Draft zdnet blog entries

% fortune -ae paul murphy

Apple Leopard Reviews

It seems to me that most of the reviews I've read of Apple's recently released MacOS X 10.5 "Leopard" releases completely missed the point.

First, the focus on user features led reviewers to miss almost all of the strategic implications. Thus not a single reviewer mentioned the change to Intel as both a key delay driver and a source of increased security concern; nobody mentioned decreased PPC performance as a consequence of x86 oriented kernel and compiler change; nobody mentioned the (apparent and surprising) lack of specific adaptations to the ARM architecture; and, I saw nobody discussing anything they saw in the feature set in relation to its role on the iPhone.

Now, obviously, I didn't read all available reviews -but among those whose efforts I did read only one: John Welch writing for information Week, mentioned DTrace - and that in passing:

For developers and sysadmins who have a need to monitor the low-level activities of any application on a given Mac, Mac OS X 10.5 now comes with its own implementation of DTrace.

Similarly, none of them mentioned ZFS or its relationship to "Time-machine", and exactly nobody mentioned that the new Spaces capability (essentially the standard Unix multi-screen capability) now supports separating the display from the rest of the machine.

The second thing I noticed is part of this same business of missing the forest for the trees: ignoring the thing's strategic implications to focus on feature comparisons. The issue there was one of valuation with both the PC and Mac camps blindly following their respective party lines when drawing conclusions based on their feature reviews. Thus the Mac reviewers established their credibility as independents by critizing some minor points but professed, in the end, to love everything about it, while the PC people established theirs by praising everything they choose to admit seeing while finding some remarkably compelling reasons to ultimately reject the product.

My personal favorite example of this syndrome at work appears as a "Freudian slip". Here's PCworld's leader:

Mac OS X 10.5, alias Leopard, was released last week and PC World Editor in Chief Harry McCracken gave it a try. Harry came up with a list of 25 things that could be improved. What do you think of Leopard? Have you has any problems? (sic)

Click on the story link, and you got the actual article - headlined:

21 Quibbles I Have With Mac OS X 10.5 Leopard

Oops! but unfortunately the wishful thinking behind this error reflects the review. Here's McCracken's own introduction:

Apple's operating system upgrade is full of good stuff, but there's room for improvement. Lots and lots of it. Herewith, my list of gripes, requests, and puzzlements.

...

First, a disclaimer: I like Leopard, aka OS X 10.5, the Apple operating-system upgrade that hit stores on Friday evening. True, Apple's list of 300+ new features includes several dozen I'll never touch. (A Danish dictionary! Analysis templates, whatever those are!) But even when I filter out everything that doesn't matter to me, I'm left with a long list of stuff that'll make my computing life meaningfully better. Compared to Windows Vista, Leopard is a meatier, more polished, more immediately useful, less annoying OS upgrade.

But while I've been working my way through everything that's new in Leopard, and being impressed by much of it, I've also come across a fairly long list of quirks and gotchas--the kind of stuff I hope Apple will iron out in Leopard updates. Without any further ado, here's my list...which will probably get longer the more I dig into the upgrade.

If someone published a detailed feature by feature comparison between a Volvo S60R and a Chevy Aveo in which the author eventually announces a preference for the Aveo because it can be serviced by a minimum wage teenager at Joe's Garage, you'd be outraged, right? Uh huh, a MacOS usability review by Fahmida Y. Rashid on CRN's channel web headlined: Review: Can Leopard Take On Vista? ends this way:

Conclusion: Mac OS X has attracted users all along with its easy to use interface and lack of security problems. Leopard enhances what was already a stellar operating system with several improvements making it that much harder for Windows users to ignore.

That said, Test Center wonders how much of the business community will move off Windows to adopt Leopard. Along with hardware considerations, there is a vibrant and robust development community for Windows applications. While many business software will not work on Mac OS X, with so many management tools moving to the Web browser, that may no longer deter business users. Perhaps the channel should stick with Windows XP for a little while longer.

But if you ignore the partisan reviewers and ask what the real bottom line on "Leopard" is, the answer turns out to be the iPhone - because the current Mactels are this generation's Apple IIIs. What's going on is that 10.5 is a mixed bag reflecting both short and long term agendas. In the short term it cleans up some x86 issues and offers some new user features raising the bar for Microsoft's next effort -particularly with respect to time machine because this will be hard for Microsoft to duplicate while Apple's adoption of ZFS makes it a one release throwaway for MacOS X.

In the long term, however, what 10,.5 is about is positioning Apple's application developers to jump to the integrated server/playphone world of the future - that's why there's so much Solaris and Java stuff there.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.