Welcome to the MacNN Forums.

rmendis · Mar 16, 2003, 03:50 AM

Originally posted by Mediaman_12:
The Copland project was attempting to do exactly what you guys want (a next gen OS that doesn't break any old apps), and that never achived anything...

This is *not* Copland.

I think that fiasco and wasted R&D at Apple has left many people skeptical or pessimistic about the development and advances of Mac OS technology.

(In some ways those disasters paved the way for the return of Steve Jobs to Apple along with the technology he developed at NeXT - and of course the true advance in OS - Mac OS X).

However, to develop a 'Classic mode' for Mac OS X is not all that sophisticated or complicated as even i first thought.

As i said, the technology is there already...
1. BlueBox full screen mode on login to a 'Classic mode' user account
2. Running Carbon apps inside BlueBox
3. (Possibly) Carbon Mac OS X Finder - for it realy to be a Mac OS X Classic Mode rather than simply running Mac OS 9.
4. (Possibly) Cocoa running on Classic...to share Applications, etc.. between the two environments.

The only new thing it may require is that the BlueBox be implemented as a lowerlevel process instead of a Cocoa app.

What isn't necessary (though I thought it might be before) is to host Mac OS 9/Classic kernel on Mach. This has in the past been the stumbling block for Apple and could be the point on which people seem to dwell on

JLL · Mar 16, 2003, 06:35 AM

Originally posted by Amorph:
What I'm hoping for is for AppleScript 2 (more precisely, OSA) to be built on CoreFoundation, so that sharing data types between AS, Carbon, Cocoa and Java is all basically transparent. If auxiliary languages like Perl, Python, Ruby et al can join in on the fun, so much the better. But for the sake of writing and automating scriptable applications, it would be amazing to have the same data structure look like an NSDictionary to Cocoa and a record in AppleScript.

You do know that AS2.0 is released in beta form?

JLL · Mar 16, 2003, 06:38 AM

Originally posted by Terri:
Well just today I had a printer return a bunch of G4s because they said it was too confusing for their production people to go back and forth between Classic and X. They are buying a bunch of Mac 9 booting machines that they found someplace.

IMHO Apple blew it by making Mac OS X work so different. How hard would it have been to make X look and work more like 9. Don't get me started on the Open/Save dialog boxes in X.

I think you blew it buy forcing Mac OS X machines on people that run Mac OS 9 apps.

calumr · Mar 16, 2003, 07:03 AM

Originally posted by rmendis:
The only new thing it may require is that the BlueBox be implemented as a lowerlevel process instead of a Cocoa app.

Aaaarrgh! This thread! It burnses my eyeses!

Seriously - Apple showed Mac OS 9 in a coffin at the WWDC for a reason - they want to move forward. Letting developers continue to get away with not porting their legacy apps is not what Apple want to encourage.

Sure, there are probably a few ancient Classic applications that are no longer being developed for whatever reason, but if you absolutely have to run them then just keep some old hardware around that can boot into OS 9, but please don't continue this technically misinformed thread ("BlueBox be implemented as a lowerlevel process instead of a Cocoa app" for a good example).

Terri · Mar 16, 2003, 08:29 AM

Originally posted by JLL:
I think you blew it buy forcing Mac OS X machines on people that run Mac OS 9 apps.

The latest machines from Apple do not boot anything, but X. What are people suppose to do when they want a new machine?

Printers get files from lots of places that they have to output. Most are in Quark, but many are also in PageMaker. Are they suppose to never upgrade their hardware?

All I am saying is that by not making the interface so different in X a lot of these problems could have been avoided.

JLL · 08:43 AM

Originally posted by Terri:
The latest machines from Apple do not boot anything, but X. What are people suppose to do when they want a new machine?

Apple's still selling a Power Mac that can boot in Mac OS 9!

Printers get files from lots of places that they have to output. Most are in Quark, but many are also in PageMaker. Are they suppose to never upgrade their hardware?

Just as they have to have a Windows machine for handling files from Windows only apps, they could keep a couple of Mac OS 9 machines for Mac OS 9 only apps.

The Mac OS 9 apps don't require the latest and the greatest hardware, so even a couple of Beige G3s will do the job perfectly.

All I am saying is that by not making the interface so different in X a lot of these problems could have been avoided.

It's not that different and I think you're making a huge mistake by installing all those apps you do to make Mac OS X "work" like Mac OS 9 - you're making it more confusing than it needs to be.

