Uh, yeah. It will be a 100% replacement of all the OS code? And yet somehow all of the new code will just work and be bug-free?
The point being no original code has not needed a re-write.
Uh, yeah. It will be a 100% replacement of all the OS code? And yet somehow all of the new code will just work and be bug-free?
I won't argue with you. However I worked in some companies where the mandated practice is to duplicate entire code sets for the same functionaility on the differnt hardware or OS's; rather than just a one line check for that particular command. so the fix has to be implemented twice. I hated doing updates for those companies as it meant double the work, double the testing, and douple the size of a release.
Sometimes you have to do that because of system calls are different. Are you a developer? Do you look at what is under the hood to get to the binary and assembly level (languarge the machine actually understands, not the english-like or code we write in)? I am glad I only had to do that a few times and abandoned that years ago since software companies write apps and OS's to create programming languages that break it down for me, so it looks like I am only clicking a checkbox.
I am done with this argument. non-developers and non-system people do not know what it takes under the hood. I do not care what platform - in the end it all boils down to code to make it work for specific hardware, and code to interpret pretty code (as in the case of Visual studio, Java, etc). you have code in the OS or firmware that breaks down your pretty code to run the hardware. Some do it more efficiently than others.
The point being no original code has not needed a re-write.
The point being no original code has not needed a re-write.
Of course Apple should work on new versions, but what is clear is that there were huge numbers of problems with 10.5 evidenced by all journalists, the traffic on all Mac sites including Apple and here and feedback from Apple Stores.
Except for the simplest apps, all software ships with bugs, it's just a question of what is acceptable.
Your original post read as though you were saying developers had twice as much work to do to support PPC and Intel, which is why I said it was overstated. If you're now saying that you really meant that the resulting compiled binaries will be larger, then I would agree with you more (although it's important for people to remember that Universal Binary apps are not twice as large as single architecture apps because most of the size in an application is taken up by resources, etc that are common across all architectures. Similarly, some UB apps actually have 4 architectures in them: 32 bit and 64 bit PPC and 32 bit and 64 bit Intel, but they aren't 4 times as large as single architecture apps).
Having said all that, I'm still confused by your comment about the "if..then..else" statements: This reads as if the created app makes decisions at runtime based on the architecture. In fact, the compiler creates separate executable binaries, one for each architecture, that are then bundled into the app bundle. In effect, you have multiple compilers running on the single source file generating different compiled apps.
I went ahead an upgraded to 10.5.3 the other night and I haven't had any problems. Really, I have not had any OS problems with 10.5.2 either (but that's just me); I do run into issues with Safari, though.
The problem is that if Apple does not address issues with an update in a timely manner, users complain-problems not being resolved quick enough.
If they seed .n right after .m is released to the public, users complain-it's too soon.
Then there is the rumor or two about 10.6, and users complain-too soon, no support for my Mac.
Either way, Apple can't win. But on the bright side, with Apple, we're a lot better off than the M$ world! It's all about perspective. Maybe we can use our energy to steer Apple toward fixing those specific issues we know are out there-maybe some user driven milestones.
This is reflected in the major rehauling of the updates. It could well be 10.5.9 will feature NONE of the code of 10.5.0 through updates. Admission that there was something seriously wrong with Leopard. With 500MB updates, we are showing huge levels of rewrites.
With all due respect, this is the stupidest thing I've heard on the internet in a long time.
I can absolutely, 100% guarantee you that 10.5.3 does not contain "huge levels of rewrites". How naive and ignorant of software development are you? Blah, blah, blah
You need to come back to reality.
The reality is Leopard had problems.
As to my references to code, I am writing as a layman. When I write 'code' I really mean features needing rewrites to perform as they are intended. The bare bones of how you achieve that and UNIX coding from the 1970s leaves me, and most Mac users (although admittedly not so much here), cold.
10.5.4 was so small (55 megs) I think all it changed was the version number in About This Mac.
People don't expect support forever, they just expect it for a reasonable length of time, and 3 years is on the short side.
10.5 = proof of having too many irons in the fire is a bad thing
I don't know what's funnier; the people who are debating the name of 10.6 (who cares!!?), or the PPC users who are so surprised by the "abandoning" of their G5's.
I don't know what's funnier; the people who are debating the name of 10.6 (who cares!!?), or the PPC users who are so surprised by the "abandoning" of their G5's. I wouldn't be surprised if you guys were also pissed they removed classic support too.![]()
Which is why removing PPC is a good thing. It's dead. It's been dead for a while. Get over it!