[CharlesS]

Originally Posted by OreoCookie 

You're right it was a mistake to tell the world they'll make a 64-bit version of Carbon, Adobe thought they could get away with not adopting standard technology once again.

Apple has been saying from day 1 that Carbon is a stop-gap solution -- all we did not know was when they would actually pull the plug.

We've had this discussion before. I was there, and the way I remember it, Jobs walked out on stage and announced that Carbon was "the basis of all life." I remember Cocoa and Carbon being billed as equal partners. I remember a whole bunch of the Foundation APIs getting backported to the C-based API, but not the reverse. I remember Carbon getting the majority of attention, due to the big companies using it, and I remember a lot of really basic stuff like drag-and-drop file promises missing from Cocoa and not getting fixed all the way until 10.2. I remember flame wars on this board between zealots insisting Cocoa was better without really understanding what it was (kind of like 64-bit now) and other folks claiming neither was better, they were just different. And, I remember a lot of panic on cocoa-dev (like this or this) about whether it was just a matter of time before Apple axed Cocoa in favor of Carbon due to the large clients using the latter, and whether Carbon was a safer way to go, due to Apple's history of killing off APIs perceived as superior (QuickDraw GX, OpenDoc, etc.).

It's not nearly as cut and dried as you present it.

[Chuckit]

Originally Posted by OreoCookie 

Apple has been saying from day 1 that Carbon is a stop-gap solution

I don't think so. Feel free to go through the old Carbon docs and find where it says that — I don't think it's in there. What Apple did say is that Carbon was a first-class application framework for OS X, and they were going to rewrite the Finder as a Carbon app to prove it. In fact, Apple was saying right up until the last minute that all of Carbon would make the leap to 64-bit — and even now, only some of the framework is being killed off. You're making this out like Apple was hounding everyone to switch to Cocoa and making it so obvious that it was the right choice, but that was never how it seemed to me — actually, Carbon was a vastly more powerful framework until quite recently. Much like cutting Rosetta would be, it was just Apple jerking people around. (OK, even more so, actually, since Rosetta actually is obviously a stopgap rather than a first-class API.)

Originally Posted by CharlesS 

I remember a lot of really basic stuff like drag-and-drop file promises missing from Cocoa and not getting fixed all the way until 10.2.

Worse than that. Cocoa only just recently got support for alias functionality — in Snow Leopard.

[OreoCookie]

Originally Posted by Chuckit 

I don't think so. Feel free to go through the old Carbon docs and find where it says that — I don't think it's in there.

I'm not sure whether it's in the documents, but this has been mentioned in many keynotes as early as 2000 (forward to 4:50). He speaks of a gentle migration from left to right (Classic-Carbon-Cocoa) and says new apps should be written in Cocoa. Also articles on OS X at that time clearly state that Carbon is a transitional API. So perhaps it's not in the documentation (I wouldn't know), but it has clearly been communicated that way -- and people who listened understood this message.

Of course, a lot has happened between the year 2000 and 2009, but this was clearly the initial plan. And yes, Carbon was extended quite a bit, because a lot of developers were (and are) still using it. However, it seems that Apple was spreading the news already in 2007 at the very least.

Originally Posted by Chuckit 

What Apple did say is that Carbon was a first-class application framework for OS X, and they were going to rewrite the Finder as a Carbon app to prove it.

No, Apple has used Carbon in the Finder (and also some other apps) as a testbed for Carbon. The Cocoa equivalent was Mail. (You can trust me on this.)
The initial plan was to kill off Carbon much sooner and I'm fairly certain Apple would have if it weren't for two major software vendors (one sits in Redmond, the other makes a popular image manipulation software).

Originally Posted by Chuckit 

In fact, Apple was saying right up until the last minute that all of Carbon would make the leap to 64-bit — and even now, only some of the framework is being killed off.

I absolutely agree, that's a mistake on Apple's part to give a clear plan to developers what the future of the different APIs will look like. (I guess Apple simply changed its mind at one point, deciding it would be a good time to get rid of Carbon.)

[Spheric Harlot]

Originally Posted by P 

You're right, mostly, but I don't see how stopping Carbon support helps. A new Fireworks interface is not better than the old Carbon one - quite the opposite - and they're not going to make a Cocoa interface on a cross-platform app. For the record: If Apple had said from day 1 of the 64-bit transition that Carbon would not be supported, I would not have said a thing about it and Adobe would have 64-bit Cocoa PS on OS X today.

Possibly.

But without the tremendous growth in user base Apple has seen in very recent years, there also would have been a considerably greater risk of Adobe just pulling the plug altogether.

Photoshop works just fine in Parallels and/or Boot Camp + small user base + disgruntledness with Apple's communication politics + Adobe having developed into a bunch of complacent, ****witted bastards = Why the hell develop for Mac OS X?

