Apple Seeds OS X Mavericks 10.9.2 Build 13C62 to Developers

in apple they call federighi "mister 30 fps" for a reason sadly

Time for an executive decision and remove him from OS X engineering. Nothing against him, I'm sure he's a brilliant programmer, but he's been in charge of 3 OS X releases (Bertrand officially left in '09, Federighi helmed 10.7 but wasn't officially placed on Apple's board until '11), about average for an Apple's SVP of OS X engineering. Since OS X moved to an annual release cycle to mirror iOS (compared to 1 ½ - 2 years previously), aspects seem [more] half baked, rushed and unpolished. It shouldn't take .1-.2 updates to debug core components when 6-12 months for more 10.X development would make more sense. We'll be ready for 10.10 soon! That's ridiculous.

Time to give someone else a chance (or coax back Serlet) and return to an OS X development cycle that makes more sense than rushed annual releases.

Loving that they are actually working out all the bugs and keep polishing before a final release.

The last update caused iCloud and Keychain to go crazy, prompting login popups for many keychain access options that would not go away. Also, System Preferences froze if I attempted to view iCloud settings, and Safari froze immediately upon startup every time, rendering Safari unusable.

This morning, I did a clean install back to OS X 10.9.1. Don't think I'll be updating to 10.9.2 until the public release..which is a shame.

Didn't experience any of those issues on my MBP.

Very soon.

So many beta releases. It's almost like they're just beta testing their new beta delivery software. :)

60 fps animations would be a dream come true on my 2010 MBP.

at apple they call federighi "mister 30 fps" for a reason sadly

Google "Mavericks 100% kernel task" and peruse any of the hundreds of threads on the subject.

Works for "Snow Leopard 100% kernel task", "Lion 100% kernel task", and "Mountain Lion 100% kernel task" as well.

tl;dr The kernel does a lot of things.

Time for an executive decision and remove him from OS X engineering. Nothing against him, I'm sure he's a brilliant programmer, but he's been in charge of 3 OS X releases (Bertrand officially left in '09, Federighi helmed 10.7 but wasn't officially placed on Apple's board until '11), about average for an Apple's SVP of OS X engineering. Since OS X moved to an annual release cycle to mirror iOS (compared to 1 ½ - 2 years previously), aspects seem [more] half baked, rushed and unpolished. It shouldn't take .1-.2 updates to debug core components when 6-12 months for more 10.X development would make more sense. We'll be ready for 10.10 soon! That's ridiculous.

Time to give someone else a chance (or coax back Serlet) and return to an OS X development cycle that makes more sense than rushed annual releases.

Do you really think they have someone more competent around or will be able to coax back a man as busy as Serlet? Serlet wasn't fired or forced out, he left on his own accord to pursue things he considered more interesting (apparently work on cloud computing) and is currently not only managing his own startup, also on the board of Parallels Inc. Versions helmed by him weren't prefect ether (specially 10.5, which I personally like to call Apple's Vista) and I suspect you're just being blinded by nostalgia thinking 10.7-9 are considerably buggier than what came before or weren't around reading people's comments with all the issues they had with 10.5.

As for the yearly release cycle, I get the feeling you're not really seeing the forest for the trees as you can't directly compare one version they've worked on for a year with another version they've worked on for two. Knowing that there's a number of things I'd like fixed (chief among them how badly OSX lags behind development in OpenGL), but are too big for ".1"-updates, I actually prefer a yearly release cycle rather than once every two years.

My favorite versions of OSX have been the ones where they've just stopped and focused on under-the-bonnet work like 10.6 and 10.9. I personally think every version of OSX since 10.5 has been fairly weak in the "new features" department, so you can't really blame Federighi for it on his own.

Also, even thou Federighi may have officially helmed 10.7, big software projects like major versions of an OS are often planned years ahead, meaning that while Federighi may have lead development, Serlet was in charge when the actually planning of the project was done.

Time for an executive decision and remove him from OS X engineering. Nothing against him, I'm sure he's a brilliant programmer, but he's been in charge of 3 OS X releases (Bertrand officially left in '09, Federighi helmed 10.7 but wasn't officially placed on Apple's board until '11), about average for an Apple's SVP of OS X engineering. Since OS X moved to an annual release cycle to mirror iOS (compared to 1 ½ - 2 years previously), aspects seem [more] half baked, rushed and unpolished. It shouldn't take .1-.2 updates to debug core components when 6-12 months for more 10.X development would make more sense. We'll be ready for 10.10 soon! That's ridiculous.

Time to give someone else a chance (or coax back Serlet) and return to an OS X development cycle that makes more sense than rushed annual releases.

I agree with you. However, there can be another reason: poor foundation. Perhaps, Lion served as a poor foundation for all the OSs after. ML and Mavericks are only an improvement (with some more features) on Lion. You know what they say about fixing something that's fundamentally broken to the core, it can't be. During these developers betas, even though they are taking their time, it seems that they are trying to plug leaking holes while new ones appear as a consequence. So it could be that the core of Lion was poorly made and they've been just trying workarounds to fix that. Workarounds, that cause other problems to appear as a consequence.

OK, let's put this into perspective. I give you 2 scenarios:

Scenario 1 - 2-year release:

Apple introduces 20 new features, and it takes 2 years to perfect them before introducing 20 more on the next release. Stable OS.

Scenario 2 - Yearly release:

Apple introduces 10 new features in an ODD release, and it takes 1 year to address them, and they are still half-baked. Then POOF...! No more bug fixes as Apple releasing 10 more new features on an EVEN release.

Imagine those old Macs that meet the minimum requirements to run the ODD release but NOT on the EVEN release. So that poor Mac will have to be contented with the former 10 half-baked features. Even if the said 10 half-baked features from the ODD release are finally perfected at the end of the EVEN release, so what..? They are still stuck with an ODD release with 10 half-baked features. If they do get thru that update, they still have to be contented with the latter 10 half-baked features. (This also sounds like the OS X development team never focus on fixing anything no...?)

Now re-apply this analogy of a poor Mac user to Scenario 1, what will he/she get by the end of a 2-year release...? 20 stable features regardless of whether or not he/she makes it to the next major OS release...

Are you following me so far...?

So, how can you ask someone to skip one ODD release and update on EVEN release when at the end of the EVEN release he/she may get 10 stable features and 10 half-baked features as opposed to an end of a 2-year cycle where he/she gets 20 stable features...?

Your scenarios are imaginary. You either get 10 stable features every year and 20 stable features every 2 years, or 10 half baked features every year and 20 half baked features every 2 years. This is not a mix and match situation. The amount of people working on OS X is more or less the same whether Apple chooses yearly releases or not so the stability of the OS and the new features does not change depending on the release cycle. Only the amount does.