Apple’s declining software quality

Accelerated growth and capex.

I once saw a very nice chart about the number of users for a great, cool start-up everyone knows about. The CEO illustrated: "You see that inflexion point on the release day? That's what happens for every release. If we postpone it by 3 months, we not only lose 3 months of new customers (painting the 3-months area between lower and higher sales). We also miss on the offset." and he painted the unlimited area between the lower sales and higher sales. And he talked about the recurrence of the inflexion for each release, and he talked about the immobilized capital inventory.

Declining? I didn't know software quality could go any lower than 0.

Declining? I didn't realize quality could go lower than 0.

In my time using Apple Mac gear since the Mac 128K days, Apple has put a moratorium on new features and focused exclusively upon stability and bug fixes for an OS release only twice in my memory; once under Mac OS, and once under OS X. Software quality at the polish and attention to detail level is a Sisyphean entropic struggle for everyone, though. It's just that Apple's enormously successful "It just works" marketing message from 20006-2009 continues to doggedly trot out with its fans today, so Apple gets tarred with the same bar to clear today despite an enormously more complex operating environment since then.

It doesn't help all the review channels (magazines, newspapers, blogs, vblogs, videos, etc.) focus their attention exclusively upon new features in a new release. You need someone as persuasive as Jobs to pull off convincing those channels to review refinements to add stability as new features. The people Apple has put forward to hopefully capture that role hasn't resonated with the public yet.

But software QA should never be an episodic, herculean, release-bound effort. There are ways to market it positively, but I'd always rather use the finite PR time surrounding a release marketing and selling the quantum leaps that set me apart from the competition, not the incremental steps.

In the age of Big Data, instrumented apps and online-inline updates delivery, a possible initial pass at identifying problem areas to shore up that is actually hitting people in their daily workflows is to simply track on the search engines the popularity of complaints about major bundled applications, subsystems and components. An OS X release should not, for example, let the first-page results for Contacts be about entries disappearing [1], three updates into its lifecycle.

There is still plenty of space for innovation here, I think. There are auto crash reports, but no systematic, automated means for a publisher like Apple to send targeted offers to users sending crash reports that pass a specified threshhold (number of crashes, kind of hardware, stack trace pattern, etc.) to update their app to an instrumented one that give developers not just the results of instrumenting the one or a handful of users who happen to persist enough through the now-familiar troubleshooting dance, but hundreds or even thousands of users at the same time, opening up opportunities to automatically search for commonalities and assist with the manual troubleshooting.

[1] https://www.google.com/search?q=mac+contacts+%22el+capitan%2...

Along with a few other issues with OS X updates over the last month or so, iTunes continually downloads previous purchases that have already been downloaded like 5 times.

This is an incredibly stupid bug and it's eating up my disk space (until I clean it up of course). I am literally talking about 10 entire albums at a time every time I make a purchase. It's been going on for the last year. I'm not sure why I still use iTunes anymore to be quite honest. I buy my songs through iTunes and then they get uploaded to google music, and I think I've just been going about it as a matter of habit all this time.