I *highly* doubt we would have a 64-bit Cocoa Photoshop today, if at all, if Apple had been more forthright.


And please, "killing off" is so barbaric. "Deprecating" is much more…subtle.

[Spheric Harlot]

Originally Posted by P 

I doubt it. Apple is many things, but it's not clueless. I think they want to advertise that certain apps are old, so users understand why their computer is suddenly less snappy.

That too, but I'm sure it's an honor for a developer to be associated with an application that forces a tedious system download/installation and runs comparatively unsnappy.

Might that not be an incentive towards developers?

Originally Posted by P 

Sure, but that's a different question though. That iTablet would have a significantly different interface. Current x86 Mac apps would only ever run badly on it. I can certainly see them skipping Rosetta on that.

Recent indications seem to point towards iPhone-like hardware (not talking necessarily about the internals) running full Mac OS X, and I'm pretty sure that this is the only thing that really makes sense - but we'll see.

Every one of the rumors I've read so far seems to lack the meat - i.e. the actual thing that makes the product *desirable*.

An underfeatured, overpriced iPod touch deluxe is as useless a product as an affordable, underpowered MacBook touch.

Unless Apple pulls another one of their OMG-so-blindingly-obvious-in-hindsight rabbits out of the hat...

We'll see.

Since they made Exchange support THE major SL feature. And they should care because they can't grow any more in their current market segments.

I think there's a huge difference between "The Corporate Market" and travelling businessmen or employee working from home needing to access the corporate database.

The former is the price-driven, Dell-dominated grey desktop box-in-every-office market.

The latter is a market that Apple is quickly gaining a good foothold in, and real Exchange support (as opposed to the uselessly incomplete support built into Office 2008) cements that by eliminating the need for virtualization in a good many cases.

And that's fine, and their way of supporting older apps probably works fine with them, but indications are they are looking to grow outside that niche. Killing Rosetta will hurt them in that market.

You make a good point about past politics making a good indicator for future decisions, but I think that this is not too relevant in the business market they're actually looking at, because

a) it's not the set-up-a-generic-image-and-deploy-across-4,000-generic-boxes-representing-a-2.4-million-dollar-investment-market, and

b) Apple's engagement in this market doesn't come with a pre-Intel legacy they need to support.

[- - e r i k - -]

Originally Posted by OreoCookie 

The initial plan was to kill off Carbon much sooner and I'm fairly certain Apple would have if it weren't for two major software vendors (one sits in Redmond, the other makes a popular image manipulation software).

I think you are also forgetting a third major software vendor in Cupertino that makes a popular video editing suite…

[mackandproud]

I didn't install Rosetta at first, since I didn't think I'd need it. Turns out there is still software out there which still requires it, and I installed it since it was necessary after the SL initial install. This is likely to be the case for a lot of people: we will skip it in the initial install, only to discover some older, legacy programs still need it.

[OreoCookie]

Originally Posted by - - e r i k - - 

I think you are also forgetting a third major software vendor in Cupertino that makes a popular video editing suite…

No, I haven't forgotten.

[P]

Originally Posted by Spheric Harlot 

That too, but I'm sure it's an honor for a developer to be associated with an application that forces a tedious system download/installation and runs comparatively unsnappy.

Might that not be an incentive towards developers?

It's just once, though. If it did that warning the first time every PPC app ran, then yes, but it does this once in forever.

Originally Posted by Spheric Harlot 

Recent indications seem to point towards iPhone-like hardware (not talking necessarily about the internals) running full Mac OS X, and I'm pretty sure that this is the only thing that really makes sense - but we'll see.

Maybe. I have absolutely no idea what a potential iTablet will be. The best idea I ever had was a Kindle DX sort of thing - big iPod touch - bad that seems so...mundane.

Originally Posted by Spheric Harlot 

The former is the price-driven, Dell-dominated grey desktop box-in-every-office market.

It's focused on TCO. If MS had released another Vista-sized disaster, Apple would have got a look - and they DID get a look after Vista, at least in some areas.

And it's not desktops anymore: Practically everything is a laptop, which helps Apple - as long as they keep one laptop at a serious price level.

Originally Posted by Spheric Harlot 

You make a good point about past politics making a good indicator for future decisions, but I think that this is not too relevant in the business market they're actually looking at, because

a) it's not the set-up-a-generic-image-and-deploy-across-4,000-generic-boxes-representing-a-2.4-million-dollar-investment-market, and

b) Apple's engagement in this market doesn't come with a pre-Intel legacy they need to support.

