December 10, 2005

The Silly Season.

Ok, normally I'm not in the rumor-debunking business. Because, well, I like a good rumor as much as the next guy, and, unbelievably, Apple doesn't actually call me up and tell me all their plans any more often than they call you. (Note to Bertrand Serlet: Ok, they call you a lot, but not the rest of us.)

Also, it's a bad idea for a developer to say anything about rumors, pro or con, because even touching the field gets one the reputation as a purveyor of rumors, and then all your Apple friends stop talking to you, lest they be seen associating with someone of your caste.

And, heck, it is the lead-up to MacWorld Expo (booth 710!), so we should expect some crazy rumors. Hell, the ones I believed the least have been the ones to come true, recently. For instance, up until the midnight before WWDC 2005 I was still explaining to all my friends why Apple would never announce Intel processors at a developer conference one year in advance. (Note to friends: sorry about that.)

Plus, the whole reason people plant fake rumors is so they can point out how many times it gets mentioned, so my mentioning it is really just feeding their ego.

But, seriously, sometimes a rumor is SO OBVIOUSLY A FAKE that it makes me grit my teeth, and when I see people discussing it I just want to grab them all and conk their heads together.

So it is with the Dharma / Yellow Box rumor currently mentioned on a surprising number of sites. Let's take this apart, shall we?

Starting with the really obvious:

1) The person says the project's new codename is "Dharma" just after it was revealed to be the codename of the project in "Lost", which just showed up in iTunes. The person signs his name as "John Locke," who is a fictional character from a TV show and "somewhere near Hawaii," which is the state in which Lost is filmed.

Oh, but wait, some will say, what if that's just his clever way of having fun with the secrecy around this project? And what if the project name is a double-triple cross, like that one episode of X-files where the government uses pro wrestlers as their men in black, so nobody ever believes you when you try to tell them you were visited by them?

Well, then, how about:

2) Real rumors don't spend most of their text justifying why they make sense. If someone has real information, they just say it. A real rumor would be more of the form, "someone close to the project told me at a conference that they've revived the Yellow Box..." not, "As we all know, in the beginning of computer history there were many architectures, which determine the lowest-level language that a computer understands... blah blah blah... and THAT's why you should believe me about Dharma."

Or, maybe you'd believe:

3) The letter-writer explains that he's leaking this information, anonymously, to a French site because he's afraid of the DMCA. Except, you know, he's anonymous. And he's leaking the information from the United States anyways, so he could still get nabbed if he weren't. And U.S.-based rumors sites are still going full-swing, so it's not like they wouldn't run his piece because it's too-hot-for-television.


4) Bertrand Serlet is supposed to be running the project, while he's also Senior V.P. of Software at Apple. Yes, he's got time to be running a project which is essentially just a tune-up of an old compatibility layer. That'd be a good use for a V.P. Also, he's in the States, although he does still have a great French accent.



Seriously, people. Apple doesn't WANT your current Intel machine to run Mac OS X software. If it could, they wouldn't be able to sell you a new machine in June of 2006. Trace the dollars! How would Apple profit from this?

It's been demonstrated (by the previous Yellow Box) that big developers won't just write their software once for Yellow Box and then "call it good" on Windows. Hell, Adobe can hardly be convinced to come out of the CFM closet, much less dump their Windows codebase in favor of a Cocoa one.

Sure, it'd be great for small developers. Heck, I think mine was the only company in the world to actually SELL consumer software bundled with Yellow Box in the old days (OmniWeb for Windows!), and, to be truthful, most of the copies we sold were from big organizations who wanted Yellow Box for their custom applications and discovered it was cheaper to buy OmniWeb for Windows and just use its Yellow Box license than to buy licenses from Apple.

