Saturday, August 15, 2015

Finding iPad’s Future

Neil Cybart (via John Gruber):

A product that carries so much brand relevancy that it still represents the entire tablet market now finds itself the leader of a category that has lost all momentum as other product categories marginalize the tablet form factor. Although Apple is still selling more than 10 million iPads per quarter, there is something about the iPad that just doesn’t sit right with me. We have gotten to the point that the status quo will likely lead to the iPad and the modern-day tablet becoming irrelevant over time. A new direction for iPad is needed based on a fundamental rethink of tablet computing.


A quick look at iPad and tablet shipment data would show that things have gotten bad in recent quarters. However, in reality, things are much worse than quarterly shipment data would suggest. The seasonality found in the tablet segment makes it difficult to see these long-term problems. A much better way at understanding what has been taking place is to look at the year-over-year change in shipments on a trailing 12-month (TTM) basis, highlighted in Exhibit 1. This smoothing effect highlights that the iPad and tablet have been on the decline for years and things continue to worsen with the overall tablet market hitting negative territory for the first time. All momentum has been lost.


Many didn’t see it, but tablets were quickly turning into content consumption devices where price was a leading purchase decision.

We now find ourselves with a tablet market where Apple and Samsung are losing share to "Others," which is represented by dozens of firms selling mostly generic tablets used to consume media, depicted in Exhibit 2.


The iPad market is in trouble and if there are no changes made to the lineup, Peak iPad is on the table. Peak iPad is a simple concept driven by the belief that underlining structural changes to the tablet market would result in the iPad losing most of its value propositions, leading to a permeant decline in sales. For example, Peak iPod is alive and well as even though Apple is still selling iPods, the product category will never reach record quarterly sales. Meanwhile, while some argued that we had seen Peak Mac, we instead were just in a sales slump that quickly reversed itself with a revamped product line. The Mac’s value propositions were still alive and well. In a world where smartphones are getting larger and laptops are getting smaller, the Peak iPad theory is starting to look more likely as time goes on.

Lukas Mathis:

The PC market relies on upgrade sales. The plastic spoon market relies on upgrade sales. The pants market relies on upgrade sales. But a device as young as the iPad should not be relying on upgrade sales to this degree. If Apple thinks that the iPad’s sales are falling because of a long upgrade cycle, the implication is that the iPad has already reached a large portion of all people it’s ever going to reach.


Better hardware would help, but I think it’s very important to acknowledge that the thing standing in the way of productive work on the iPad is not its hardware. It’s iOS.

iOS is a cumbersome system for even reasonably complex productive tasks.


Fixing this problem does not mean «giving access to the file system.» When I say that Apple needs to fix document management, people sometimes assume that I’m saying that they should bring something like the Finder to iOS. I’m not. The Finder approach to file management is broken. It was designed for a time when people had a tiny number of apps, and almost no storage space. That time doesn’t exist anymore, and neither should the Finder.

I think this last part is wrong. When the iPad was released, the jury was still out. After eight years of iOS, we’ve seen that no one has figured out a good alternative to the filesystem. iCloud and iOS seem to be converging on a design that is less capable and possibly more complicated for non-basic uses. Meanwhile, the Mac has gained iCloud Drive, sandboxing, and rootless, which offer many of iOS’s simplicity and safety advantages without (in theory) compromising functionality.

Of course, the Finder does have problems. But something like the Finder—as an icon on the home screen, rather than the home screen itself, like on the Mac—would straightforwardly solve a lot of iOS’s current document management problems. And, tucked away as an icon that many people wouldn’t need, it wouldn’t cause much trouble.

Update (2015-08-20): Jason Snell:

I think we’re all down on the iPad because we got too excited about it to begin with, and the hangover hasn’t faded yet.


I suspect the iPad’s initial sales were so good that Apple was fooled into thinking it didn’t need to improve the iPad—which is a mistake that the company may be paying for now. But Apple sure seems to be paying attention to it now!