My point is that corporate IT loves the way Microsoft clearly communicates support windows. Apple doesn't, and is in fact secretive, and that is a problem for anyone wanting to push an Apple solution. Previously, you could argue that Apple at least had a good past record, with apps written for 68k Macs still running in Classic on 10.4, and you could argue that that made the support window thing less relevant from an application perspective. Now Classic is gone, Rosetta clearly on the way out, and that argument is hard to make.

[Spheric Harlot]

Classic died in 10.5, what, five years after they made OS X the default boot?

If Rosetta should go away in 10.7, maybe 2011?, it will have been five years since the last PPC machines.

Nothing changed here in terms of track record.

The fact that (some) apps written in 1985 still worked in classic in 2007 is more a testament to how long in the tooth the Classic OS was...

[P]

Originally Posted by OreoCookie 

That's correct. But in a professional environment, the necessity to upgrade is a fact of life. Needing to upgrade after six, seven, eight years is not forcing the hand of many customers.

Maybe I'm biased from personal experience, but I have to work every day with one such installation, where an ancient OS has to kept alive because of a certain piece of software doesn't work on newer versions. That software is dead, replacement is very expensive and just won't happen. That incompability is probably for a good reason, but eliminating compability layers as some sort of good housekeeping maneuver, always drives me crazy. If some starts mentioning "to drive lazy developers to update already", well I'm sorry, but you have no idea what you're talking about.

Originally Posted by OreoCookie 

I've seen other numbers based on browser usage. I couldn't find them, but the pattern was rather similar: OS X users were adopting new versions of OS X much quicker than Windows users would adopt new versions of Windows.

Yes, but 10.1 was indisputably better than 10.0, 10.2 than 10.1, 10.3 than 10.2 etc. Vista is not indisputably better than XP, XP is not indisputably better than 2000, etc. Vista for all its flaws has moved forward in some areas, but has major problems in others. XP is a very minor update to 2000, really (even if SP2 added some more features) and a lot of things change if you're coming from Win 98/Me. It's not obviously better at everything the way the first OS X versions were.

[Simon]

Originally Posted by P 

And it's not desktops anymore: Practically everything is a laptop, which helps Apple - as long as they keep one laptop at a serious price level.

(emphasis mine)

I disagree with that statement. I think Spheric was correct when he said there remain basically two corporate desktop markets:

Originally Posted by Spheric Harlot

I think there's a huge difference between "The Corporate Market" and travelling businessmen or employee working from home needing to access the corporate database.

The former is the price-driven, Dell-dominated grey desktop box-in-every-office market.

The latter is a market that Apple is quickly gaining a good foothold in, and real Exchange support (as opposed to the uselessly incomplete support built into Office 2008) cements that by eliminating the need for virtualization in a good many cases.

It is absolutely true that corporate systems that run >$1000 and are used by medium to upper level employees are migrating to notebooks very quickly. You'll have a tough time finding a manager or junior VP who doesn't want a notebook (even if they hook it up to a monitor, KB, and mouse at the desk). That's essentially what Spheric referred to as the "travelling businessmen or employee working from home" market.

But that is by far not all of the corporate market (and by corporate I mean "desktop", not server). There are still countless simple desktop systems installed in every company. They're used by lower level employees, secretaries, call center people, customer reps, etc. Essentially what Spheric called the "Dell-dominated grey desktop box-in-every-office" market. And this huge chunk of the market is by no means migrating from $300 Dell desktops to $800+ notebooks. This is the one market (beside gamers maybe) where the dull gray box will stick around for many years to come.

Now I fully agree with you that it makes sense for Apple to focus on notebooks. It appears that's what they like to do and what they do best. But, if your premise to Apple entering the corporate market is that they can do so by only selling notebooks, you have to be aware that automatically means you'll only be talking about a specific subset of the actual corporate market. I too believe that Apple doesn't stand to gain much from trying to win market share in the $300 gray box segment, but that doesn't mean the segment doesn't exist or isn't important in order to become visible in the corporate market.

Apple should chose to offer only notebooks to the corporate market if they come to the conclusion that that's the only area where they can make money. But if they're serious about gaining substantial market share* and/or becoming an important player in the corporate market, Apple will have to reconsider a notebook-only approach.


*) Note that since Apple owns most of the high-end home and consumer market, the corporate market is the one area they should be looking at if they plan to further increase their overall market share with computers. Of course, I'm tacitly assuming Apple wants to make more money not just by selling phones and gadgets, but also by selling more Macs.

[P]

Originally Posted by Spheric Harlot 

Classic died in 10.5, what, five years after they made OS X the default boot?

If Rosetta should go away in 10.7, maybe 2011?, it will have been five years since the last PPC machines.

Nothing changed here in terms of track record.

The fact that (some) apps written in 1985 still worked in classic in 2007 is more a testament to how long in the tooth the Classic OS was...

