Jean-Louis Gassée (ex-Apple, ex-Be): Apple’s Next Macintosh OS
mondaynote.com
mondaynote.com
I think all of Apple's R&D efforts are now focused squarely on iOS and mobile. So when Apple reached into their bag of tricks for the next version of OS X and the next Mac hardware, all they had was iOS stuff. That's the new MacBook Air, that's OS X Lion.
It's convergence, but only on a purely superficial level. A Mac is still Mac, OS X is still OS X. I don't see Apple wasting engineering resources on changing the Mac into something it's not.
Real convergence will come from iOS working its way up, not OS X slimming its way down.
There are plenty of apps that exist and run right now (Minecraft, for example) that can't go into the app store. So you can think of the app store as sort of a platform "ideal" for Apple even while they leave the old APIs around.
I do welcome the UI possibilities. UI innovation was languishing for a decade, as companies essentially spent their time tweaking the 1984 Mac and 1993 Mozaic UIs. And then Apple changed all that. Good for them. Bring some of that to the Mac; I'm all for it.
On the other hand, the trend toward the computer being something that some company controls, and I am allowed to use as they see fit, is not something I want to be a part of.
I recently bought a home computer. I run Linux on my office desktops & laptop, but, after much reflection, I decided I could not inflict even the relative friendliness of Ubuntu on my family. So we got an iMac. Great machine. We love it.
But now it's looking like, in a few years, there might be no computer at all that my family would enjoy using, and that I would allow in the house. Not a pleasant thought.
a) there's a lot of supposition in the whole "Apple is going to lock the Mac down" line of thought.
Personally I think it's unlikely they will simply because I think it would be a bad decision commercially. I can see an OS that is more restricted by default but software developers, designers and corporates simply won't accept heavy restrictions and will look elsewhere if the changes impact them.
Regardless though, it seems odd to me how worked up people are getting about something which is, at best, a prediction about something which might happen two or three years (at the earliest) down the line.
b) I love the idea that you might not allow a more locked down Mac into the house despite the implication that your family would enjoy and benefit from it.
Surely you judge each device on it's merits? Personally I'll accept the restrictions of the iPad because I have no need for it to be more open than it is but I won't accept those same restrictions on a computer.
You shouldn't impose your needs for a compute on your family and more than it would be reasonable for a professional chef to insist that their son or daughter had a professional standard cooker or knives when they're completely excessive for what they do?
True. Actually, I doubt they will lock it down.
However: Apple's online stores, beginning with iTunes, and heading into the various app stores, have been a huge success, and, from a business standpoint, they would be foolish not to do something almost identical for the Mac. (I know they're going to do it; it's the "almost identical" part that I'm pointing out.)
And then, of course, from developers' point of view, Apple has created something wonderful. It is now possible to sell a file for $1/copy, never deal directly with end users, and make decent money at it. So given a Mac app store, developers signing on is obvious.
The result is that, even if the OS itself is not locked down (which, as I said, I consider unlikely), we end up with a situation in which getting quality software requires giving Apple control of your machine.
That is what I consider unacceptable. Quality software is why I bought a Mac. Linux's opennness, scriptability, etc., is great, but there is nothing on Linux with anywhere near the polish and usability of iMovie, GarageBand, etc. Nor will there be for some time, I think. And when that kind of thing moves over to the we-run-your-computer world, then I don't see any options left for a home computer I'd be willing to buy.
It's an install and download mechanism which you can use or not use.
But I doubt it will be. For example, Apple's current app store is not merely an installation mechanism; it also allows them, to disable apps they don't want running on your machine. And after 30 years of flailing around, Apple has finally found a business model that works for them. I expect them, therefore, to copy as much of it as they can on the Mac.
The reason they don't completely screw people's machine isn't technical, it's commercial, it would be utter suicide. That applies to Apple as much as anyone else and the app store doesn't change that (the kill switch for iOS has never been used in the three years it's been there).
The difference is similar to that between a plumber I call in now & then to fix some pipes, and a plumber who gets a key to my house, and my written permission to do anything he wants, whenever he wants. Both of them have the capability to rearrange my plumbing in any way they want. But the latter is scarier.
I do think the good software is going to start being app-store-only, and that will require giving Apple full access to my machine. Good software is why I got a Mac. Not being able to get good software without giving Apple the ability to delete files of mine that they don't like, is why I might not get another.
We'll see.
