.
- In 2013 the codebase was a beast, and this was like v4 or v5... I can only imagine adding major new features to it.