But, Apple doesn't really need to attract small developers to Cocoa. They've got a pretty rich small developer community already. Small developers are starting to figure out on their own that using Microsoft's tools to write software for Microsoft's operating system and then compete with Microsoft's monolithic consumer apps is putting your head in the lion's mouth. (If you disagree, let me make you a deal: I'll sell you all the tools you need to write software to compete with Delicious Library. Just send me a check for $600 every year and, uh, I'll get back to you... maybe. I totally promise the tools will be really great, too. Really.)


Now, I should give the standard disclaimers: I don't have any inside information, I've just been in the industry for a long time. Maybe I'm wrong. Maybe, come January, Apple will announce they're releasing Yellow Box and it's at a low enough price that we developers can bundle with it and it's small enough that we don't have to require users to download and install 5GBs of libraries with our apps.

And, I promise, this will be the LAST RUMOR I EVER BOTHER TO DECONSTRUCT.

Until the next one.

Labels: ,


Anonymous matt said...


December 10, 2005 3:38 PM

Anonymous Gernot said...

Look at the leaked Google Earth beta. It's ugly, and doesn't have a decent Cocoa-ish Interface because it has to be portable. In an ideal world, there would be a convenient cross-platform way to implement a first-class UI, that looks and feels like it's at home on the mac, and works just as well on windows. Currently, there is no such thing, and we are very far away from it.

Great mac apps stay mac apps, however the first thing people ask if you tell them that you've developed a great Cocoa app is "how portable is it? Can we expect a windows version?". And if you've developed a really portable app, it looks like crap on the mac. Making it easier to make your apps portable and be a great mac app is totally the right thing to do. I think Apple knows that, and "Dharma" doesn't look unrealistic to me. It would certainly make users happy, it would make the developers' lifes easier, and I don't think Apple would sell less macs if they do it.

December 10, 2005 3:50 PM

Blogger Hieronimus said...

My 2c?

I think you *do* know people in Apple (you've said it yourself before) and you're just reinforcing the rumor (mentioning Wine when you previously encouraged us to Wine?)

I think you are part of Apple's PR machine, although an independent maverick.

Not that any of this is a bad thing, or that you're a bad guy. I just think we're being wrapped in a triple riddle. Or something. Too many beers.

December 10, 2005 4:42 PM

Anonymous Anonymous said...

Don't believe him!
I'm running a developer beta of MS Delicious Library 4 for Windows 98, using Apple iYellowBox 3.2!
You can believe me because of the fact that I know the secret version number of the yet-to-be-released iYellowBox software.
But, no i can't tell you what it is.

December 10, 2005 4:54 PM

Blogger Wil Shipley said...

"Only the true Messiah denies His divinity!"

December 10, 2005 5:20 PM

Blogger John Siracusa said...

"Develop for it? I'll piss on it!"

-- Bill Gates on NeXT, revealing the true origins of the "Yellow Box"

December 10, 2005 5:50 PM

Anonymous David Young said...

60% of rumors are a manifestation of someone's unfulfillled desire. The rest probably have to do with alcoholic beverage consumption.

December 10, 2005 6:00 PM

Blogger Jonathan Ellis said...

On the last point -- it's not like Apple is above competing with ISVs, either. Audion, konfabulator, omniweb...

December 10, 2005 6:14 PM

Anonymous Bryce Kerley said...

"Look at the leaked Google Earth beta. It's ugly, and doesn't have a decent Cocoa-ish Interface because it has to be portable."

To be fair, it doesn't have a good Windows interface either, and didn't back when it was single-platform.

December 10, 2005 6:53 PM

Blogger Sören 'chucker' Kuklau said...

"In an ideal world, there would be a convenient cross-platform way to implement a first-class UI, that looks and feels like it's at home on the mac, and works just as well on windows."

I sincerely disagree.

"Currently, there is no such thing, and we are very far away from it."

And I hope we'll never get to that point.

Cross-platform UIs will always bear a "lowest common denominator" principle: in order to make something work the same way on multiple platforms, you have to make use of as little platform-specific features as possible.

