I’ve used iOS apps since the App Store first opened in 2008. As a software developer, the market fascinated me. There was the initial flurry of simple apps, followed by more sophisticated apps. Then the iPad came along, providing a bigger screen for even more powerful apps. The future for apps seemed bright.
Today things have become vastly more complicated: multiple devices with different screen sizes and hardware capabilities, different operating system versions, and many more software APIs. (And that’s just iOS.)
Along the way, we’ve seen changes in business models. At first, most apps charged a fee up front. Then, some apps explored the ad model, while others had separate free “light” and paid “pro” versions. Later, when Apple added in-app purchases, the freemium model became commonplace, with the app being free to download but certain features requiring payment.
The gold rush of those early days — complete with stories of lone developers becoming millionaires overnight — quickly dried up as developers raced to the bottom to see who could charge the least. Today, more money is being made on the App Store, but the lion’s share of that money goes to a small group of large software companies.
Throughout its nearly ten years of existence, the App Store has had a turbulent upbringing. While the App Store may be a senior citizen in Internet time, as a marketplace, it’s barely out of diapers. But we’ve now reached a point where I believe the App Store will either morph into something genuinely useful or fade away as a fad.
I don’t mean that the App Store itself will go away — it won’t — but it could disappear as a business opportunity for most developers. In this dystopian future, the only profitable apps left will be a handful of entertainment apps by huge companies and “business essential” apps, such as those made by banks or news organizations for their customers.
The looming threat that I see is abandoned apps. They have always been cluttering the edges of the App Store to an extent, but the number of abandoned apps has grown lately for three reasons:
- The age of the App Store is such that even many wildly popular and successful apps have reached their natural end of life. It’s rare even in the desktop world for an app to exist for more than decade — technology just changes too much for many programs to stay relevant. Mobile apps live fast and die young.
- Apple recently began deleting apps that developers haven’t updated in years, under the assumption that they aren’t being supported.
While Apple has required that apps be compiled for 64-bit for over a year, old 32-bit apps won’t even launch in iOS 11 (see “Apple to Deprecate 32-bit iOS Apps,” 15 May 2017).
Individually, none of these factors would be cause for undue alarm. But bringing all three together could result in a catastrophic tsunami for smaller developers.
Good Intentions, Grave Consequences — Apple’s intentions are good. Customers downloading apps that haven’t been updated in years is bound to create a poor experience, and lack of developer support undoubtedly generates complaints to Apple.
Getting rid of 32-bit code is also sensible: it reduces app sizes, iOS can drop old APIs and 32-bit-only code, and everything new runs smoother and better. It’s also a good way to “encourage” customers to upgrade to more recent hardware (older devices are 32-bit-only and will not run iOS 11).
However, Apple’s solutions to these issues have serious problems.
The biggest is that, starting with iOS 9, performing a backup with iTunes no longer copies apps to your computer. To restore an app, you must redownload it from the App Store. But if Apple has removed the app for being too old or not 64-bit, the app is gone — there’s no way to download it again!
On the Mac, if a developer abandons an app you rely on, you can easily make backup copies and reinstall it if needed. If an app won’t run on a new version of macOS, you can theoretically boot from an older version or run the app in a virtual machine. Worst case, you can usually find a way to at least migrate your data to another app.
In iOS, the situation is different. Because Apple exercises total control over which apps are allowed to run and how you get and install them, there is no way to get abandoned apps to work (short of jailbreaking, which introduces its own set of non-trivial problems).
And because iOS doesn’t give users access to the file system, and apps themselves are sandboxed (meaning that one app can’t access another app’s data), if you have data in an abandoned app, that data is most likely inaccessible.
To get your data out assumes:
- That you still have the app and it still works
- That it provides an export feature
- That the format of the exported data is editable or useable
- That you have a replacement app that can import the data
Even if all those assumptions turn out to be true, exporting is rarely simple.
For example, I have an old iOS drawing app with about 50 drawings I made in it over the years. To get those out, I have to open each drawing and email it to myself, setting several export options each time. It will take me hours to export all my pictures, and I’ll lose aspects of the data in the process, such as text no longer being editable, layers being merged, or parts of the drawing being rendered differently in another app.