Five and a half, yes. But there was a good reason for it: Apple didn't want to port Classic to Intel machines, so it was dying anyway. So far, noone has produced a good technical reason for removing Rosetta.

[P]

Originally Posted by Simon 

(emphasis mine)

But that is by far not all of the corporate market (and by corporate I mean "desktop", not server). There are still countless simple desktop systems installed in every company. They're used by lower level employees, secretaries, call center people, customer reps, etc.

3 years ago, I would have agreed. Today, I don't. I just walked around the office: There are a few desktops in finance and payroll. There is one or two standing fixed on desks, acting as dumb terminals. There are some UNIX workstations. Everything else around here is laptops - at least 90% if you exclude the workstations. My guess is that the price difference has dropped enough that we buy laptops for most staff.

Originally Posted by Simon 

Essentially what Spheric called the "Dell-dominated grey desktop box-in-every-office" market. And this huge chunk of the market is by no means migrating from $300 Dell desktops to $800+ notebooks. This is the one market (beside gamers maybe) where the dull gray box will stick around for many years to come.

Apple makes the mini, which would work well here. The price could come down, though.

Originally Posted by Simon 

Now I fully agree with you that it makes sense for Apple to focus on notebooks. It appears that's what they like to do and what they do best. But, if your premise to Apple entering the corporate market is that they can do so by only selling notebooks, you have to be aware that automatically means you'll only be talking about a specific subset of the actual corporate market. I too believe that Apple doesn't stand to gain much from trying to win market share in the $300 gray box segment, but that doesn't mean the segment doesn't exist or isn't important in order to become visible in the corporate market.

Apple should chose to offer only notebooks to the corporate market if they come to the conclusion that that's the only area where they can make money. But if they're serious about gaining substantial market share* and/or becoming an important player in the corporate market, Apple will have to reconsider a notebook-only approach.

I'm not arguing anything of the sort. I'm only saying that what I see around me is the office moving to laptops only (the desktops I saw were older P4s with one exception) and that this trend helps Apple. It was very much a throwaway comment, though.

[OreoCookie]

Originally Posted by P 

Maybe I'm biased from personal experience, but I have to work every day with one such installation, where an ancient OS has to kept alive because of a certain piece of software doesn't work on newer versions. That software is dead, replacement is very expensive and just won't happen.

Sure, I'm aware of cases like this. For instance, a friend of mine works with a über-high tech electron microscope. He can monitor and write chip structures that are smaller than anything Intel, AMD and the likes can deliver. However, the software does not run on versions of Windows based on the NT kernel -- which means they are stuck with Windows 98. They had to buy used pc hardware for a piece of equipment that costs $500k+!

Originally Posted by P 

That incompability is probably for a good reason, but eliminating compability layers as some sort of good housekeeping maneuver, always drives me crazy. If some starts mentioning "to drive lazy developers to update already", well I'm sorry, but you have no idea what you're talking about.

Huh?
I'm just saying this is a problem for a small minority. Again, I'm not talking about Rosetta (because we don't disagree about Rosetta at all), I'm focussing on Carbon. Carbon works, it's just no longer developed actively and people are and have been advised to use Cocoa instead. I don't think Apple will take out Carbon within the next two releases. So if you would like to run Office 2008 four years from now, I don't see any problems. Same goes for Adobe apps. I assume both are `must not break apps' for Apple.

The problem with Carbon is that Adobe wants to make a 64 bit version of, say, Photoshop. This feature is simply not available -- and this is not the same as killing backwards compatibility.

Personally, upgrading the OS broke only one app I needed to pay for: when I went from Tiger to Leopard, I could no longer use Photoshop 7 which I have gotten for free from a friend who has bought a license to a new version of the Creative Suite.

Originally Posted by P 

Yes, but 10.1 was indisputably better than 10.0, 10.2 than 10.1, 10.3 than 10.2 etc. Vista is not indisputably better than XP, XP is not indisputably better than 2000, etc. Vista for all its flaws has moved forward in some areas, but has major problems in others. XP is a very minor update to 2000, really (even if SP2 added some more features) and a lot of things change if you're coming from Win 98/Me. It's not obviously better at everything the way the first OS X versions were.

Well, but the pattern is consistent and continues with 10.6 (if you look at the sales figures).

[P]

Originally Posted by OreoCookie 

Huh?
I'm just saying this is a problem for a small minority. Again, I'm not talking about Rosetta (because we don't disagree about Rosetta at all), I'm focussing on Carbon.

Then we don't disagree on much. I understand that Apple didn't port Carbon to 64 bit: My only complaint with that is, as I said, that they promised to do it and then backed down on that promise. I also think that this removal of Rosetta from the default installation is an ominous sign that it will be removed and we're apparently both opposed to that.