[1800-whatever]

Hi,

Anybody here that has experience with, or knows how difficult it is to port applications from Linux to OS X?

[depolitic]

More details!!!

Is it a server app, desktop app, X11 app etc Is it C or C++ or Java - Also see this

http://developer.apple.com/documenta...section_1.html

[1800-whatever]

o.k. depolitic,

It's about this application, I would LOVE this to be ported to the Mac
http://www.plasticanimationpaper.dk/
at the moment there only is a Windows version and a Linux version in beta.

I stumbled upon this thread in the forums :
http://www.plasticanimationpaper.dk/...opic.php?t=767

No it looks like they are willing to port it to the Mac but don't know how to go about it.
Can I ask anybody that knows to do me a big favour and jump in here ?

(You don't seem to have to be registered to post in their forum. )

[depolitic]

Just send them to:

http://developer.apple.com/

I would guess they have three basic choices, - they could take the Linux version and with a small recompile it should run under X11 for the Mac this would be the simplest way. However this is not a very user friendly way to run the app under a macintosh.

Next they would be using carbon API, to port it just Like photoshop is written. This will give you an native Aqua Application, with a minimum work.

http://developer.apple.com/carbon/

Or they could go the the Cocoa path which will give them the most feature rich Mac application, but probably means a little more commitment on their part. However this will attract the highest interest from macintosh users. And will future proof the application.

http://developer.apple.com/cocoa/

Also they may be interested in this OpenGL info for the mac -- http://developer.apple.com/opengl/

These guy's should be able to find their way from their. Then get a couple of mac beta testers and a petition going and you may get them going.

[Angus_D]

Originally posted by depolitic:
Or they could go the the Cocoa path which will give them the most feature rich Mac application, but probably means a little more commitment on their part. However this will attract the highest interest from macintosh users.

This is just total bull. You can create a "feature rich" Carbon application that fully adopts the Mac OS X user experience. Users, more often than not, don't have a clue which framework the application uses. This is the way it should be: they shouldn't care.

[Thinine]

Carbon is dead. Especially for Tiger, if you aren't using Cocoa, you're missing a lot a functionality Cocoa apps will get nearly for free. Now, if it's a good Carbon, that may be fine, but it'll be more future proof if they can make at least a Cocoa wrapper.

[Catfish_Man]

Originally posted by Thinine:
Carbon is dead. Especially for Tiger, if you aren't using Cocoa, you're missing a lot a functionality Cocoa apps will get nearly for free. Now, if it's a good Carbon, that may be fine, but it'll be more future proof if they can make at least a Cocoa wrapper.

Eh? Every version of Carbon so far has lessened the differences in capabilities between it and Cocoa. Quickdraw in Carbon, however, is definitely not the way to go these days (especially once 10.4 comes out).

[Angus_D]

Actually, I am starting to be of the opinion that the writing is on the wall for Carbon. Unless the two camps can work something out such that Cocoa uses Carbon more extensively (so Cocoa becomes an application framework built upon the "low-level" Carbon API, moving into a position more akin to the old MacApp), Carbon will be going away.

[techtrucker]

I have to agree with Angus_D, I too have the feeling that Carbon will pe phased out. Granted, it will take a looong time (MS Office, Adobe CS, Macromedia, to name a few heavy hitters),but if I were starting a Mac app today it would have to be Cocoa. As hard as it is to accept sometimes we have to deal with losing backward compatibility in the interest of forward progress.
To "correctly" port a Linux app, IMHO, would involve writing it in Cocoa, and either translating the program logic or if you're lucky in some cases copy and paste existing C code.