A different type of problem is that many iOS apps don’t involve data creation but instead run as a service, downloading data from the cloud. A perfect example of this is the Check The Weather app, which requires an Internet service to show you the current and projected weather in your area. Unfortunately, Check The Weather’s developer recently decided to shut down the weather network as it was costing too much. Thus, an app that I paid for and worked great for me for years, suddenly stopped working, with no warning.
Obviously, it’s not inherently Apple’s fault that Check The Weather’s developer couldn’t afford to keep paying for the weather network. Indirectly, though, Apple’s policies play a role, most notably by promising users that all app updates will always be free. When developers can’t charge for updates, that eliminates a common source of revenue that could fund ongoing development and operations. Plus, the App Store doesn’t provide developers any way to contact their customers to alert them to untoward situations or alternative apps.
I found out about Check The Weather when I contacted the developer to see why it had stopped working. He pointed out that he’d written a blog post on his Web site months earlier. How was I supposed to know that? I have nearly 600 apps on my iPad: I’m supposed to check all those developers’ blogs?
A Professional Environment? — Apple likes to promote the idea of iOS as being capable of professional work. In particular, Apple would like us to see the iPad Pro as a laptop replacement for many workers. For the purposes of argument (if only that), let’s assume that Apple’s contention is true.
What kind of professional environment allows apps to vanish with no way to reinstall them and potentially loses data you’ve accumulated for years?
For a casual home user, losing data to an abandoned app may be no more than an annoyance, but for a business or professional, it could be hugely damaging.
Businesses tend to be slow to change (don’t fix what’s not broken), especially if there are migration issues. Just look at how many companies still use Windows XP!
Apple’s policy, which encourages abandoned apps, is going to cause a lot of businesses to delay or avoid upgrading to iOS 11 for fear that they’ll lose access to apps and data.
As ever more software falls by the wayside this year with Apple’s forced march to 64-bit, more people will suffer the consequences. For many people, that may not be much of an issue, but for others, it could be a serious problem.
Looking at my iPad and iPhone, I currently see 101 and 58 obsolete 32-bit apps respectively. That’s about 25 percent of all the apps on my devices!
(To see the above list on your iOS device running at least iOS 10.3.1, go to Settings > General > About > Applications.)
Yes, a number of these are little games I haven’t played in years, and some are apps that have new versions as replacements (either I deliberately didn’t upgrade or didn’t know there was a new separate version).
A few of these apps were abandoned years ago, such as Zite and Bento. But if an abandoned app still works for me, shouldn’t it be my choice to stop using it or not?
Some of these apps may get new versions before iOS 11’s mandatory 64-bit cutoff, but short of checking with each developer, I have no way of knowing until that happens.
More problematic is the fact that roughly a quarter of my obsolete apps contain data. The list includes lots of writing and drawing apps, six photo editing apps, several news apps, and a number of “content” apps (language databases, apps with books in them, and so on).
These aren’t rare or uncommon apps, either. Many are from major developers or big businesses. I see apps from Fox, HGTV, Sci-Fi Channel (I refuse to use their new inane spelling), Adobe, and many others.
I have an app called Martha Stewart’s CraftStudio that I use for making my own greeting cards. Martha abandoned it a while back (I like to imagine her using Xcode), though the version I have still works fine. Unfortunately, when I upgrade to iOS 11, it will die and I’ll lose all the greeting card designs I’ve made.
What’s most disturbing is that I paid for many of these apps. Although I didn’t pay that much for any given app, in aggregate I spent hundreds of dollars on apps that I anticipated would have longer lifespans.
Worst of all, some of my obsolete apps are linked to hardware. For instance, years ago I bought a toy car that’s controlled via an app on my iPhone. That app is on my obsolete list. When it goes, the car is useless. Good thing it’s not a $1500 drone!
Another 32-bit app lets me use tiny flash memory “hard drives” wirelessly for additional iPad/iPhone storage. (When I travel, I store movies on these for watching on the plane.) Once I update to iOS 11, I won’t be able to use those drives anymore, and they were quite a bit more expensive than the average app. Not good.
I’ve heard other horror stories of hardware that stopped working when its app was abandoned. But actually experiencing it is making me think twice about buying any device that requires an app, whether it’s a security camera, kitchen gadget, toy, or whatnot.
All this means I’m going to be a lot more cautious about spending money on apps in the future. When the App Store first launched, I was accustomed to desktop prices, and as a programmer myself, I was happy to throw a few bucks to developers for their work. Now that I’m seeing so many developers abandoning apps and Apple rendering other apps artificially obsolete, I doubt I’ll be buying much from the App Store.