Here's an example. If you were to implement a text field the same way for your Mac OS X, Windows XP and *nix (via, say, Gtk+), you wouldn't benefit from spell-checking on Mac OS X, nor on Gtk+, because Windows XP happens to not support spell-checking in text fields at all, and because Mac OS X and Gtk+ implement it in different ways. So you could either leave the feature out altogether on Windows XP and implement it differently on Mac OS X and Gtk+, but then you'd have a lot of platform-specific code, not to mention Windows users would feel left out. Or you could add even more platform-specific code and write a custom implementation on Windows, thereby probably breaking support for random piece of software X -- say, a screen reader. Thanks to your spell-checker, vision-impaired people on Windows could no longer use their screen reader on your text fields because it no longer recognizes text properly. So you could add an option, thereby cluttering your Preferences window, thus pissing off your UI Design department.

Finally, you would realize it isn't worth the effort and leave out spell-checking altogether, for the sake of cross-platform. Hooray.

Cross-platform GUIs are bad. Cross-platform *back-ends* (with architecture-specific optimizations, if applicable) are good. Check Skype for an example: they have the same basic code (although not up-to-date on all platforms) on Windows, Mac OS X, Linux (Qt) and Windows CE/Mobile/Pocket PC/whathaveyouit'sallthesamethinganyway. Yet, their UIs are (arguably) quite fitting into their specific environments. The Mac OS X UI uses drawers (maybe even overzealously) as well as sheets, uses the fully customizable NSToolbar, makes use of Growl for notifications, etc. It feels like a Mac app. And yet it works just fine with people who are on other systems.

And yes, this post was way too long. My apologies.

December 10, 2005 7:12 PM

Blogger narya said...

> The person signs his name as "John Locke,"
> who is a fictional character from a TV show

??? ;-)

December 10, 2005 10:56 PM

Anonymous Gernot said...

I agree that the kind of Cross-platform UIs like you describe them are a bad thing. But if you're using Cocoa the way Apple wants you to use it (and the way it is the most fun) you'te using Objective-C, CoreData, Interface Builder, Bundles, etc... A lot of things that require a 100% rewrite when you want to move it to other platforms. Apple should make porting easier to encourage the use of those technologies, if they don't they aren't used!

Apple already did it with some portions of Carbon for porting iTunes. This is a cross platform program with a UI that works. Ok, it doesn't look like a windows program, but these times even windows programs don't. And it doesn't behave badly the way you describe it. Providing tools that don't require ports of Cocoa programs to be rewritten completely would make a lot of sense for Apple.

Apple wants Developers to use Cocoa. Developers want to use Cocoa. But they don't as much as they and Apple want to, and the reason is portability. Apples whole Application Unit is tool to show the world which kind of apps are possible if you use the tools right, yet too few do. I'm pretty sure a Cocoa-on-Windows project exists, and if it's just another just-in-case project in the basement of building 5 that gets pulled out of the hat at WWDC 2009.

December 10, 2005 11:51 PM

Blogger Hieronimus said...

To Sören 'chucker' Kuklau:
I believe Skype is developed using the big daddy of x-platform development: the Qt library from trolltech. And it is exactly that Qt library that allows great x-platform development without working on the lowest common denominator.

December 11, 2005 5:00 AM

Blogger Hieronimus said...

And truth be told, Apple's attitude towards ISVs is a bit scary. So you have an idea, you develop it, you set up your website, and then they go and develop their own version with a team that has infinite resources compared to you.

I don't know if I want to go there.

December 11, 2005 5:02 AM

Anonymous leeg said...

You forgot OpenStep for Solaris - that'll be back again soon enough. And it's part of a project being overseen by Phil Schiller and Andy Bechtolsheim.


B.A. Barracus.

December 11, 2005 5:22 AM

Anonymous Anonymous said...

They are certainly doing it, or else that would mean they are dropping itunes for windows.

We all know they are currently porting itunes to cocoa for the intel macs.

The simplest way for them to keep the windows version to sync is to update their "yellowbox"compatibility layer".