Update (2015-08-31): See also Accidental Tech Podcast #132.

19 Comments RSS · Twitter

"After eight years of iOS, we’ve seen that no one has figured out a good alternative to the filesystem."

I haven't seen any examples of anyone really trying.

I *am* seeing plenty of examples of people struggling with the Finder (or other file system-driven document management systems), so if that's the solution to iOS's document management problem, it's a terrible one.

Of course, if I can choose between having a Finder-like App on iOS, and not having any file management, I'd rather have the Finder. But the idea that it is impossible to come up with anything better than the Finder seems implausible.

If we were in a pre-Ford era, Google would be the company selling you a horse buggy. Apple would be the company selling you just a buggy, because horses are kind of hard to maintain, so let's get rid of them. But now I have to push the buggy to get it to go anywhere, so that's not great, either.

So you could ask Apple to give you back the horse. But if they do, you also get all of your horse problem back. What you *really* want is not for Apple to give you back your horse, but to give you a car instead.

Assuming the Lukas comment here is from Lukas Mathis, whose blog I've been devouring via RSS for a while now, and while I agree with his basic contention that problems are baked deep in the OS, and that while a Finder would certainly help, it wouldn't solve all the problems.

I've also been noting his case that while Surface solves productivity issues far better than the iPad, it still leaves a world to be desired as productivity tool.

At the end of the day, I've always though this is really just a form factor issue, at base. While there is certainly long-hanging fruit to be fixed on both platforms, I just can't imagine solutions that will make either platform a decent productivity tool except for relatively edge-case scenarios. There really is a simple case to be made that the laptop form factor has immense advantages that simply can't be overcome, or even really approached, by tablets.

So the Surface will improve as a productivity device, and the iPad will too if Cupertino ever bothered to deal with it. But, still, except for edge-cases, my theory is that it's always going to be a massively inferior, stop-gap productivity solution. Form factor is destiny, to a not-insignificant degree...

@Lukas I’m not saying it’s impossible, just that no one seems to have any idea how to do it. So I think it’s premature to say that horses should not exist. It also may turn out that document management has a certain amount of essential complexity, so that there may exist a solution that’s better than the Finder, but not markedly so.

@Chucky I tend to agree with you about the form factor. Even with better software, I wouldn’t see myself spending a lot more time with a tablet.

"I've been devouring via RSS for a while now"

That's me. Thank you!

"There really is a simple case to be made that the laptop form factor has immense advantages that simply can't be overcome, or even really approached, by tablets."

This is true in some sense: hardware keyboards are better for text input, a mouse is better for precise interactions, the hinged screen design of laptops work well in many situations where tablets don't. However, here's how I use my Surface: if I'm in a meeting, taking notes, or I have an idea at night and need to write it down so I can go back to sleep, or if I want to draw, I use it as a tablet. But when I go to my desk, I put it into the dock, and now it *does* have a keyboard. And when I'm in a train and want to respond to some email, I attach the keyboard cover, and now it *does* have a physical keyboard.

Tablets can change based on the current need, and offer the correct form for whatever you're currently doing.

Tablets can change based on the current need, and offer the correct form for whatever you're currently doing.

Is that a lot more true for Surface? Because using a keyboard with my iPad feels like the worst of both worlds.

"I’m not saying it’s impossible, just that no one seems to have any idea how to do it."

I don't know if it's true that nobody has any ideas. I think the problem is rather that nobody is in a position where the incentives to fix this problem line up correctly.

On the desktop, people are very change-averse, so anything Apple or Microsoft do with file management is received with extreme pushback.

On mobile, both Apple and Google are quite content with just ignoring the problem, at least for now. I'm not sure if either company really accepts that this *is* a problem on mobile. On phones, it's probably not a huge issue, and tablets barely seem to matter to either company.