Terri · Mar 16, 2003, 08:57 AM

The Mac OS 9 apps don't require the latest and the greatest hardware, so even a couple of Beige G3s will do the job perfectly.

It's not that different and I think you're making a huge mistake by installing all those apps you do to make Mac OS X "work" like Mac OS 9 - you're making it more confusing that it needs to be. [/B]

You mean things like Labels, WindowShade, FruitMenu and the Application menu. Sorry, but no one can seem to find the built in Labels in X and the dock turns into a giant mess very quickly in a heavy production work.

75% of their work comes in as Quark docs, often with lots of problems. The rest comes in as PageMaker, Illustrator, often old versions that the newer Illustrator does not output correctly.

They want the latest, fastest hardware since time is money. Also each machine costs money to support so less machines mean lower support costs.

They want and need Mac OS X, but will need to run classic for a long time. Using old hardware is not the answer, but Apple is not giving them what they need in X so they are turning to 3rd party hacks for their needs.

JLL · Mar 16, 2003, 09:31 AM

Originally posted by Terri:
You mean things like Labels, WindowShade, FruitMenu and the Application menu. Sorry, but no one can seem to find the built in Labels in X and the dock turns into a giant mess very quickly in a heavy production work.

By installing an app menu, the user is presented with two ways of handling open apps - that is not very user friendly.

By installing an Fruit menu, the user is presented with two ways of launching apps - that is not very user friendly.

75% of their work comes in as Quark docs, often with lots of problems. The rest comes in as PageMaker, Illustrator, often old versions that the newer Illustrator does not output correctly.

OK, this sounds like a Mac OS 9 only operation - why even think of introducing Mac OS X?

They want the latest, fastest hardware since time is money.

Quark isn't much faster on a G4 than on a 604 - and you're wasting your clients money by using DP machines.

Also each machine costs money to support so less machines mean lower support costs.

Then stick with Mac OS 9.

They want and need Mac OS X.

For what? QXP, PageMaker and old Illustrator versions?

And didn't they return the machines? Couldn't have wanted it that bad.

rmendis · Mar 16, 2003, 09:39 AM

Originally posted by calumr:
Seriously - Apple showed Mac OS 9 in a coffin at the WWDC for a reason - they want to move forward. Letting developers continue to get away with not porting their legacy apps is not what Apple want to encourage.

Yes, but pretty much all major developers have ported to Mac OS X already.

So it would be no big deal if Apple were to release this 'Classic moed' feature of Mac OS X subtly. It doesn't/won't require much more work and it is an excuse in fact to remove Classic envionment from Mac OS X.

Encouraging the use of one or the other environment now would more likely work to Mac OS X's (Aqua) advantage.

Also the reason Mac OS 9 was showcased in a coffin at WWDC and *not* at MacWorld is that Mac OS 9/Classic *API* is dead. That is to developers it is...not to users.

JLL · Mar 16, 2003, 09:42 AM

Originally posted by rmendis:
So it would be no big deal if Apple were to release this 'Classic moed' feature of Mac OS X subtly. It doesn't/won't require much more work and it is an excuse in fact to remove Classic envionment from Mac OS X.

What makes you think that it doesn't require much work?

Terri · 10:03 AM

By installing an Fruit menu, the user is presented with two ways of launching apps - that is not very user friendly.

They often keep the Dock hidden and this makes Classic and X work the same, nothing confusing here.

OK, this sounds like a Mac OS 9 only operation - why even think of introducing Mac OS X?

Photoshop, what do you think they use to edit the images? Illustrator, what do you think they use to edit the illustrations? Not to mention all the other apps they are also running. Not having the machine go down every time an app crashes is a real plus as well.

Quark isn't much faster on a G4 than on a 604 - and you're wasting your clients money by using DP machines.

Faster networking, faster postscripting to the RIPs, FireWire 800, being able to read the CDs from the newer machines, the list keeps on growing each month.

They returned the machines for mirrored doors machines that do boot 9. Ironically the mirrored doors machines seem less stable in 9 then the earlier machines. In fact they seem almost unusable in 9 with constant screen freezes, even with 9 stripped down to almost nothing.

Then there is the problem of machines that are running 9 seem to have a problem reading many of the CDs that are burnt under X.

Staying in 9 is not the answer, but some apps, like PageMaker, will never be updated.

Wasting money on making 9 more like X is not the answer. So far the best solution seems to be making X work more like 9 so that there is not so much confusion going back and forth.

I don't think some people realize how important some of 9's features, like Labels and WindowShade are in production.

