MacOS X and XCode
Hey Folks --
Hear me out before you shoot my head off. I have several friends at Apple and have even interviewed for a position in the XCode group so I at least kind of know what I'm talking about. That having been said, this rant includes a lot of personal speculation.
Ask yourself, why would Apple want to control it's own destiny when it comes to developer tools.
The obvious answer is that by giving away easy to use, powerful developer tools for free you empower people to create applications. If you pay attention to Apple, however, you know that Apple rarely invests in something these days unless it is critically important to the company's future. Let me assure you Apple is investing heavily in XCode.
I really believe that the real reason Apple wants a strong set of in-house development tools is that they plan to release MacOS X on Intel. I know, I know -- the notion has been discredited before.... but having met many of the new engineers on the XCode project, I am no longer so sure. They aren't your typical 'Apple Engineers', lets just put it that way.
So you can now attack me but I felt compelled to share. I know, I know... the G5 is here. Why would Apple need to go Intel. Good question, I wonder the same thing.
Hear me out before you shoot my head off. I have several friends at Apple and have even interviewed for a position in the XCode group so I at least kind of know what I'm talking about. That having been said, this rant includes a lot of personal speculation.
Ask yourself, why would Apple want to control it's own destiny when it comes to developer tools.
The obvious answer is that by giving away easy to use, powerful developer tools for free you empower people to create applications. If you pay attention to Apple, however, you know that Apple rarely invests in something these days unless it is critically important to the company's future. Let me assure you Apple is investing heavily in XCode.
I really believe that the real reason Apple wants a strong set of in-house development tools is that they plan to release MacOS X on Intel. I know, I know -- the notion has been discredited before.... but having met many of the new engineers on the XCode project, I am no longer so sure. They aren't your typical 'Apple Engineers', lets just put it that way.
So you can now attack me but I felt compelled to share. I know, I know... the G5 is here. Why would Apple need to go Intel. Good question, I wonder the same thing.
Comments
Can you clarify?
Originally posted by Placebo
It took me a couple tries to wade through your post, but I still don't get the point you trying to make. Xcode isn't the first free devtool that Apple has released. Apple has always had developer tools available for free, and I don't see what you mean by "giving away ease of use".
Can you clarify?
Before, Apple was heavily reliant for real software development tools. When Apple made the jump to PowerPC, Metrowerks played a huge role in the transition, for example.
Now, Apple has taken the development tools that they had at NeXT and has invested *significant* engineering resources into them. Part of it is that obviously Apple needs those tools for their in-house development. Another part of it is that they want developers to have good tools for free.
But there is more to this... you don't assign/hire the caliber of engineers that they have working on this type of project unless there is something more. I suspect Apple is going to release MacOS X on Intel. Now I may be way off base. BUT I don't think there is any doubt that Apple wants to control its own destiny when it comes to being able to move to new platforms.... something is going on here.
I know someone will chime in with, "Developer tools tend to have hard core engineers assigned to them." To which I agree. But developers with these backgrounds seem very, very out of place. Can't say more than that.
Yellow Box? Now you're talking.
It would make Apple's life easier to be able to cross-compile to Windows. It would make their developers' lives easier (as CodeWarrior does now). It would position the Mac as the go-to developer platform for targeting just about everything, and encourage Mac versions of products.
This would be a resurrection of another one of NeXT's dormant capabilities, not anything new. But if they really, really, want developers to love them and use their dev tools, this would be a major step forward.
Originally posted by Amorph
Yellow Box? Now you're talking.
It would make Apple's life easier to be able to cross-compile to Windows. It would make their developers' lives easier (as CodeWarrior does now). It would position the Mac as the go-to developer platform for targeting just about everything, and encourage Mac versions of products.
This would be a resurrection of another one of NeXT's dormant capabilities, not anything new. But if they really, really, want developers to love them and use their dev tools, this would be a major step forward. [/B]
Yellow Box functionality would be huge.... so let's take it the next step. Does Apple intend to release more applications on Windows? What does this mean for the Mac?
Let me give you an example of the way Apple thinks these days. Why release a calendaring application. There are a billion out there. Sure you need one to sync with, but it's really not something that makes sense for Apple to invest manpower in. Well, it turns out that Apple is sick of paying for their meeting maker app and wanted to build one on their own for company use. So they start the iCal project which enhances MacOS X and will ultimately serve an internal corporate purpose.
