Skip to main content
4 of 5
added 97 characters in body
bmike
  • 241.3k
  • 80
  • 433
  • 958

My recommendation is you’ll need to run Xcode 9 on your mini and then get one used Mac for every other contributor. They can connect to the build server on iOS 9 and either use Xcode themselves or Screen Share as the second account each Mac allows fully supported.

You will waste far more effort, suffer low quality code and delay working around tooling issues than you’ll gain by not funding your development team with tools that work. Buy broken Mac at scrap prices and fix them up or cook dinner and trade labor with someone that can repair Macs. Do anything you can think of to get your team Xcode running on macOS.

It’s like training dentists remotely with no tools but books and then hoping they perform as part of a team still without tools. Your customers will not be fooled when they see the results. Worse, if your iOS app is going to the App Store, you’re competing against every team that had funding for 4 years college, the best training and tools.


Now, I hope to be proven wrong soon. I think we’re a year or so away from the Xcode server to be good enough to really run as the build server João hopes to find, but in October 2017 they aren’t ready. The iPad should be a $250 solo, design, test, build machine as swift playgrounds mature. Once that is hooked into Xcode in the cloud or running remotely, it should be game on for windows, Linux, iOS and Mac to be on somewhat more equal footing to program for iOS since someone might invest the effort to reverse engineer the remote Xcode programming functions. Today, Mac is privileged to develop for iOS and to operate otherwise puts your team at a huge disadvantage.

For low cost programming options, you might need to scope to python or swift alone without the iOS framworks to keep the budget low for 1:1 hardware deployments that you could fund with a grant or course fees.

bmike
  • 241.3k
  • 80
  • 433
  • 958