Very interesting post detailing what was needed to bring a very old, very complex application to the new Macs. Kudos to the Mozilla folks for being on top of it hand having a build ready very close to the hardware release.
Mozilla is probably in a better position than some folks. There’s a lot of legacy in their codebase, but part of that is vestigial support for a huge number of operating systems and architectures. As I recall, Adobe was one of the worst hit by both OS X and the Intel switch because most of their products were tightly coupled to old macOS APIs and a lot of their fast paths were carefully optimised for PowerPC AltiVec. The Windows and Mac versions of their products shared far less code than you might expect, so they couldn’t just use the x86 code paths they were using on Windows for a bunch of things and when Apple killed Carbon they didn’t have platform abstraction layers for the services that it provided and so had to do a major rewrite to get to Cocoa.
Very interesting post detailing what was needed to bring a very old, very complex application to the new Macs. Kudos to the Mozilla folks for being on top of it hand having a build ready very close to the hardware release.
Mozilla is probably in a better position than some folks. There’s a lot of legacy in their codebase, but part of that is vestigial support for a huge number of operating systems and architectures. As I recall, Adobe was one of the worst hit by both OS X and the Intel switch because most of their products were tightly coupled to old macOS APIs and a lot of their fast paths were carefully optimised for PowerPC AltiVec. The Windows and Mac versions of their products shared far less code than you might expect, so they couldn’t just use the x86 code paths they were using on Windows for a bunch of things and when Apple killed Carbon they didn’t have platform abstraction layers for the services that it provided and so had to do a major rewrite to get to Cocoa.