Welcome to the MacNN Forums.

ajbaker · Apr 29, 2005, 09:46 AM

I'm amazed there are no comments here about the ArsTechnica review of Tiger:
here

Do not even bother following the link if you dont have a few spare hours: it is both long, involving and addictive!

What amazes me even more is the abilities Tiger adds for metadata, and file typing far and beyond what Spotlight is all about - and I hadnt even heard about it until now. A year or two ago these boards were ablaze about file extensions. Now they are finally in their last few years of life - thank God (RIP) - and there is not a single mention on here. What's that all about?!

Mithras · Apr 29, 2005, 11:11 AM

Yeah, it's an incredible review. Great in-depth stuff on metadata and filetypes. John is a Mac OS review god.

mitchell_pgh · Apr 29, 2005, 11:13 AM

That review made my head hurt. Talking about bleeding edge detail. Amazing article!

Dog Like Nature · Apr 29, 2005, 11:50 AM

I loved that review. John really does know his stuff, and it's nice to see his wishes for rich metadata being (slowly) fulfilled by the Mothership.

UTIs are the future!

mAxximo · Apr 29, 2005, 12:28 PM

As one that's been hating mandatory file extensions from day one I can only rejoice when reading John's reviews. This one in particular showed a little light at the end of the tunnel on this issue.
The shameless lack of attention and sheer ill-conception of the X �Finder� is another (if not the main) gripe I've been having with OS X all these years and nobody writes better about it than Siracusa.

pman68 · Apr 29, 2005, 12:36 PM

Best review of Tiger period.

workerbee · Apr 29, 2005, 01:12 PM

Thanks a lot for the link ajbaker, much appreciated.

:: off reading::

placebo1969 · Apr 29, 2005, 01:25 PM

Originally Posted by mitchell_pgh

That review made my head hurt. Talking about bleeding edge detail. Amazing article!

I read about the first 1/3 of the article last night. Yikes. I wasn't quite sure what I was reading, but it sounded cool.

P · Apr 29, 2005, 01:37 PM

That was the best Tiger review ever, and possibly the best OS X review ever. I learned a lot fo things I never knew about the OS. Now I might have to buy it...

tigas · Apr 29, 2005, 01:42 PM

Ditto. The Panther review was hard to beat but John delivered in spades!

nforcer · Apr 29, 2005, 02:02 PM

It took like an hour to read but it was interesting. I wish more people would do thorough reviews like this. There are so many reviews of things where the reviewer seems to hardly even test it, and just regurgitates a press release.

E's Lil Theorem · Apr 29, 2005, 02:14 PM

Awesome review indeed. With all the interruptions I had, it took me a long time to finish it.

Originally Posted by placebo1969

I read about the first 1/3 of the article last night. Yikes. I wasn't quite sure what I was reading, but it sounded cool.

Originally Posted by nforcer

It took like an hour to read but it was interesting. I wish more people would do thorough reviews like this. There are so many reviews of things where the reviewer seems to hardly even test it, and just regurgitates a press release.

I agree with you, but most of the time the target audience of the reviews you're talking about isn't as tech savvy as your typical Ars reader, which is why the reviewers just end up adding their own color to the product description the company (that made the product) puts out.

Eug Wanker · Apr 29, 2005, 03:46 PM

AnandTech's Review

It's pretty long, but much more mainstream than Ars's very technical review.

Dog Like Nature · Apr 29, 2005, 04:23 PM

Originally Posted by Eug Wanker

AnandTech's Review

It's pretty long, but much more mainstream than Ars's very technical review.

Ouch. Here's a quote from the third page:

Tiger is a lot more polished than it was during any stage of the beta program, but I would hardly call it a finished product ready for retail release.

Seems a bit harsh, don't you think?

mrmister · Apr 29, 2005, 05:03 PM

Siracusa is the only review you need, if you're the kind of person who would go to a Mac forum--so if you're here, that's what you should be reading.

Eug Wanker · Apr 29, 2005, 05:11 PM

Originally Posted by Dog Like Nature

Ouch. Here's a quote from the third page:

Seems a bit harsh, don't you think?

No, I would agree with AT's assessment. Tiger is buggy. I think Tiger 10.4.0 is more buggy than 10.3.0 was, and 10.3.0 definitely had its problems.

But given all that's changed, I'm not terribly surprised. That's why it's always a good idea to wait on OS upgrades if you're conservative.

mr. natural · Apr 29, 2005, 06:04 PM

I haven't gotten far in the article but I agree with his assessment of Mail's new toolbar icon, etc., look.

It is as butt ugly as anything put out from Redmond and almost ruined my glee with installing Tiger. :/

Thank god for the article's link to "cagefighter." It'll help a little.

I'll be registering my complaint too.

Now onto the good stuff.

mAxximo · Apr 30, 2005, 12:39 AM

Originally Posted by Dog Like Nature

Ouch. Here's a quote from the third page:

Seems a bit harsh, don't you think?

I think he's spot on. If I thought Panther was beta-grade Tiger feels like 10.0.

mitchell_pgh · Apr 30, 2005, 01:11 AM

Originally Posted by mAxximo

I think he's spot on. If I thought Panther was beta-grade Tiger feels like 10.0.

I think 10.0 and even parts of 10.1 were beta, but 10.2 was solid.

Panther... beta-grade. No way.

CharlesS · Apr 30, 2005, 04:02 AM

Originally Posted by John Siracusa

The second thing in favor of the BFS approach is reliability. It's a simple fact of life that file system driver code is held to a higher standard than almost any other code�even kernel code. Certainly application code can afford to be much more lax. If an application crashes, it's annoying, but an application can always be relaunched. If the kernel crashes, well, that's pretty bad, but at least you can reboot.