I think XCode is in the same vein. Apple wouldn't invest this kind of manpower just so that developers could cross-compile. They must have some kind of internal purpose for it... more than just keeping iTunes in sync across platforms.
Originally posted by dogboy
What does this mean for the Mac?
BSD* is dying
If Apple can use its powerful development toolset to sway some body of developers to develop to its APIs using something like "Yellow Box", it could be great boon for OS X and Apple.
This is akin to the original Java strategy (well, not the ORIGINAL original, but the new original...after the setop box language birth). Cross-platform development. But better than Java.
With regard to OS X for Intel. I don't know. It is a very small assumption (for me) that OS X for Intel lives within Cupertino (behind closed doors). But the issues about OS X for Intel are more business and political. I think OS X for Intel might just be the "shotgun" Apple needs to keep Microsoft honest. It is the "end game" scenario.
Apple is a different kind of company than Microsoft (in soooo many ways it is hard to count ;-))...Apple is a different company than Dell.
They ARE a hardware company. They ARE a software company.
Maybe better yet, they are a "solutions" company. They want to sell you "the whole widget" so you can pull it out of the box, get to work and not worry about everything else.
Microsoft sells junk. Dell sells junk. Yeah, its all cheap to acquire but you pay for it in Tums and Excedrin for keeping them damn things running (and virus-free). Macs/OS X aren't perfect...but they are a damn shot closer than Windell/Wintel.
I don't know if I WANT OS X on a piece of junk Dell.
I want a daman computer that just works so I can get my job done. Period. I think Windell/Wintel is weak in this area. Fast? Sure. Cheap? Yeah. Cumbersome? Oh yes. Clunky? Yep.
Worth it? Not for me. For some people it is. Fine.
I also don't know if OS X on Intel makes much business sense for Apple.
Sure, they can turn themselves into a software company (a la MS). I don't think they own any manufacturing plants any longer so their direct capital investment might be easy to get rid of. It is in the transition where things get tricky.
I say wait a few years (at least). Let's get the OS 9 transition completed. Then let's get rid of Classic. Now we have a platform to move forward from. Apple owns (via NeXT) patents for something called "fat binaries". This enables the development tools to actually build a binary application capable of running on multiple CPU architectures (SPARC, PA-RISC, Intel, PPC, etc.) It is a compiler switch. Boom.
Now...instead of OS X on Intel...how about OS X on Sun workstations...then HP workstations...then IBM workstations...then well you see...pick away at the fringes first. Maybe double revenue in the course of it all.
Now you've created a viable secondary desktop/workstation platform. During all of this comes...say..."Yellow Box"...get some of those Windoze developers on the bus.
Finally...if you really want to...OS X on Intel. Maybe Apple is able to grab 25-30% market share over a few years.
Suddenly we have a ballgame.
I don't know. It's Friday afternoon. I should be going home. I'm rambling. Probably senselessly. But those are my thoughts.
Originally posted by dogboy
But there is more to this... you don't assign/hire the caliber of engineers that they have working on this type of project unless there is something more.
Why not? Apple obviously needs good development tools for internal use, and it's vitally important to encourage third party app development. And if Apple were going to release Mac OS X on Intel, the Xcode group would have little to do with it. Port Quartz, Carbon, and Cocoa, and then most apps, including Xcode, will just work. (There's probably some processor-dependent stuff in Xcode, but not a lot).
. I know, I know... the G5 is here. Why would Apple need to go Intel. Good question, I wonder the same thing.
And the answer is, they don't. There's 3 reasons why Apple would release an x86 version:
1. The PPC is too slow. If IBM hadn't come along with the G5 Apple probably would have introduced Macs with Xeons or Itaniums this summer.
2. The PPC is much faster than x86. In this case Apple can release the x86 version without much fear of cannibalizing Mac hardware sales. The G5 is great, but not to this level yet.
3. Nuclear war against Microsoft. If MS discontinued Office for the Mac, or took other actions aimed at completely eliminating all non-MS alternatives (as may happen in the Longhorn/Palladium timeframe), Apple would go all out to stop them. This is obviously very risky, and would probably not succeed (but it *might*, which is part of why Microsoft continues to tolerate Apple's existence).
