Next OS X/iOS Version Upgrades Should Concentrate On Quality And Stability — Not Buggy Feature-itis – The ‘Book Mystique

On stage at Apple’s World Wide Developers’ Conference in June 2009, Bertrand Serlet, the company’s Senior Vice President of Software Engineering at the time, announced that the forthcoming OS X version 10.6 Snow Leopard would feature “no new features,” — something Mr. Serlet suggested was “unprecedented” in the computing industry. In fact, Snow Leopard did have some new feature content, but Mr. Serlet’s point was that the main thrust of the version upgrade was to polish and refine the performance of OS X 10.5 Leopard with bugfixes and some mainly under-the-hood tweaking, rather than adding a raft of new features likely in need of refinement.

When Snow Leopard shipped ahead of schedule a couple of months later as a $29 upgrade to Leopard, it included “hundreds of refinements,” “new core technologies,” and out of the box support for Microsoft Exchange which allowed allows users to use Mac OS X Mail, Address Book and iCal to send and receive email, create and respond to meeting invitations, and search and manage contacts with global address lists. Apple affirmed that to create Snow Leopard, its software engineers had refined 90 percent of the more than 1,000 projects that comprise Mac OS X (which the OS was still called), an exercise that was successful in producing what is arguably the most solid and stable version of OS X ever — an achievement that still stands. Going on five years and and four full OS X version upgrades later, many Mac users whose aging machines are still capable of supporting 10.6 are still using it for production and productivity, with its rock-solid stability combining with it being the last OS X version to support Rosetta emulation for running “Carbon” software applications ported from native Power PC
to Intel.

That would include me. I have the factory install of OS X 10.9 Mavericks on my mid-2013 Haswell MacBook Air, and I’ve arrived at a sort of a working detente with Mavericks. I still haven’t forgiven Apple for ruining the Spaces feature in OS X 10.7 Lion and subsequent versions, but do I enjoy the speed and quietness of the MacBook Air. However for serious productivity work, my late-2008 aluminum MacBook running Snow Leopard is more often than not my go-to machine, partly because the user interface is slicker and more efficient, and partly because of several Carbon apps for which I have yet to find satisfactory Intel-Native substitutes. I can also boot that computer into OS X 10.8 Mountain Lion from a second hard drive partition, but find I don’t like it as well as Snow Leopard for production work.

So I agree with TekRevue’s Jim Tanous, who contends that what Mac users need isn’t more new features in the OS, but rather a Snow Leopard style “no new features” bugfix upgrade. Three months on, I still haven’t upgraded to OS X 10.10 Yosemite on the MacBook Air. I don’t love Mavericks, but at least it’s pretty stable, which I keep hearing Yosemite is not. Problems cited by Tanous echo what I’ve been reading on various boards, forums, and comments pages for the past three months; UI slowdowns and system freezes that require daily reboots to clear (shades of Windows in the old days) and Wi-Fi connectivity issues — all of which would make me apoplectic. Stability and dependability are among the very top qualities I value in a computer OS, and reportedly Yosemite still falls well short even after the several update patches released so far are applied. The long and short of it says Tanous, who has Yosemite installed on his 2013 Mac Pro and 2014 MacBook Pro, is the realization : “I no longer trust OS X. In fact, OS X is unusable in its current state.”

He hastens to clarify that by “unusable,” he doesn’t mean he can’t boot into Yosemite, which he says takes 8 to 10 seconds longer to boot than Mavericks on the same hardware, adding insult to injury. What help does mean is that for production work, “I can’t trust it. I’ve had too many crashes, too many freezes, too many reboots to rely on the operating system to get my work done in a timely and efficient manner, and for me, that’s all that really matters in the end.” I’ll second that contention vicariously.

Tanous observes that no new feature, technology, or interface tweak justifies diversion of engineering resources away from proper testing and quality control, and while he says he’s long used both Windows and OS X, heretofore he’s generally preferred OS X for typical workaday tasks like writing, research, and video editing, working with apps like Chrome, Word, and Photoshop. However he says that since upgrading (?) to Yosemite in October, he’s actually finding Windows 8.1 more enjoyable to use — an epiphany that occurred him during a stretch working at home on his gaming Windows PC due to an illness when he noticed that he wasn’t experiencing any of the frustrations or anxieties about work-destroying bugs that he’s been living with on Yosemite for the past three months. He finds that running Windows, those key productivity apps and the overall operating system run just fine, and it has yet to ever crash or freeze on him, while in Yosemite, the user experience is littered with bugs, slowdowns, and outright system lockups — flakiness that disappears when he boots back into Mavericks on the same hardware.

Not news inclining me to take the plunge and upgrade to Yosemite on a workhorse machine. Tanous thinks the problem is a consequence of Apple’s adopting a yearly release cycle for OS X updates beginning in 2012, when they released Mountain Lion just one year after Lion debuted, while Tiger, Leopard, and Snow Leopard had all enjoyed runs of 20 to 30 months apiece.

Former Apple executive Jean-Louis Gassée addresses this issue in his Monday Note blog this week, observing that the incidence of recent “software accidents” in both iOS and OS X raises questions about Apple’s management of its ongoing increase in R&D spending, saying that for the past six months or so, he’s become increasingly concerned about Apple software, citing specifically Yosemite’s “painful gestation” with damaged iWork apps, and the chaotic iOS 8 launch, quipping that the longtime and up to recently largely truthful Apple “It Just Works” brag is now more appropriately: “it just needs more work”.

Mr. Gasseé cites several other respected Apple watcher commentators who have made similar assessments, and notes that for himself, his default position on Apple software in OS X has changed from ‘probably good’ to ‘probably not OK’, observing that they seem more interested in pumping out volumes of upgrades, citing not only the operating system raggedyness, but also a range of specific angularities with Pages, iCloud, iTunes, laying the blame in part on Apple’s obsession with “dumbing down” features and functions in aid of OS X/iOS compatibility, which in Mr. Gassée’s estimation isn’t working, noting that iWork on iOS leaves much to be desired, and observing that”creating even a moderately complex document on an iPad is an unpleasant, frustrating experience.”
To say nothing of more bugginess in Pages, and with iTunes presenting an even sorrier spectacle than Apple’s productivity apps.

Back on the OS upgrade topic, for a more positive assessment of OS X 10.10 Yosemite, see this collection of observations by iMore editors:
http://www.imore.com/os-x-yosemite-review-3-months-later

However, even iMore’s managing editor Peter Cohen, who has been using Yosemite and likes it for his purposes, says that sometimes the perils of a major operating system upgrade aren’t worth the trouble unless you’re willing to accept the risk of being an early adopter, so this is an instance where he recommends doing as he says and not as he does. On a more encouraging note, Cohen relates that he’s been using the Yosemite 10.10.2 update beta for some time, and is optimistic that it may have finally squashed the WiFi dropout bug that’s plagued many early adopters. If that proves the case when 10.10.2 goes live, I may be persuaded roll the dice and upgrade.

Nevertheless, Jean-Louis Gassée cites a comment by Marco Arment earlier this month to the effect that we don’t need major OS releases every year and/or for each version upgrade to have a huge list of new features, and what we do need is for our our computers, smartphones, and tablets to work well as a first priority to enable us to enjoy new features released at a healthy, gradual, sustainable pace. That mirrors my own perspective on the topic. Ergo: bring on the bugfix patches, but what we really need is another Snow Leopard-like “no new features” OS X upgrade to fix Yosemite! and for that matter while they’re at it, a “no new features” iOS release to fix version 8.