Third-party developers have zero incentives to fix the problem, because the people who would most profit from a solution are the people least likely to use a third-party tool for file management. People who *do* use third-party tools for file management are power users; hence, third-party file management tools add more features (Path Finder) or offer specialized file management for specific use cases (Eagle Filer). They don't fix fundamental file management problems in a generic manner. Also, many of the required fixes would need OS-level support.

As to what ideas for solutions might be, think about how you manage files. Here are some of the things I do:

I sort file lists by date, because I have a rough idea of when I created or edited a file.

I know who sent me a file, so I search for email from that person, and re-download the file.

I know what application I created a file in, so I search for all documents from a specific application.

I know what kind of topic a document is about, so I search for keywords I expect to be in the document.

I search by file type, because I know what kind of thing a file is (e.g. an image).

It would be easy to extend this list, maybe also by observing what normal people do to retrieve files. Now, if you think about these things — time, people, applications, topics, keywords, file types — these are all things that your computer knows about. Your Mac knows who sent you a file. Your Mac can analyze the contents of a file to figure out what it is about. Your Mac knows when you created or edited a file. Your Mac knows what app you used, and so on.

So why does it not surface that data, and allow you to easily and visually browse your documents based on time, based on people, based on applications, topics, etc.?

(As an aside, the application I work for in my day job has to solve many of the same problems as a file system. It manages tens of thousands of file-like entities, and different people have access to the same repositories, so one person's order might be another person's mess. The way we do it is by using metadata, and by using context, to auto-organize things based on what we know about the relationship between them, and what we know about what the user is doing.)

Finally, I'm not saying that horses shouldn't exist. Horses still exist, and for some use cases, they're better than cars. But these use cases are limited, and the people who use horses for work are trained professionals. I think the analogy holds :-)

"Is that a lot more true for Surface? Because using a keyboard with my iPad feels like the worst of both worlds."

I think the reason why it doesn't work well with the iPad is that you can't *control* it with the keyboard. All you can do is enter text (well, almost all). I suspect that the result of this is that people used to notebooks somehow mentally switch to "notebook usage mode" when they use an iPad with a keyboard, when really, it's still an iPad, it's just easier to enter text now. That makes the interactions feel weird.

On the Surface, you are free to mentally switch to "notebook usage mode" if you want to. Attaching a keyboard means you can use it like a notebook — or use it like a tablet, but type text using the keyboard. Or do both, and switch between mouse and touchscreen depending on what feels better at the moment.

Also, the way the Surface's hinge works, it pushes against the center of the screen, so touching it in "laptop mode" feels stable. Touching an iPad in a dock or docking keyboard doesn't feel good, because most docks don't offer much resistance or stability. It's just a small thing, but I think it makes a difference.

One area where the iPad still has an advantage, at least for me, as a productivity device over a MacBook (and I love my 11-inch Air), is built-in cellular networking. For example, while in Colorado for a week, with little to no wifi access, I had two occasions where I needed to type out a lengthy email and send. I had two bars worth of 4G in our cabin, and it managed the text just fine.

I can certainly see a future for myself where, if MacBooks gained built-in cellular networking, I wouldn't use an iPad any more, except for maybe Comixology. 8^)

My kids, however, well, their primary computer is an iPad, so it will be interesting to see how that plays out, too.

PS: I apologize for the long comments, and for the space between the words "Eagle" and "Filer".

"I apologize for the long comments, and for the space between the words "Eagle" and "Filer"

Not my blog, of course, but IMHO, long worthwhile comments are fine, but posting "Eagle Filer" should get you shipped off to the Gulag.

"posting "Eagle Filer" should get you shipped off to the Gulag"

I humbly accept that punishment, as it is fair, justified, and fitting the crime. I'll use Apple Maps to find my way there.

@Lukas I appreciate your long comments, and apology accepted. I’m left not quite understanding your position, though, since you started off by saying “That time doesn’t exist anymore, and neither should the Finder,” but now you seem to be retracting that. Are you agreeing with my idea that (like horses today) the Finder should exist but that everyone shouldn’t need to use it? Or are you drawing a finer distinction?