None of these conditions exist right now. I have no doubt that Apple has OS X running on PCs internally, and may even have somebody working on keeping the architecture-dependent parts of Xcode working on x86. But I can't see any reason for an imminent x86 release.
Now I'm thinking WINE style compatibility... no Windows needed.
The other alternative is using x86 in Servers perhaps, or other such things. Keep in mind, the G5 is still a Pro chip. Sony is jealous.
Consumers? I keep thinking about this. Why don't more switch? Familiarity. There are 30 million (or so) people still on some version of Windows 9x. They will only upgrade when their hardware dies and the 95+% market share has made any non-Windows OS the equivalent to non-English speaking.
For consumers, it doesn't matter what processor OS X is running on. For some enterprises it does, but not the main reason.
Apple has little to definitively gain from OS X (client) on x86.
The revolution might occur through the backdoor: OS X Server. Impress the IT staff and IT management might authorize a test box, which could become a production box, which could encourage switching within the IT department, and so on.
The OS Domino Theory, people. Linux has the momentum for server switching, but does not have the follow through to the desktop. OS X has both.
But again, your notions of OS X domination once an x86 port rolls out? Forget them. Their not going to switch. Just don't burden yourself with it. Waste of time, effort and hope. Forget about 49% in 5, 10 or 20 years. Cast away your dreams of a chewy 30% of the market. (Okay I'm over doing it)
10%, a good and healthy target rooted in reality.
Screed
If Yellow Box for Windows was released, the quality of Windows applications would go though the roof, eliminating a huge advantage of Mac OS X (better applications).
Barto
Yellow Box -- lets you run software written for the Rhapsody OS
Red Box -- lets you run DOS/Windows software
http://www.lowendmac.com/musings/boxes.shtml
Google is your friend.
Just because it seems like a good idea to do a thing does NOT mean that Apple is, in fact, working on it.
Admittedly, the case for the argument that Apple is maintaining development of OS X on Intel hardware is quite strong; to the point where it is accepted as fact. I don't dispute the argument that such a project exists at Apple; not in the least.
However, Apple has demonstrated that it is not so forward-thinking as everyone believes. iTunes for Windows, for instance; Apple placed an ad for a Windows programmer in early 2003 (no earlier than February) that explicitly mentioned iTunes for Windows development. (Still, seven months is a great turn-around for an app)
As another example, the iTunes Music Store for Canada has only begun negotiation in the last two weeks according to reports out of macrumors.com - who published communications between the Canadian Musical Reproduction Rights Agency (CMRRA) and Canadian Recording Industry Association (CRIA) - the very people that Apple MUST come to agreement with in order to even consider implementing their iTMS in Canada. In what other country on the planet would it be easier to deploy iTMS, given the logistics of copyrights and law that must be taken into consideration? Yet its taken Apple almost seven months to make the first call.
3.1416's comment "[Apple] ...may even have somebody working on keeping the architecture-dependent parts of Xcode working on x86" pushed me to think along the lines that I've just described. I know it was only an off-hand remark, one which I would LOVE to believe, let alone know to be true, but any effort to believe it has to be thrown up against my experience with Apple over the last 12 years, and that experience led me to relay those two examples of how Apple, in retrospect, clearly wasn't as forward thinking as I'd hoped.
I agree with the argument put forth by Chris Cuilla that Apple would only be in a position to consider moving to Intel after the OS 9 to OS X transition is complete, after Classic has been put to sleep, or possibly when Microsoft decides to drop support for Mac.
I think the XCode Dev Tools, which are really only an update to the Dev Tools that existed since OS X 10.0, came out of Apple's own in-house needs - like their development of iCal, as someone mentioned here, Keynote, and to a more subtle extent, WebObjects - which is an Apple-owned technology used to power their most ambitious in-house efforts, for example, Apple's WebSite, The Apple Store, and the iTunes Music Store. I think the Dev Tools simply needed an update to help Apple take their core toolset further along - to make it possible for them to more easily follow their existing software roadmaps for iTunes, Web Objects, etc.
Of course, I have no inside information. I have not met actual programmers at Apple as DogBoy has. I'm just an armchair theorist.
Originally posted by Brad
Blue Box -- lets you run MacOS software
Yellow Box -- lets you run software written for the Rhapsody OS
Red Box -- lets you run DOS/Windows software
http://www.lowendmac.com/musings/boxes.shtml
Google is your friend.
Thanks Brad....learn something new everyday.