Some economic analysis of Apple's move to x86

[Mike Hearn]

On Sat, 2005-06-11 at 11:40 -0700, Scott Ritchie wrote:
> If it works for Wine on Mac, why wouldn't it work for Wine on Linux?

Well, Mac users/developers tend to write Mac software which isn't
portable anywhere. That's my primary concern.

> Maybe if they do some sorta half-baked frankenstein port mixing APIs,
> but in that case all Wine did was reduce the expense of porting to Mac,
> rather than raise the expense of porting to Linux.  Such a port,
> incidentally, would mean Wine was missing a key part of the Windows API,
> and extending Wine rather than converting to use the Mac APIs may be a
> lot more efficient.

A (good) Mac port of an app basically implies rewriting the GUI anyway,
you can't get native look'n'feel without it. That's not *such* a huge
deal on Linux, partly because expectations are lower, and partly because
the GUI is more similar to Windows anyway so you can do a lot more with
theming and such.

> Not releasing a working port is...a strange business decision.  

They didn't want the tech support problems, unfortunately our reputation
for crappy binary portability has this effect sometimes ...

> Well, the good news is that whatever it means our strategy in the coming
> months is still the same - make Wine a damn fine piece of software for
> Linux.

Yes, that'll hopefully be my focus as well. My primary concern is that
as software is ported or written from scratch, it targets open source
APIs as then that software can easily run on Linux, MacOS, Windows,
SkyOS, whatever. Whereas if people are running proprietary operating
systems they tend to write apps for large and hard to emulate APIs like
Win32 or Carbon/Cocoa, which isn't good for the free market.

thanks -mike