Will they release and licence it separately ? That is a completely different question...

December 11, 2005 6:13 AM

Blogger cjwl said...

I would not be surprised if they had this as an internal project, it's not a very big step, considering what they have on Windows already. It makes a lot of sense for Apple to use this for their own application efforts, they can come out with key cross platform applications faster and that are easier to maintain. It would not even be a big deal to make it part to the QT distribution.

I would be very surprised if they released it, supported it and made it something developers could count on. I'm not sure how this would help them. Maybe only the dev. tools run on a Mac as a way to bait more developers to OS X? To sell developer boxes? If they're going to write for Windows anyway, why not have them on a Mac doing it, and attract them to Cocoa? Could they get Aqua going on Windows before Vista comes out to show up MS? Most of the scenarios I can come up with are a bit of a stretch, but then again I'm not running Apple.

Windows is such a mish-mosh of UI kits that they could leave Cocoa/Aqua as-is and users would barely blink at the different appearance or behaviour. The most popular programs on Windows don't use the standard controls and have their own UI kits anyway.

It's something that could happen, but I doubt it will happen anytime soon. If it does, I would be afraid of using it considering what they did with YellowBox in the past.

December 11, 2005 6:34 AM

Anonymous patrick said...

All I can say is: if someone of the statue of John Locke (1632-1704) rises from the grave to spread the good news that Cocoa will be set free, it must be true.

The fact that the person who leaked this info / started the rumor lacks the imagination to even think up a non existing alias bodes well for the reliability of his message; he just couldn't have made it all up :-)

Besides: it makes sense: sooner or later, but likely sooner, Safari will come to Windows. iTunes and Quicktime are already there. A suite of media players, with a common, (non-standard, non-Windows) interface. Why not base them on a common, portable (subset of) Cocoa? Will it be a private framework? Probably. Will it remain private? Who knows...

December 11, 2005 6:37 AM

Anonymous Anonymous said...

To Sören 'chucker' Kuklau:

Look at cross-platform toolkits like qt or even gtk+. They all look fine on Linux and on Windows. KDE, written using qt, looks great on Linux, and Photoshop Album, also written with qt, looks great on Windows.
However, applications created with these toolkits look like crap on the Mac. If Cocoa would be cross-platform, then developers who like the Mac but have to develop for Windows and/or Linux, too, could use Cocoa. They could develop a great Mac OS X application (including spell checking, sheets, etc.) and then publish a Windows version with ease. It's possible that this application would not make use of all Windows technologies, but to be honest, that's not really important, because Windows users do not expect applications that conform to a standard.
E.g., look at Firefox and Thunderbird. They use Mac-like toolbars, even on Windows. Have you ever heard a Windows user complain about that? I certainly haven't. (However, I have heard Mac users complain that while the toolbars are Mac-like, they don't work exactly like "real" Mac toolbars.)

So I believe that a cross-platform Cocoa would be a great cross-platform toolkit, because applications created with it would look great on the Mac, and would be good enough on Windows to please Windows users.

To ogami itto:
No, the Mac version of Skype isn't written using qt. If qt was that good on the Mac, it would be a great toolkit even for Mac-only apps. Unfortunately, it isn't...

December 11, 2005 6:39 AM

Anonymous the Snail said...

Don't listen to him! Ssssss!

December 11, 2005 7:07 AM

Anonymous Jay Tuley said...

"like that one episode of X-files where the government uses pro wrestlers as their men in black"

Don't forget Alex Trebek.

Anyway, the one thing in that page 2 rumor that did sound good was a port of Safari. Having Safari on windows is key to having more websites work with Safari as sites can be tested for it without getting a Mac. I think American Airlines site still doesn't work for buying tickets with Safari.

December 11, 2005 9:15 AM

Anonymous Anonymous said...

This is the most authoritative source for the "I'll piss on it" quote that I could find out of the hundreds of urban legend type postings on the subject:
"Develop for it? I'll piss on it." - Bill Gates, in response to InfoWorld's Peggy Watt asking if Microsoft would develop applications for the NeXT Computer.