Terri · Mar 16, 2003, 10:02 AM

So it would be no big deal if Apple were to release this 'Classic moed' feature of Mac OS X subtly. It doesn't/won't require much more work and it is an excuse in fact to remove Classic envionment from Mac OS X. [/B]

Now there is a nice simple answer.

What is the big deal to bring some of 9's features to X?

Seeing how many people I know are running many of the hacks that bring back 9's features it seems to be what many people, myself included, want.

If you don't like them then don't use them.

So far I have yet to find a replacement for tabbed windows and don't get me started on the almost unusable open/save dialogs.

JLL · Mar 16, 2003, 10:44 AM

Originally posted by Terri:
Photoshop, what do you think they use to edit the images? Illustrator, what do you think they use to edit the illustrations?

Both apps runs just as good (if not better) in Mac OS 9.

PWasting money on making 9 more like X is not the answer. So far the best solution seems to be making X work more like 9 so that there is not so much confusion going back and forth.

Again, my advice would be, that they should keep one or two Mac OS 9 machines for handling Quark and PageMaker.

The right tool for the right job - I bet that they don't run one color jobs on four color Heidelbergs.

JLL · Mar 16, 2003, 10:46 AM

Originally posted by Terri:
Now there is a nice simple answer.

What is the big deal to bring some of 9's features to X?

Seeing how many people I know are running many of the hacks that bring back 9's features it seems to be what many people, myself included, want.

This thread is not about bringing Mac OS 9 features to Mac OS X, but about integrating Classic with Mac OS X more than it is today.

Terri · 11:29 AM

Originally posted by JLL:
Both apps runs just as good (if not better) in Mac OS 9.

Guess you haven't used Illustrator 10 much, it's a real dog in 9, but runs very nice under X. Photoshop is also better with the huge files that they have to open under X.

Mac OS X is the right choice, classic works very nice. The problem is the interface of X, the production people find it lacking.

Again, my advice would be, that they should keep one or two Mac OS 9 machines for handling Quark and PageMaker.

Did you even read my post? They are sick of a bad font crashing Quark and taking down the whole machine. Anyway Quark 6 will be out soon and that will be for X only which still leaves PageMaker, FrameMaker and lots of other apps that will most likely never be ported to X.

They also don't want to pay the support costs of having additional machines running 9. The machines runniung X almost never crash or cause any problems other then the fact that the production people don't like using the interface as Apple ships it.

The right tool for the right job - I bet that they don't run one color jobs on four color Heidelbergs. [/B]

They have 6 color Heidelbergs and direct to plate Xerox machines that run some kind on *nix RIP. It is unbelievable how much faster machines running X postscript to these machines. They also have a bunch of other presses and film machines.

There is also the problem of the fact that they run Mac OS X servers and machines running 9 have problems with them, mostly with permissions.

The more operating systems you run the more problems you have the more it costs to support them.

rmendis · Mar 16, 2003, 11:33 AM

Originally posted by calumr:
...technically misinformed thread ("BlueBox be implemented as a lowerlevel process instead of a Cocoa app" for a good example).

Normally I don't respond to flaming, however, perhaps I didn't make myself clear.

You know it is possible to have a BSD/Darwin tool or app?
There are several GUI apps/enviroments that run on 'command line' OSs. Like those on Linux for example.

(In fact, you can even build 'faceless applications' or tools using Cocoa that don't use the AppKit. i.e basically Foundation/CoreFoundation stuff. Some of the UNIX like functions that shipped with OPENSTEP for Windows - Cocoa for Windows if you like, were built this way).

So, I don't see why you think it would be impossible to re-create a full-screen mode BluBox environment that is a BSD/Darwin process. After all it doesn't require Quartz or Cocoa to dispaly itself.

JLL · Mar 16, 2003, 11:47 AM

Originally posted by Terri:
They have 6 color Heidelbergs and direct to plate Xerox machines that run some kind on *nix RIP. It is unbelievable how much faster machines running X postscript to these machines. They also have a bunch of other presses and film machines.

You totally missed the point!

There is also the problem of the fact that they run Mac OS X servers and machines running 9 have problems with them, mostly with permissions.

No problems here.

calumr · Mar 16, 2003, 12:11 PM

Originally posted by rmendis:

So, I don't see why you think it would be impossible to re-create a full-screen mode BluBox environment that is a BSD/Darwin process. After all it doesn't require Quartz or Cocoa to dispaly itself.

