Last Week on My Mac: What the eye does see, Apple doesn’t grieve over

The release of the macOS Sierra 10.12.4 update last week marks the midpoint of Sierra’s working life. Six months ago we were still grappling with its PDF problems shortly after first release; in six months time we will be discovering the issues brought by 10.13.

So far – a few issues excepted – Sierra has been one of Apple’s better major versions of macOS. For most users, its kernel has proved as robust as any before, and a big improvement on El Capitan, which seemed to panic for the sheer hell of it. Although some third-party developers have had to sweat blood to work around and through problems in Sierra, I’ve been surprised at some of the really old apps which still run sweetly.

Apple’s security team has been busy keeping pace with a succession of threats and some worrying vulnerabilities, but I don’t get the feeling that they are just papering over cracks. The increasing demands made on installers, for example, have seldom stopped me from running what I want to, and have steadily made macOS more secure.

At this mid-term point, my only two concerns are that Sierra doesn’t seem to be doing a whole lot new, compared with El Capitan, apart from not panicking every few days, and that there are still plenty of unfixed bugs.

Novelty is always difficult to judge. When you’ve been using an operating system daily, often very heavily, for over sixteen years, you tend to get slick at using more traditional ways rather than using newer tricks. I seldom use Continuity features, for example, because I instinctively move documents around by sharing and iCloud, when I need to.

Bugs are another matter, though. Sierra’s stability has undoubtedly benefited from automated problem reporting to Apple, which has enabled the engineers to devote their attention to events which generate crash and spin reports. Most of the bugs which stand on the list which I publish here do not result in panics, crashes, or even spins: only the user can see what is going wrong.

Problems becoming apparent in 10.12.4 with its handling of multiple displays are good examples of this. Unless large numbers of users flood Apple’s support services with reports of the issues, the macOS software engineers are unlikely to see this as a common or severe problem. Which brings me on to consider Apple’s systems for obtaining user feedback.

Compared with its automated gathering and analysis of engineering faults in macOS, Apple is remarkably deaf to problems which only the user can perceive. When you have updated to 10.12.4 and discovered that your iMac’s display stays asleep when external displays have awakened correctly, your Mac doesn’t send a report to Apple. Phone Apple support and it might be one of the few reports which it passes back up the line. But most of us just want to know how to solve or work around the problem.

A few might ask in Apple’s user support forums, where their issues might get noticed by Apple support staff. But, as ordinary users, that’s about as close as they can get to reporting a bug.

Developers – why only developers? Don’t other users matter? – have access to Apple’s Bug Reporter, a fairly buggy web reporting system which does nothing to encourage the reporting of bugs. I won’t dwell on its many flaws here, and cannot be open about it as its exchanges are often covered by NDA. Suffice it to say that reporting a bug using Bug Reporter often takes a lot of time and effort, but provides no return at all to the person making the report.

A developer who discovers a bug affecting one of their products understandably has an interest in reporting the bug, so that Apple can fix it. But what of a developer who discovers a potentially quite serious bug which affects general macOS function, but not their own product? Why should they take the time and trouble to enter it into Bug Reporter and provide all the supporting information which enables Apple to discover what is wrong, and eventually to fix it?

So, macOS steadily accumulates bugs which don’t get reported automatically, seldom get passed through user Support, and don’t affect the products of developers. And they just linger – like the Finder column width bug, for years, through successive releases of macOS.

I think that it is important that Apple knows about these bugs, and fixes them. They may not stop any user from doing anything particularly critical, but their presence is a constant reminder that, in some respects, macOS is still pretty shoddy. Which is a great shame, because I think that Apple and its software engineers have done a much better job than might appear.