• Explore the magic and the mystery!
  • The Tech Night Owl's Home Page
  • Namecheap.com





  • Does Apple Need to Rush 10.7?

    February 17th, 2010

    It is widely believed that Apple will unveil iPhone 4.0 at this summer’s Worldwide Developer’s Conference and ship it shortly thereafter. There’s also a passing level of speculation that they will also deliver preliminary information about the next great reference release for Mac OS X, 10.7, sporting an unknown feline moniker.

    The real question, however, is whether there’s a crying need for a successor to Snow Leopard, and therein lies a tale.

    Now most of you know that 10.6 arrived last August, several weeks ahead of Apple’s promised delivery timeframe, with the usual amount of fanfare. But much of the advantages of Snow Leopard lie in its potential, rather than any visible improvements. Consider Grand Central Dispatch, which provides built-in tools to allow applications to better support all those recent Macs that are equipped at least two processor cores. Consider the plight of owners of those super-expensive Mac Pros, with eight powerful processor cores and not much to do with them.

    Another advantage, enhanced 64-bit support, lets apps access more than 4GB of memory, which will surely make a difference when you are engaged in heavy-duty content creation, particularly when it involves image editing and 3D rendering. And don’t forget OpenCL, which passes off key processing tasks to the often underutilized graphics chips on your Mac.

    As you have probably noticed, precious few apps have been updated to support the first or the last, and even Apple was slow to make their products 64-bit savvy. What this means is that Snow Leopard’s promised performance improvements remain just that — a promise. Yes, 64-bit can help some, but keeping all your Mac’s processor cores active as much as possible will have a more immediate impact. Even converting a big audio file from one format to another can benefit greatly, but how many apps out there use more than one processor core?

    Now to be fair to the developers in our audience, I realize that supporting Grand Central Dispatch and OpenCL isn’t trivial. Apple provides enhanced tools to get the job done, but lots of work is still required. On the large, sprawling apps that benefit most, you can expect it will take months and months to ship upgrades.

    Just the other day, I was reading some preliminary information about the forthcoming Adobe Creative Suite 5, which will include a 64-bit savvy Mac version of Photoshop. Yes, there was a 64-bit Windows version of CS4, but Adobe gave the excuse that the switchover to Intel hamstrung their developers, not giving them enough time to complete the job and deliver a Mac version without a substantial delay. That’s because Adobe took the easy way out in the early days of Mac OS X by porting their apps to Apple’s legacy Carbon environment.

    Regardless of the pros and cons of the programming issues, it stands to reason that the expectations for Snow Leopard have yet to be fulfilled for most of you. Yes, maybe it’s a bit snappier and, for most, a little more stable. But the under-the-hood stuff isn’t doing what it was intended to do — yet!

    That being the case, where’s the rush to 10.7? Regardless of what Apple offers for the next major operating system upgrade, there will be those inevitable compatibility issues, even if the fundamentals of Grand Central Dispatch and OpenCL remain pretty much the same as they are now. Besides, how could Apple resist the temptation to tinker with a few odds and ends of the system plumbing to make it work better? No doubt there will also be interface changes and new features that developers will want to support.

    So would it make sense for them to be forced to essentially suspend their Snow Leopard compatibility projects and spend the time and money moving to an even later operating system?

    I’m sure little of this is lost on Apple. The other question is whether there’s any incentive to rush to upgrade Snow Leopard by the end of 2010 or even by the middle of 2011. Maybe there is, but I don’t see it, not when the existing operating system is working fine and Mac sales are still raging at a pretty good clip.

    While Microsoft appears to be doing well with Windows 7, in terms of early adoption and potential business migrations, there’s little evidence that they have managed to stem the gradual erosion of their operating system’s market share. They haven’t slowed Mac sales, and making the swichover from XP to Windows 7 so difficult isn’t helping matters.

    This isn’t to say that Apple will not stage a 10.7 technology demo at the WWDC. It’s possible they will, just to indicate what they’re working on, and provide a 15 to 18 month estimate for shipping a final version. That step would move the full feature rollout to WWDC 2011, which would seem a more sensible move.

    It would also help developers if Apple assured them that they could continue to deliver Snow Leopard upgrades without having to confront major compatibility problems when 10.7 finally ships. But there’s no rush for them or us.



    Share
    | Print This Article Print This Article

    29 Responses to “Does Apple Need to Rush 10.7?”

    1. "Doubting Thomas" says:

      If Apple are in such a rush to implement OS 10.7, it suggests several things.

      Number one is that the company wants to make sure they keep everyone on the upgrade treadmill.

      Number two is that there must be more problems with Snow Leopard than are generally realized. (Snow Leopard is, after all, a first generation Intel only OS product and Apple does not have a terrific track record on first generation (or Rev A) products.)

      Another thought is that it distracts attention from the other problems Apple is experiencing, such as the ill conceived iPad without Flash and the growing problems with hardware that has been released without adequate testing. Apple appears unable to get products out the door on schedule without omitting some crucial aspect of testing and development. Apple runs the risk of people eventually realizing that “the emperor has no clothes”.

      • hmurchison says:

        @”Doubting Thomas”,

        Is Apple “really” in a rush to implement 10.7? Judging from past releases 10.7 coming next year would be right on schedule.

        From http://en.wikipedia.org/wiki/OS_X :

        On March 24, 2001, Apple released Mac OS X v10.0 (internally codenamed Cheetah)
        Later that year on September 25, 2001, Mac OS X v10.1 (internally codenamed Puma) was released (6 months)
        On August 23, 2002,[75] Apple followed up with Mac OS X v10.2 “Jaguar” (11 months)
        Mac OS X v10.3 “Panther” was released on October 24, 2003 (14 months)
        Mac OS X v10.4 “Tiger” was released on April 29, 2005. (18 months)
        Mac OS X v10.5 “Leopard” was released on October 26, 2007 (18 months)
        Mac OS X v10.6 “Snow Leopard” was released on August 28, 2009 (14 months)

        So let’s average the last three releases and we get 16.66 months. Thus a 10.7 upgrade will likely come out April/May of 2011. Which means Apple needs to identify the featureset (major) at WWDC this year with plans to ship in a calendar year.

        Snow Leopard isn’t problem free but I believe it’s more stable than Leopard was by the third 10.x.x release.

        The notion that OS X 10.7 development distracts from the iPad or hardware is folly. The same core team working on OS X also work on the iPhone OS as well. The hardware team is separate from the software team so progress in one area or lack thereof doesn’t mean the other team is hampered. Snow Leopard was likely delayed because Apple had to get the iPhone out but since then they’ve delivered timely iPod Touch/iPhone OS updates without issue.

        With record sales happening for Apple quarter over quarter I think it’s clear that they are operating very smoothly with a few small roadbumps rather than resembling a rudderless ship.

        • hmurchison says:

          @hmurchison,

          Cont.

          Flash is going away

          (Virgin drops Flash)

          (Vimeo HTML5 player for video)

          Google beta tests HTML5 video

          SublimeVideo (another excellent HTML5 video player)

          The writing is on the wall. Using flash for video playback or web elements that are basic is silly. Expect developers to begin to move away from flash heavy websites as HTML5 playback builds steam.

    2. Adam says:

      @Roberto Avanzi : Unfortunately the streamlined kernel performance came at the cost of instability. For example incorrect lock handling is causing 30-second “beachball” hangs for many users. There’s no easy way to know how widespread these problems are statistically, but clearly their prevalance is greater in 10.6 than 10.5 or 10.4. Unless they rise above a few percentage points Apple probably keeps its resources on feature and performance improvements rather than stability though.

    Leave Your Comment