Bigger more established developers (like Panic, Bare Bones Software, Delicious Monster, The Iconfactory, The Omni Group, just to name a few) have already solved these problems (not in perpetuity and not without limit, of course).
I don't think that companies like these are going to place the fate of their best-selling software in Apple's hands for the relatively small gains the Mac App Store will provide them. That is not to say that I think they'll stay out of the Mac App Store either, it's just to say that it won't be Mac App Store exclusive.
I expect most of them to "try out" the Mac App Store with new software releases of smaller scope, rather than retrofit their best-selling software and offer them exclusively on Mac App Store, with the straightjacket that comes with it.
Where they go from there will depend on the results of the Mac App Store. They won't be forced to use it for their key software, so they won't if it doesn't suit their needs.
I think the Mac App Store is going to be an absolutely enormous runaway success. But not because you can find already established and popular software on there.
Let's hope so.
> I think the Mac App Store is going to be an absolutely enormous runaway success.
Can't disagree with that.
"A few days ago, I downloaded a neat little utility to silence the startup sound on my new 11” MacBook Air. How much did the developer make? Zero, it’s freeware; the programmer didn’t want to spend the time and money to set up a commercial site. How much would I have paid for it from a Mac App Store? Less than $5, more than 99 cents."
It seems to me that silencing the startup sound is essentially tampering with the Apple-designed user experience, which may not be looked kindly upon. (Also, it seems like the kind of thing that may require private API access, which is explicitly forbidden on the Mac App Store.)
"Apps must be self-contained, single application installation bundles, and cannot install code or resources in shared locations."
I don't see how the utility in question could be a single app bundle (no daemon, no startup script, etc.) and still perform its job of muting the volume on startup.
Apple made Rosetta to run older apps and Windows 7 includes Windows XP in a virtual machine in business SKUs.
Also, Intel chips haven't really implemented x86 instructions for a very long time. Most of the chip runs micro-instructions that are derived for a x86 to micro-instruction translation layer.
Universal Intel & PowerPC binaries
Carbon & "Classic mode" on OS X
Fat Binaries (68k & PowerPC)
I think if Microsoft mostly targeted consumers, rather than businesses, they would take the same approach to compatibility as Apple. But a large amount of Windows sales results from selling to corporations that have hundreds of critical legacy applications. They simply won't upgrade if their apps won't run. However, on my desktop, I doubt I'm running a single application that's more than 3 years old.
Whereas a "Classic mode", "WINE", or "Virtual Machine" style mode would seem to solve the backwards compatibility thing in a much more containable way - apps still run, the old APIs stay frozen in the custom environment but deprecated/removed outside, new developments can use new APIs without being saddled with the old.
Even though the compatibility environments have to be kept around forever, now you've only got a few API interfaces to renovate (connecting the classic environments to whatever new Windows APIs exist), instead of thousands.
I guess it's not as simple as that, though.
Microsoft shifted away from backwards compatibility with Windows XP Professional x64 (ok actually the Itanium version). With Vista, 16bit code was broken (ie some Win9x and WinMe).
Microsoft's current approach to backward compatibility is radically different from Apple's. Microsoft allows Windows to run in virtual machines. If you want to run Windows 95, just fire up your favorite virtual machine (Virtual PC, VMware, etc), create a machine, and load Win95 from your CD.
Microsoft released the first version of Virtual PC several years ago at no cost. It runs under all versions of XP, Vista, and Seven. Windows Seven Professional includes a built in copy along with XP so that an existing licensed copy of XP isn't required.
This is the thing, though: it took until 2007 to break support for code written as far back as 1993.
Microsoft allows Windows to run in virtual machines
To me, this definitely seems like a smart strategy.
Allowing virtualization is very much a customer focused strategy, but one which Apple deliberately avoids at the consumer level with OSX. It doesn't help them in their core business of selling branded hardware with a three year support lifecycle.
What hasn't been explored is virtualization as a high level organizational feature of a desktop operating system. I currently reference a legacy application which relies on 16bit code. Running it in a virtual machine is fabulous because I can save the state and come back to the exactly the same point two months later.
I have another VM which runs Ubuntu. I used it to sign up for Facebook and that's all I use it for. It appears to sidestep the commercial byproducts of Facebook use.
It's AMD/Intel that has broken 16bit compatibility with 64bit, not Microsoft. If you put your CPU in 64bit mode, it will no longer be able to run 16bit code.
Chicken || Egg;
> the backwards-compatibility burden sometimes running all
> the way back to Windows 95, maybe even Windows 3.1+Win32s.