But if a file system driver has a bug, it could potentially corrupt or destroy all of your data. No relaunch or reboot is going to fix that. And while application and kernel bugs can also corrupt or destroy files, a file system driver bug is much more likely to do so.

As a consequence, file system drivers are among the most reliable pieces of software in any operating system. Putting metadata indexing into the file system itself ensures that this code will be as reliable as the rest of the driver code. Users simply will not tolerate buggy file system drivers, period. One only has to witness the torture tests that BFS endured during development to see the how the reliability constraints of being in the file system driver can benefit a piece of code.

Um, isn't this actually a good reason not to put the metadata indexing into the file system drivers? Since it is completely mission-critical that the FS drivers be bug-free, it seems that adding extra features to the FS drivers could compromise their stability...

Big Mac · Apr 30, 2005, 06:27 AM

The fourth page of the review - the section on kernel improvements - is especially edifying. Thank you for letting us know about this, ajbaker.

Mithras · Apr 30, 2005, 07:19 AM

Originally Posted by CharlesS

Um, isn't this actually a good reason not to put the metadata indexing into the file system drivers? Since it is completely mission-critical that the FS drivers be bug-free, it seems that adding extra features to the FS drivers could compromise their stability...

Yeah, that reasoning seemed totally ass-backwards. "Any error here could destroy all of your data... so we should have Apple put Spotlight there, because by definition it will be error-free!" Ummm, that, or it will destroy all of our data!

Only major flaw in an otherwise excellent article.

JCS · Apr 30, 2005, 08:23 AM

Yeah, that reasoning seemed totally ass-backwards. "Any error here could destroy all of your data... so we should have Apple put Spotlight there, because by definition it will be error-free!" Ummm, that, or it will destroy all of our data!

You're not following my reasoning. I will try to break it down more simply.

File system driver code needs to be reliable. This needs causes file system drivers to be thoroughly tested. Integrating metadata indexing into the file system driver causes it to be thoroughly tested as well because no one will tolerate a buggy file system driver.

The real-world example given was BFS, which was created by a handful of people in a small company for an OS with a tiny user base, but was still torture-tested way beyond the expected levels for "normal" (non-fs-driver) code. (See Dominic Giampaolo's book for tales of the testing.)

Big Mac · Apr 30, 2005, 08:36 AM

I can see both sides of this debate. Complicating mission critical driver code sounds like the wrong thing to do. And if simply integrating external code into mission critical kernel components would suddenly make things so robust and bug-free, then why not throw in a bunch of other code bases for good measure? It also would run counter to Apple's modern coding philosophy that stresses cross platform compatibility to incorporate this data in the filesystem. On the other hand, Syracuse's point about the integration's benefits from a performance standpoint makes intuitive sense.

JCS · Apr 30, 2005, 08:41 AM

Originally Posted by Big Mac

I can see both sides of this debate. Complicating mission critical driver code sounds like the wrong thing to do. And if simply integrating external code into mission critical kernel components would suddenly make things so robust and bug-free, then why not throw in a bunch of other code bases for good measure?

Sorry, I meant to address that. Obviously you can't just throw anything in there, but managing file metadata is exactly what the file system driver is supposed to to. It's a perfect it. Moving that function outside the file system driver is the "odd" choice, not the other way around.

It also would run counter to Apple's modern coding philosophy that stresses cross platform compatibility to incorporate this data in the filesystem.

The data is already incorporated into the file system if you're using HFS+, it's just the indexes that are not. That's an unnatural, potentially synchronization-vulnerable separation. Obviously external storage is needed to support "lesser" file systems, but clearly if Apple though that was the best path they wouldn't have added "native" extended attribute storage to HFS+ (nor would MS have added it to NTFS, nor would things like ReiserFS be in development, etc.)

Dirk Lenz · Apr 30, 2005, 09:03 AM

Besides the fact that the reviewer has some strange thoughts on good GUI design, this is a very good and detailed review.

Dirk

UnixMac · Apr 30, 2005, 12:39 PM

Both reviews are great, and rather objective coming from PC dominant sites..

legacyb4 · Apr 30, 2005, 12:42 PM

He writes:

Spotlight performance

There's one slightly dim spot in the Tiger performance picture, and unfortunately it's going to be a part of every user's first experience with the new OS. When booting into Tiger for the first time, Spotlight has to index all attached drives. This is done with a low-priority background task, but there's no avoiding the huge amount of disk activity involved. Expect your first few hours in Tiger to be accompanied by the steady tick and pop of all of your hard disks.

The performance impact of indexing is negligible on any operations other than disk access. The CPU is not swamped and graphics draw just as fast as always. But launching applications or doing anything else that requires disk i/o does slow down noticeably. There'll be an increased cacophony of hard disk noises and many more bounces of Dock icons.

This effect was exacerbated by the slow external disk that I used to test many of the pre-release versions of Tiger, so maybe it got stuck in my head more firmly than it deserved to be. It's not as bad with modern desktop hard disks. The PowerBook's lower-RPM internal hard drive didn't fare as well, but then I suspect Mac laptop users are used to slow disk access at this point.

Anyway, it's all over in a matter of minutes or hours, depending on the size and number of your hard drives. Once Spotlight is all caught up, it stays up to date though that kernel magic that we love so much. This on-the-fly indexing in response to individual file operations was imperceptible to me, even when running off a very slow disk. Spotlight appears to index only the first 10MB or so of file contents by default, although metadata importers can choose to ignore this limit.

Have early installers noticed an increased performance AFTER Spotlight has finished its initial run? I was noticing a lot of posts about slow, perceived performance which can be explained by this.