If my memory is accurate, I used to spend $10 to $25 per month on apps. In 2016 that probably dropped to an average of less than $5. For 2017, it will be even less.
Mac Apps May Be Next — While getting rid of 32-bit cruft on iOS makes some sense, resources aren’t as constrained on the Mac side. Plus, most apps aren’t from the Mac App Store, so Apple has less control over what is installed.
Despite that, at Apple’s WWDC developer conference in June, Apple made an announcement about the next version of macOS, High Sierra, that makes it sound like 32-bit support on the Mac may be disappearing soon.
Apple’s statement is clear about the Mac App Store requiring 64-bit, but it’s more vague about apps distributed outside the Mac App Store. There Apple “highly recommends” developers go 64-bit and concludes with: “macOS High Sierra will be the last macOS release to support 32-bit apps without compromise.”
I interpret that to mean that some apps will work and others may have problems, though I have no clue what such problems would be. If “without compromise” means old 32-bit apps run slower, similar to PowerPC emulation during the switch to Intel processors, that’s very different from not being able to run them at all.
Whatever version of macOS succeeds High Sierra is over a year away, however, so there’s time to figure out a solution. For most apps, making them 64-bit is just a matter of recompiling, but that doesn’t help for ancient apps that aren’t still under development.
What’s the Solution? — So far I’ve been focusing on the problems, but what about solutions?
While I realize that Apple has an interest in culling unsupported apps from the App Store, and I agree that getting rid of 32-bit ones is a valid engineering goal, the way the company is going about it is draconian.
Why not put the old apps in a special “old” section of the App Store? Mark them with an Obsolete tag. Set the default App Store search to ignore those apps unless the user selects an “Include obsolete apps in search results” option. Provide a warning upon download that an app is obsolete and unsupported, but let me download and use it if I want.
While it seems unlikely that Apple will ever let developers communicate directly with all their customers through the App Store, Apple could contact app users on behalf of the developers. It would go a long way if Apple could provide users with information about obsolete apps, such as a link to new versions, instructions on how to export data, or suggestions of replacement apps. On reflection, I can’t see Apple ever doing that either.
But Apple could notify us when apps we buy have been removed from the App Store. This would provide a warning that the developer may be abandoning the app and would encourage users to check for more information.
(For instance, I’ve noticed a few apps disappearing when I migrated to a new iPhone. It was a shock to learn those apps were no longer on the App Store and therefore couldn’t be restored. If I’d known, I could have prepared, but I wasn’t warned.)
Apple should also do more to encourage the financial success of developers other than the big boys: better discovery of apps, allowing paid app upgrades, removal of “clone” and garbage apps, and more. If apps are being abandoned because they aren’t making money, anything Apple can do to address that problem would be a help.
To be fair, Apple has been making improvements to the App Store, notably a faster review process. And in a change to a long-standing policy that hurt developers and customers alike, Apple finally allows developers to respond to customer reviews. iOS 11 will bring a new look to the App Store; we’ll see if that improves app discovery or improves sales in any way, or if it’s just rearranging the deck chairs on the Titanic.
What You Can Do — How you respond to Apple kicking apps off the back of the train is up to you, of course. But here’s my approach — feel free to copy or modify it to suit your needs.
While I think iOS is highly capable and could be a person’s only computer, I’ve already been hit so many times by abandoned apps that I’ve become wary. I no longer think of iOS as a “professional” environment.
Since apps can disappear at any moment, I’m not going to devote much money to buying them nor will I invest significant time and energy into learning and using an app. And yes, I realize that this behavior will result in reduced sales and lowered developer interest in creating “professional” apps.
When evaluating iOS apps that I might use for real work, I’ll avoid those with proprietary data storage. I’ll actively seek out apps that let me store data in the cloud — preferably Dropbox — and apps that use standard data formats.
Before committing to an app, I’ll always check to see that it has export capabilities — and I’ll immediately test those features to make sure they actually work.
My next project is to start going through my list of abandoned apps and seek out replacements for the critical ones. For many, though, I plan to look for a Mac replacement, not just another iOS app. After losing so many favorite iOS apps, I just don’t trust the App Store anymore.
(A version of this article originally appeared in xDev Magazine in May 2017.)