Nice observation by Jason Snell from the Apple quarterly report conference call:
“It probably means nothing, but when Maestri listed off Apple’s four major product platforms, he said this: ‘We’ve built a huge installed base around four platforms: iOS, Mac OS, watchOS, and tvOS.’”
Like Jason, I’ve long wanted a return to “Mac” as the most important part of the OS name, and have suggested it a couple times on Core Intuition. It was a missed opportunity to do this transition after 10.9, when it could have cleanly gone to Mac OS 11 without the .10 and .11 silliness.
The new tvOS and watchOS branding — combined with Apple’s quote above — makes an official rebranding to “macOS” at WWDC this year seem almost likely. The next major version should be macOS 11, without the “X” and “10.x”. That would still look a little wrong compared to simply “Mac OS”, but it would be much better than “OS X”, and the lowercase would be consistent with the rest of the platforms.
Brent Simmons has a pair of posts on dropping support for older OS releases and how the upgraded percentage goes up quickly:
“When making decisions like this, I don’t think about what conditions are were I to ship today — I think about what conditions will be like when I actually do ship, and I think about conditions six months out as we do support, testing, and maintenance releases.”
Good advice. And remember, very few developers actually ship on time. Even if you think your app is going to ship before the next major version of the OS does, your app will probably be late. More people will be able to run it than you first expected.
“At VitalSource”:http://www.vitalsource.com/ we now have a dozen Xserves running Ruby on Rails and a couple others running MySQL. While it’s mostly stable now, over the years there have been several mystery show-stopper problems that no one seems to have on other platforms.
Which is why I found this “quote from Rentzsch”:http://rentzsch.tumblr.com/post/316835872/mac-os-x-server-a-fear-realized so interesting:
“The trick is to radically minimize all services performed by Mac OS X Server itself, and instead run services in VMware-hosted Debian images. At that point, Mac OS X Server becomes little more than a simple container for VMware, something it seems able to handle with a minimum of tantrums.”
It seems kind of wasteful, but Mac OS X Server really is overkill most of the time.
Related: over the weekend I checked into “Heroku”:http://www.heroku.com/. I’m impressed with what they have built. If it works as advertised, I think I’ll supplement my Dreamhost stuff with Heroku any time I need a Ruby backend.