December 11, 2005 9:27 AM

Anonymous Erik M. Buck said...

The precursor to Yellow Box, Openstep, was deployed on Windows before it was ever deployed on a Mac. Openstep was a very nice solution for cross platform development. You could use Interface Builder, Project
Builder, EOModeler, D'OLE, PDO, etc. and deploy for Openstep Mach 68K, Openstep Mach x86, Openstep Solaris, Openstep for HPUX/Pa RISC, and Openstep Enterprise for Windows NT/2000.

There is/was a difference of opinion about how "native" an Openstep application felt on each platform. From my perspective, Microsoft tried so hard to copy NeXTstep (the precursor to Openstep) user interface conventions when they built Windows 95 and Windows NT that Openstep applications felt more native than former Windows 3.1 applications felt. My company's products were much more widely deployed on Windows NT than any other platform, and we never had a single complaint about the products' integration with the platform. In fact, we provided some components of our Openstep products as ActiveX components so they could be embedded in MS Word, Excel, etc.

Three years after Apple bought NeXT, Apple refused to sell more deployment licenses for Openstep at any price thus reneging on their often repeated commitments to provide "no or low cost" Windows licenses of "Yellow Box".

I might add that folks citing Qt as representative of what is possible clearly never used Openstep Enterprise. Qt is the stone ages in comparison IMHO.

December 11, 2005 9:39 AM

Anonymous Anonymous said...

I'll just pipe up here and say, it ain't gonna happen.

I lobbied for Cocoa on windows, both from the inside and the outside, and the chances of such a thing ever coming to pass are miniscule, for all of the reasons that Wil cited, and many more.

To put it in Marxist terms, if you do too much to relieve the suffering of the masses, then you only delay the revolution. It does not benefit Apple to make life better for Windows developers.


December 11, 2005 9:57 AM

Anonymous Anonymous said...

Reading blog comments is really something funny.

I really like when someone post is said to be from Anonymous and that person signs with its initials.


December 11, 2005 3:11 PM

Blogger Wil Shipley said...

Anonymous: We all know they are currently porting itunes to cocoa for the intel macs.

Seriously, we all what? How do we know this, again? All of us?

December 11, 2005 4:55 PM

Anonymous Anonymous said...

I don't have a blogger account, I don't feel like making one, so I post without logging in and sign it. Any objection?


December 11, 2005 5:50 PM

Blogger Sören 'chucker' Kuklau said...

"Look at cross-platform toolkits like qt or even gtk+. They all look fine on Linux and on Windows."

A-wa-what now? Gtk+ looks like complete and utter *crap* on Windows. Ever used WinGaim? That horrible, atrocious GUI that only deserves to be used at all because all other instant messengers on Windows are so bloated and royally stupid as to be even less usable?

Qt? Yes, it's fairly decent on Windows and Linux. But you completely missed the point of my post: Qt looks good because it doesn't allow for anything platform-specific at all.

Same for WxWidgets, FOX, etc.

Some of these frameworks use native widgets (thankfully), others don't (e.g. Qt). Even those that do, however, manage to fail at providing a good UI where it's needed the most.

And yes, Skype on Windows and Linux is Qt. Skype on Mac OS X, fortunately, is Cocoa.

December 11, 2005 8:51 PM

Anonymous Anonymous said...

Hi Sören,

ok, gtk doesn't look great on Windows. But with the Wimp theme, it certainly doesn't look like "crap". Compare this to gtk+ on Mac OS X (within X11). THAT looks like crap.

