Slashdot Mirror
Mac OS X "Tiger" Enters Final Candidate Stage
Orangez writes "Apppleinsider.com reports that 'Tiger' reaches the final candidate stage. 'With massive software projects such as Tiger, Apple will sometimes seed several final candidate builds before one is declared gold master...'" The final release has widely been speculated to be in the next month or two.
Its far more than a point release . The fact it adds many new features to the OS it is far more of an upgrade than most. . ;)
The new search tech is fully integrated
The new G5 optimised code in the kernel is implemented iirc
The Dashboard looks very cool
A reworking of many of the key areas of the functionality of the OS
Read the article and read any review of the tiger betas out there to find out for yourself why this is more than a mere update.
The only things certain in war are Propaganda and Death. You can never be sure which is which though
For the love of God!!! Its not a point release. Man, just because Linux goes by that way of point releases doesn't mean OS X does. 10.x is not points but a full independent version, 10.x.y is a point release.
-----BEGIN PGP SIGNATURE-----
12345
-----END PGP SIGNATURE-----
Rumurs are that it will be presented this Friday (april 1st) and that it will be "unleashed" on April 15. Is it logistically possibly that right now it's not even "gold master" and that 2 weeks later millions of discs are pressed and packaged?
So I believe the 15th as release date is very improbable (by Zarquon), maybe June 6th at WWDC?
The fact that it's a a point release is basically just semantics. Apple sort of painted themselves into a corner with the name OSX. It's sort of the 10th version of the Mac OS, but the X was to make it sound cooler and sort of clever, but what comes after? OS XI? That looks weird. And a little too close to XP. So they've gone with 10.whatever, and used 10.x.x for what'd normally be considered a point release. 10.4 has been a long time coming, and it's got plenty of big changes over 10.3, such that a bigger name change wouldn't be that surprising, if apple could come up with a better name for it. That's probably why they've been making the big cat code names more official. Jaguar, Panther, Tiger...
If you want, you can complain that Apple's devaluing the normal versioning numbering system, but I don't think they'll care much if you do.
One time I threw a brick at a duck.
No, this is nothing new.
Yes, the version number seems to indicate it's not a new version but only an update. You have to simply ignore the leading "10.". It ain't that hard.
Yes, this is actually like Microsoft charging you for XP (NT5.1) after you already bought Win2000 (NT5.0) or NT4.0 or NT3.51 - the leading "10." is like the leading "NT" from Microsoft.
Yes, this is old news, but the issue comes up every time Apple releases a new version of OS X.
You know, you don't have to pay for each new OS each year. You can skip one, or hell, two if you like.
The incredibly amount of work that goes into each new major OS X version easily justifies putting a price tag on them. These aren't Windows 98 to Windows Me steps, these are considerable feature and functionality upgrades.
As to writing software for them, my understanding is that they haven't often broken backwards compatibility, and thus haven't broken forwards compatibility. If you want your app to work for multiple versions, then only use the feature set exposed by the lowest version you want your app to be capable of running on. I don't think that's creating an unfair situation for developers at all.
You're clearly being facetious, but not everyone will pick up on that. Thus, a point of explanation:
It's a mere point release, but that's an artifact of marketing. The number and magnitude of changes under the hood is incredible, with huge advances in developer productivity through tools like CoreData, CoreImage, and CoreVideo. The rendering subsystem has been worked over to the point where some operations are hundreds or thousands of times faster than they used to be, and the system takes advantage of modern GPUs to offload even more processing (formerly it was just compositing, not it's a whole lot more). Add to that new versions of Safari, Quicktime that's build on CoreAudio, and a ton of other neat stuff (Automator). You get a lot for your $140.
And remember, the 2.6 kernel was just a point release!
Looks like there will be a 10.3.9 update soon, interestingly enough.
Typically, there's a 30 or 60 day (I forget which) period, that if you've purchased new equipment you can get the new OS either for free or the cost of shipping, something like that. I know I got Jaguar for less than $25 when it came out, because I had just purchased a laptop before its release.
I worked as a Mac Genius when Jaguar came out, and there was an official cut-off date about 5 weeks before the official release. If you purchased a mac between then and the release date, you got the free upgrade. Not the stand-alone OS install, mind you, but the "drop in" upgrade discs that they toss in the boxes of new macs at the store that don't have the OS preloaded. They do have a little give around these dates if you whine enough (hope I'm not violating my NDA...=)
You CAN install the OS from scratch (you aren't forced to do one of those nasty upgrades), but you MUST have the previous OS installed for the discs to work. Which you do, so don't worry. It just means if you ever need to reinstall your OS in the event of a disaster, you'll have to install 10.3 first, then do the format-(or archive)-and-install with 10.4.
With the first link, the chain is forged.
In fact, Apple provides build configurations such that you can specificallly target, say 10.2 or 10.1 from 10.3, and be confident that you'll have the correct API & ABI versioning.
That said, with each version of OS X, shareware developers salivate to use the new features, since they often make the dirty work easier, or negligible ( for exampe, Cocoa Bindings for 10.3 ).