I can see that you want better classic compatibility, and I can see why some people would want that. And I'm sure that Apple or some determined developer could write something similar to what you described just now, but that was not the inaccuracy that I was describing.

You said that BlueBox is a Cocoa app, when it is not, that's what I meant by inaccurate.

I suppose that all I want to get across is that implementing what you suggested in your first post would probably be more difficult than you made it out to be, as although software may be described as 'portable', it still requires a lot of work to port (even for Apple).

Amorph · Mar 16, 2003, 01:51 PM

Originally posted by rmendis:
I think people are mistaking this as taking a step back or away from Mac OS X.

Not at all.
In fact, it's the other way round.

It's really developing better Classic support into Mac OS X.

Define "better". Right now, Classic is done the way it's done because the #1 reason for its existence is compatibility. There are some crucial things that it can't do (like run Mac OS device drivers) that your solution doesn't fix — the reason your HP doesn't work is because the driver model changed radically, and Apple didn't quite get it finished for a while (10.2.4 rolled in what Epson needed to be able to do borderless printing, for example). The only way to use OS 9 drivers is to boot OS 9. Not OS 9 ported to Mach, or Darwin, since those enforce preemptive multitasking and protected memory by default, and thus require a completely different driver model. Porting OS 9 to run solo on top of Darwin accomplishes nothing except to make it more difficult to launch Classic. All the problems Classic currently has re: compatibility will still be there.

It's true that Mach was designed to run under various OS', but it's also true that OS 9 was designed to run right on the hardware. The microkernel that Apple introduced with 8.5 and refined with 8.6 and then 9 broke a number of applications each time, and they were specially designed to deal with OS 9's particular (and antiquated) memory and tasking models. Mach is not. Darwin is not. Mac OS is very much on its own. We already have the most seamless and most compatible port that you're going to see.

Amorph · Mar 16, 2003, 02:06 PM

[double post]

Terri · Mar 16, 2003, 03:29 PM

I loved System 9, but Apple took it as far as they could. It's done, stick a fork in it.

Classic under Mac OS X is very good. Quark actually seems to run better in Classic under X then it does booted into 9 and it doesn't take the machine down when it crashes.

Lots of places have scanners and some other stuff that doesn't work under X and that is perfect for using under 9 with older Mac hardware. Older peripherals older Macs, makes sense.

I have an old G3 that runs my old scanner and old Epson.

The only real problem I see with X is things from System 9 that haven't been brought to X yet. So I guess what I am saying is that we don't need to make Classic for like X we need to make X more like Classic or at least have the options to make it work more like Classic.

This thread was about making Classic more like X, but if X was more like Classic then we wouldn't have people asking to make Classic more like X.

At least that is how I see it.

exca1ibur · Mar 16, 2003, 08:45 PM

From what I gathered Apple designed Classic compatibilty to be nothing more than a bridge. To run exisiting apps untill they were updated to OSX natively... Thus the coffin at WWDC about killing off OS9 development. Once the major apps are in full swing on OSX, classic will be dropped. I see no reason to have this much work put into classic, as it is designed basically as a temporary bridge. The long term goal is to migrate users from 9 to X, not build a full environment in classic so we have a division in OSes for the years ahead. We have developers draggin their feet, that is causign this gap in some areas (Quark). OSX has been out long enough for them to get it together. A perfect example is Adobe...

Phtoshop 6.0 was out when OSX shipped and it ran in Classic. When 7.0 came out they used Carbon to port it over to OSX, now the next version is said to be OSX only using many native functions of OSX technolgy. This is the idea scenario of migration from Classic to OSX. The foundation is there in classic to do what it was designed for. Nothing more. We want to move forward not stay in the present state.

my 2 cents...

Millennium · Mar 16, 2003, 09:09 PM

Oh, Christ...

This is not possible, people. Learn to deal with that.

1. Rebuild Mac OS 9 ontop of Darwin - this is what Mach microkernel architecture was designed for - to be very versatile and allow the budding of various OS 'shells' - MkLinux, NEXTSTEP (BSD), OSF/1, etc...

Um, not quite. The core architecture of OS9 was irreconcilably different from that of OSX. Perhaps the biggest difference is in drivers; they're done in completely different ways, and it just plain Is Not Possible to port from one to the other in any easy manner. Without drivers, what's the point?