And I certainly didn't miss your point. You said that cross-platform always means "without support for OS-specific features". And yep, that's true for all the cross-platform toolkits that are available. However, this is much more noticeable on the Mac than on any other platform. qt on the Mac doesn't support spellchecking, it doesn't support integration with the address book, it doesn't even support the real Mac OS X look and feel (e.g. it still uses the Tab control look from OS X 10.2) and the simulated pulse of the default button looks horrible. Now tell me what native Windows features qt apps don't support - e.g. take Adobe Photoshop Album. When I first tried that application, I didn't think at all: "Oh, this must have been developed with a cross-platform toolkit." And that is my point. Windows users do not expect many "Windows native technologies" that are not available in cross-platform toolkits. Mac users do expect many of them and that's why there's no single application on the Mac that's written with a cross-platform toolkit without people immediately noticing that.
And that's why I said that Cocoa for Windows wouldn't neccessarily suffer from the same problems. It would be native on the Mac, of course, and so it would support all the Mac-native technologies. And on Windows, no one would miss anything, because Windows users do not expect applications to support any "native technologies".

I believe that YOU missed MY point (maybe that's my fault) and I hope I have made myself clearer now.

December 12, 2005 6:07 AM

Anonymous Magnus Nordlander said...

Anonymous wrote: The simplest way for them to keep the windows version to sync is to update their "yellowbox"compatibility layer".

Ehm, no, not necessarily. They still have to deal with the whole "We don't have IOKit on Windows"-deal, so it's not like they can just recompile anyway. I'd say it's actually more likely that just use some Windows-native thing.

December 12, 2005 6:47 AM

Blogger Topher said...

I have been advocating this for some time. I think that the best way to go about re-releasing the Yellow Box for Windows (probably renamed Cocoa for Windows, of course) is by including the framework as part of QuickTime for Windows. That way Cocoa apps could just list QuickTime as a requirement and be all set. I also believe that the UI should use Aqua widgets, even when running on Windows (as iTunes does). Developers are going to go "w00t!" over a develop-once-deploy-on-both-major-platforms workflow, and users are going to like Aqua. It's human nature to want things to be consistent, so the more that users get used to the ever-growing Aqua-themed apps, the more inclined they will be to just get a Mac where everything looks and feels the same.

Of course, the development environment should be Mac-only, so all those devs that want to benefit from the sudden increase in market share need a Mac dev box.

I think that the large influx of developers would lead to an increase in viable Cocoa software, and that the increase in software would lead to rising levels of user interest, which of course just feeds back into more developer interest.

December 12, 2005 7:53 AM

Anonymous Anonymous said...

Seeing how QuickTime is being slowly rewritten to use new OSX-only technologies (CoreAudio & CoreGraphics on the bottom, QTKit on the top), it isn't clear how they'll be able to retain feature compatibility between platforms. I assume that at some point all the guts in the middle will be replaced by a new architecture to kind of glue CA/CG and QTKit together. How will that run on Windows where those Core* technologies don't exist at all?

This has definite impact on iTunes which uses QT for its media handling and CODEC support. As QT continues to evolve more towards the Cocoa Way, iTunes on Windows will need to keep pace.

I guess I don't believe there will be a Cocoa for Windows, but I do believe there will be pieces of it ported. Apple will just refuse to license it to ISVs.

December 12, 2005 9:26 AM

Anonymous Anonymous said...

"Seeing how QuickTime is being slowly rewritten to use new OSX-only technologies (CoreAudio & CoreGraphics on the bottom, QTKit on the top) .."

QTKit is not a clean wrapper nor an abstraction. It allows access to lowlevel Quicktime functions and thus is far from clean. It's just an additional tool, nothing more, nothing less.

December 12, 2005 2:41 PM

Anonymous Anonymous said...

Yellow Box on Windows is an interesting idea but I do believe the political environment wouldn't allow it. Balmer/Gates will do everything they can to control every aspect of every electron in every machine. As we have seen they have no problem breaking the law and destroying companies and markets.

What Apple might do is place that Yellow Box on Linux, BSD and getting Xcode into the Open Source community. It's possible to push some of the Open source community in the direction of the Mac.

Think of the possibilities...

December 12, 2005 3:15 PM

Blogger Danny Cohen said...