Obviously, the big development houses, Adobe, Quark, etc will not generally use these new features.
lorem ipsum, dolor sit amet
I don't know about 10.4 (and if I did know I wouldn't be allowed to say) but I'm guessing that it's not that much different to the way iSync works now in 10.3.
You can sync between one Mac, your phones, iPods, Palms, etc. without requiring a .Mac subscription. However for Mac to Mac syncing you do need .Mac
However, if you have access to your own server somewhere, it is possible to fake it to look like .Mac Here are some instructions on how to do that.
Note: I haven't actually tried it myself (yet)
- It's sad that this has to be said in every single Tiger thread, but Core Image/Video will not refuse to work on older Macs. It has an AltiVec fallback path that is slower than the GPU path but produces the same results.
- The real importance of CI/V is not how cool it looks applying Photoshop filters to movie trailers; it's having an advanced image and video transformation infrastructure built into the OS and available to all developers. Apple is clearly planning for the future here, and the real benefits of CI/V will not be felt until months after Tiger ships and apps start appearing that were designed taking blur/distortion/etc for granted. That 10-20% is only going to grow in the future.
"If I don't want it, it's utterly worthless" is one of the most persistent and insidious memes on Slashdot. Please don't succumb to it.Before even cracking a book, I'd probably start out with Apple's own "Getting Started" developer documentation. It's included with Xcode, and the introductory material is clear, fairly concise, and offers a lot of tutorials and code samples so you can easily see the concepts in action (this is particularly helpful if you haven't worked extensively with MVC development before).
If you intend on developing end-user stuff, be sure to check out the Human Interface Guidelines -- Mac users have expectations for how applications should "feel" and Apple has spent a lot of time and money developing and revising the HIG over the years. If it feels like a typical mediocre X11 app, it'll get torn to shreds by rabid users.
As for books...
Aaron Hillegass' Cocoa Programming for Mac OS X is an excellent primer and my personal favorite. It's not cheap at US$44.99, but well worth it. The first few chapters are essentially a Cliffs Notes version of Apple's free introductory material, and from there the book tackles a little bit of everything -- Objective-C basics, bindings, custom views, localization... you name it.
O'Reilly's Learning Cocoa (aka Learning Cocoa with Objective-C in its second edition) by James Duncan Davidson isn't horrible, but isn't the best. It also isn't as up-to-date as the Hillegass book, but they'll both be dated pretty shortly with Tiger coming out in the next few months (or a couple weeks, if you believe the rumor sites).
Once you get out of the starting gate, there aren't a whole lot of applicable books (but Cocoa and Objective-C are fairly easy to pick up). This isn't necessarily a bad thing, as Apple's mailing lists are a great resource and the developer and API documentation is quite good in most areas.
"wouldnt it be nice it they made OSX for a x86"
...
But they do, they just don't sell it.
In an interview last year, an Apple executive confirmed that an x86 port of OSX, aka Marklar does exist in Apple's labs and that they are keeping it on par with PPC development.
Before the release of the G5, Steve Jobs said in another interview that they do not plan to move to x86 but that they like to keep their options open.
If you take these two statements and add one and one together, it should become obvious that they have no intent to change their business model from making and selling "hardware including software" to "software including hardware" or even "software only". In other words, Marklar is just an insurance policy against unpredictable disaster scenarios where Apple would be forced to move to another CPU and as a result, Apple have a stronger negotiating position with IBM.
Consequently, for as long as IBM do a good job on fostering PPC, for as long as PPC is competitive, Apple have very little reason to move.
And should they ever decide to move, or should they decide to offer OSX on x86 in addition to PPC, their business model will almost certainly remain the same, meaning OSX will continue to be made to run on Apple hardware only, regardless of CPU compatibility.
So, you would then see an x86 Mac with exactly the same treats as today, from OpenFirmware to Apple's own motherboard designs, not compatible with other x86 hardware. In fact, such an x86 Mac might even have a custom x86 CPU, made only for Apple, ie bolted on AltiVec compatible SIMD. Without specific hacks, OSX would not run on other x86 machines. Likewise, Windows would probably not run on such an x86 Mac without some extra software from Microsoft, eg. Virtual PC or Mac/x86. Such an arrangement would also likely have Microsoft continue MS-Office development for the Mac - even more reason for Apple to choose such a path if they ever were to go x86.
So, whether or not Apple will release OSX on x86, if you want OSX on non-Apple x86 hardware, you will almost certainly have to rig your own.
Mind you, you can do this within limits already today. Darwin, the core of OSX, is available for x86 and it's a free download
http://www.opendarwin.org/en/downloads
You can get GNUstep and run it on top of Darwin x86
http://www.gnustep.org
GNUstep is the GNU implementation of OPENSTEP, the foundation on which Cocoa is build. In addition, GNUstep has some, but not all of the things Apple has added, so you get Cocoa compatibility within limits. This is as close as you can get OSX on x86 today. It's free, but it requires a little more effort than an OSX installation on a Mac. And if you want the OSX eye candy, you will also need to do a bit of DIY. If you do, consider becoming a contributor to the GNUstep project.
Thus, it comes down to paying a little extra for convenience or save some money and put in some work. You can't have it both ways. Remember, there ain't no such thing as a free lunch.
the macintosh asterisk mailing list http://www.astm