Another thing: there are many aspects of the original Mac OS which were programmed using practices that most half-decent programmers wouldn't touch with a hundred-foot pole nowadays. I'm not talking about cooperative multitasking or anything like that: I'm talking about things like the fairly heavy use of global variables in QuickDraw. For you non-programmer types, this basically means that to do certain things in QuickDraw, you had to overwrite a very specific point in the system's memory, but anything else could overwrite it at any time and you had no real way of tracking whether or not that had been done, or of preventing other apps from doing it. This is such a nasty hack that it's not even possible to do the same way in Mach or in Linux; memory protection gets in the way. That alone keeps you from just porting it over.

in fact, there is nothing stopping Apple from gently migrating the Mac OS X unix shell to Linux (ala MkLinux): "Linux for Mac OS X".

Actually, there is. MkLinux, you see, is not really Linux as we know it. It's a very clever Linux-alike, but many very imporrtant parts of the guts were ripped out and replaced with Mach's versions. To migrate the OSX Unix shell to MkLinux would be basically making three layers instead of two: BSD over Linux over Mach, instead of just BSD over Mach.

What's the fascination with Linux, anyway? Not to disparage Linux at all; it's a great OS in its own right and I use it often, but there's no real reason for Apple to port.

2. Back port Quartz to Mac OS 9. The Quartz rendering + compositing engine is again a very portable system.

Says who? Ever tried it? Quartz relies very heavily on things like preemptive multitasking, which is not possible to backport to OS9. They tried it several times, and it couldn't be done.

3. Give Mac OS 9 the Aqua theme along with transparency and compositing.

Impossible to do well without Quartz. See above for why that can't happen.

Then Carbon and Classic apps being seemless as they are already in Mac OS 9, will inherit the Aqua look & feel.

Theoretically. But, again, since this cannot be done, this point becomes moot.

4. Port the Mac OS X Carbon Finder to Mac OS 9.

Why would you want to do that? OS9 has a perfectly good Finder of its own. Some would say it's even better, in fact, and while I wouldn't agree with that outright, I will admit that good arguments can be made.

5. Now i'm not sure if Cocoa will be portable to Mac OS 9...even though Carbon is and they share the same Core foundation. Say it is, then with a little work, Cocoa should port to Mac OS 9...it will at least require retrofitting NSEvents.

It is true that Cocoa is being reimplemented on top of Carbon. The preliminary work is already present in Jaguar. But it will be a long time -perhaps years- before that is complete. As part of this, NSEvents is being reimplemented, presumably on top of Carbon Events.

Cocoa is actually pretty portable. There's a version of it for Windows, though the only way to get it nowadays is to buy WebObjects. In addition, the GNUstep project has a Cocoa-alike running on Linux, though with a few modifications. For example, they don't use .nib files, preferring their own .gmodel format. Why they do that I don't know, given that they obviously know how .nib files work (they have a convertor program).

Viola!

A viola is a large stringed musical instrument, generally played with a bow. The french word for "there", which is what I think you were trying to use, is "voil�".

You will basically have Classic, Carbon and Cocoa seemlessly integrated as one.

You make it sound so simple on paper. Unfortunately, there is this little thing called, I don't know, reality which pretty much tears your plan to ribbons. This is basically a "Bring Back OS9" plan, isn't it?

Backward-compatibility is all well and good, unless it starts to hamper actual innovation and advancement. This has chained the Intel architecture and Win32 API for years now. Eventually, the bandage has to be ripped off, and while that does hurt for a short while, it's the only way for the injury to fully heal.

exca1ibur · Mar 16, 2003, 11:51 PM

Backward-compatibility is all well and good, unless it starts to hamper actual innovation and advancement. This has chained the Intel architecture and Win32 API for years now. Eventually, the bandage has to be ripped off, and while that does hurt for a short while, it's the only way for the injury to fully heal.

NICE!!

You need to get a patent for that. Thats the best saying I've heard for backwards compatibility. Good one.

rmendis · Mar 17, 2003, 02:08 PM

Originally posted by Millennium:
Quartz relies very heavily on things like preemptive multitasking, which is not possible to backport to OS9. They tried it several times, and it couldn't be done.

For Cocoa to be ported to run on Mac OS 9/Classic, it would only require parts of Quartz - namely Quartz rendering. It won't require the Quartz window server or compositor.

Now the Quartz render should be very portable.

It should be similar the Adobe's renderer for the SVG plugin (for IE5 and Netscape) on the Mac: They do use the same display model and format (PDF). Also there may be strong similarities to Adobe's Bravo engine for Java 2D. It wouldn't surprise me if the SVG plugin and Quartz rendering is based on that...though that is only a wild guess

Anyway given that the SVG plugin does work on Mac OS 9/Classic, i would think that the Quartz renderer should do as well. If not, the possibility of retrofitting it to do so exits.