Keep Mac software on the Mac platform. If people want to use an OS made by Microsoft or a computer made by Dell or HP, let them be screwed with poorly designed software lacking amazing features (ex. Delicious Library and Delicious Mysterious Next Product).

December 12, 2005 4:31 PM

Anonymous Jeff LaMarche said...

Wil - I've got to say I agree with you on the likelihood of this being true. But I'm sincerely hoping it's true.

December 13, 2005 8:49 AM

Anonymous Chad said...

I ultimately say, I'll believe it when I see it.

It would be interesting to see if Cocoa came out on Windows, but how well would it work? Would it do everything it should? And what about if you need UNIX or Carbon support in cases where Cocoa doesn't provide the functionality.

If it did happen, it might help promote more Cocoa development in this world, but I think it depends if Apple wants to spend the resources on supporting Cocoa for multiple platforms.

December 14, 2005 6:30 PM

Anonymous Anonymous said...

To the person commenting about quicktime on windows, if you check the install folder, core audio and core video are installed as well.

December 17, 2005 1:43 AM

Anonymous Anonymous said...

The intel macs are the yellow box.

December 17, 2005 11:36 AM

Blogger Troy Phillips said...

I agree with Wil that this is an obvious fake rumor (just the like the Mac moving to Intel rubbish).

I imagine that Apple do have Cocoa for Windows running internally and I just thought of one circumstance that they might release it externally - use TPM to lock it to Apple hardware! If a company switches to Macs, then there might be some people who must stay with Windows (VPC no good enough?). If they have Cocoa for Windows, then they can ease the migration from Windows to OS X.

Unlikely yes, but more likely than Apple releasing it to the whole world I think.

December 21, 2005 11:06 PM

Anonymous Theo said...

I was surprised, very, when Apple actually went ahead with the switch to Intel (and I'm pretty damn sure that just about everyone else was as well, although the usual Mac amnesia has since set in, now making it seem obvious and natural).

While I also seriously doubt the seriousness of the Yellow Box rumour, I wouldn't put it past Apple. Apple is going to have to face the fact that Macs will dual boot Windows in future anyway and that that could dissuade Windows developers from porting their software to the Mac.

If Apple decides that attack is better than defence, there are a number of very good reasons for reopening the Yellow Box for Windows and porting Cocoa, the iApps, and possibly XCode and IB (although this could work against them).

Firstly, it would open the Windows market to the thousands of small independent software makers using Cocoa on the Mac.

Secondly, it would strengthen the Mac platform, much as .Net has strengthened Windows.

Thirdly, it would possibly bring even more people to Apple as customers in that the iApps are not free. The iPod really took off when it became available for Windows as well the Mac. Apple mainly wants customers, and if Cocoa on Windows will bring customers in droves, I am sure that Apple will release it.

Of course, in reality, I have no idea. But I love this wild speculation before a Mac event. It's better than movies.

January 09, 2006 12:53 PM

Anonymous Anonymous said...

Qt4 on Mac is using native widgets and all it's drawing of objects goes through coregraphics.

It is not emulated in any way.

It also compiles on the Intel Mac with no problems. Basically Trolltech made no assumptions about the endianness of the processor and it was all about good engineering.

Toon boom studio uses Qt and that looks pretty native to me.

February 03, 2006 8:59 AM

Anonymous Anonymous said...

The big doubt I have about the current Yellow Box resurrection rumors is that the architecture has changed signficantly since the original NeXTStep for Windows. IIRC, the original Yellow Box was *just* the NeXTStep (now, Cocoa) framework hosted on top of another operating system.
Today's Cocoa framework has dependencies on Carbon, CoreFoundation, OpenGL, QuickTime, etc., etc. Given these dependencies, I tend to think that the creation of a Yellow Box for Windows that would be a near complete compatibility environment for Mac apps to be hosted on Windows would be a very significant task. Not at all like the original Yellow Box.

April 05, 2006 10:50 AM


Post a Comment

<< Home