And given that the Core* stuff has been factored out of both Carbon and Cocoa, i would expect that Cocoa is more portable than before...particularly to a platform on which Carbon has already been ported to - namely Mac OS 9/Classic.

cwasko · Mar 17, 2003, 02:45 PM

Originally posted by rmendis:
[Spun off discussion from the "What's new in Mac OS X 10.3 Panther?" thread]

...

Consider this:
1. Rebuild Mac OS 9 ontop of Darwin - this is what Mach microkernel architecture was designed for - to be very versatile and allow the budding of various OS 'shells' - MkLinux, NEXTSTEP (BSD), OSF/1, etc...in fact, there is nothing stopping Apple from gently migrating the Mac OS X unix shell to Linux (ala MkLinux): "Linux for Mac OS X".
2. Back port Quartz to Mac OS 9. The Quartz rendering + compositing engine is again a very portable system.
3. Give Mac OS 9 the Aqua theme along with transparency and compositing. Then Carbon and Classic apps being seemless as they are already in Mac OS 9, will inherit the Aqua look & feel.
4. Port the Mac OS X Carbon Finder to Mac OS 9.
5. Now i'm not sure if Cocoa will be portable to Mac OS 9...even though Carbon is and they share the same Core foundation. Say it is, then with a little work, Cocoa should port to Mac OS 9...it will at least require retrofitting NSEvents.

Viola!

You will basically have Classic, Carbon and Cocoa seemlessly integrated as one.

...

My god man! How is this a good idea? Isn't the simplest solution to spend time optimizing OSX *AND* getting the lazy-ass developers to support OSX?

I just don't see any benefit to trying to support an old, dated system when it has *no future*. Sure, there are things that run better in it... then why not just *run it*. I have a 7100/80 av with a 440MHz G3 in it if I ever want to run any old tech crap.

dfiler · Mar 17, 2003, 04:11 PM

Originally posted by Millennium:
Oh, Christ...

This is not possible, people. Learn to deal with that.
...

You make it sound so simple on paper. Unfortunately, there is this little thing called, I don't know, reality which pretty much tears your plan to ribbons.

Thank you for another voice of reason!

No matter what these armchair operating system commentators claim, reworking an entire OS architecture is not quick or simple.

Once again, let me point out that nearly everything detailed above is already accomplished via Carbon or was attempted during the multi-hundred-million dollar boon-doggle known as Copland. How many more billion should Apple waste just to prove that porting Cocoa, Aqua, and OS X drivers to OS 9 is an incredible waste of money?

Friggen ridiculous

Amorph · Mar 17, 2003, 04:20 PM

Originally posted by JLL:
You do know that AS2.0 is released in beta form?

Actually, I didn't. I knew it was in process, but that's it.

Time to go digging, I suppose. Thanks.

Amorph · Mar 17, 2003, 04:39 PM

Originally posted by rmendis:
For Cocoa to be ported to run on Mac OS 9/Classic, it would only require parts of Quartz - namely Quartz rendering. It won't require the Quartz window server or compositor.

So you're asking Apple to rework the "chain of responsibility" provided by the Quartz server into the flat event model used by MacOS, all transparently within the Cocoa frameworks? Not gonna happen.

And given that the Core* stuff has been factored out of both Carbon and Cocoa, i would expect that Cocoa is more portable than before...particularly to a platform on which Carbon has already been ported to - namely Mac OS 9/Classic.

Not all of Carbon has not been ported to Classic. It's eminently possible to write an OS X-only Carbon app, and it's even sensible to if you're concerned about ease of development and performance.

There is no way to accomplish what you want. OS 9's architecture is too fundamentally different to run on Mach, and its drivers make assumptions about the way the machine is run that cannot be gotten around with a simple KEXT (or surely Apple would have done so?).

OS 9 is dead. Classic is the best Apple can do without crippling their future in the name of legacy support for an OS that should have been overhauled ten years ago.

JLL · Mar 17, 2003, 04:39 PM

Originally posted by Amorph:
Actually, I didn't. I knew it was in process, but that's it.

Time to go digging, I suppose. Thanks.

Don't get your hopes up - I just realised that it was Script Editor 2.0b that I was thinking about - D'oh!

Terri · Mar 17, 2003, 04:55 PM

Classic may be dead but this thread just will not die.