@chris For me, that advantage no longer holds because my iPhone’s AT&T account now includes free tethering and more bandwidth than I am ever likely to use. On our last road trip, we had four nights in a row of hotel Wi-Fi that didn’t work, and I had forgotten my Thunderbolt-to-Ethernet adapter, but it was easy for my MacBook Pro to just use the phone’s connection.

"I humbly accept that punishment, as it is fair, justified, and fitting the crime. I'll use Apple Maps to find my way there."

As we both know, that's cheating. If you use Apple Maps to get to the Gulag, you'll probably end up in the Caribbean or Tahiti. Use Google Maps, so you'll correctly end up in Siberia.

@Michael: I think the Finder, as it is today, should no longer exist. It should be replaced with a higher-level document management system that is consistent with modern computer usage, and modern computing environments (e.g. it should seamlessly tie in with cloud services like Dropbox, it should use metadata to organize documents automatically, it should allow users to access files in a way that makes sense to them, it should not just show the file system — that's an implementation detail).

I don't equate "access to the file system" with "the Finder". So while I think that the Finder needs to go, I think access to the file system should still be available to the people who want it. That doesn't mean having a Finder. It can mean leaving this open for third-parties (e.g. Path Finder), or it can mean having a file system power user app that most people will never see or use (similar to Terminal). Positioning this explicitly as a power user application means that it will no longer have to take an uncomfortable middle-ground between being accessible to everyone (because that's the first thing they see when they boot up their Mac), and providing power user features to those who know what they're doing.

(Conceptually, it actually kind of used to be like this before Mac OS X. While the Finder just showed the file system, the file system was much simpler back then, and the real complexity was hidden in files' resource forks. So normal people saw the simple document management tool, but power users could download ResEdit and access a deeper hierarchy that was hidden from regular users.)

@Chucky: Dang, I thought I'd found a loophole...

"My kids, however, well, their primary computer is an iPad, so it will be interesting to see how that plays out, too."

This! When it comes to the future of iPad I think this is the most interesting question. Right now iPad seems to have gone from bright to bleak future, but if we look at it long term things could be different. I've not conducted proper research but from what I've seen the iPad is the primary computer for kids these days. If my kids could wish for anything they would both wish for an iPad (and I wish I could afford them each an iPad).

What happens when this generation grow up? Will they continue to use tablets or will the large phones and traditional laptops make them abandon tablets?

*pops head*

While I think that Apple should do better than a “Finder for iOS”, it is important to recognise that this is a secondary matter compared to the sorry state of the iPad operating system and the fact Apple hasn't really proposed solutions to the matters of document filing, exchange between apps, or management in general in the first place (or if they have, they have done so way too slowly). It seems in particular that nobody has really tried to improve on the “Finder” model: Apple hasn't, obviously, and its competitors were too content to just slap a traditional filesystem view on their gizmos and be able to boast about providing something Apple doesn't provide. It doesn't necessarily mean something different is viable (I believe it would be, but that's only my belief), but so far we don't know.

Mind you, things will be improving with iOS 9, but their track record so far indicates that Apple hasn't really got a vision for this aspect of computing.

@Pierre That’s quite a bind. Apple doesn’t seem to have a vision, and they’ve locked the system down so that no one else can implement one, either.

[…] T being an actual car. There are not even rumors about improvements to the store. And, like the iPad filesystem, no one else seems to have a better idea, either. At this point, the problems have been common […]

@AdrianB makes a good point. See Ben Bajarin's iPad Pro review (, specifically the section "The Most Eye Opening Observation" where he writes about how his daughter and her school mates are using iPads.

Saying that computer type X is more productive than type Y depends on what you are producing. Making music or a movie involves very little use of an alphabetic keyboard. Force touch in conjunction with the stylus is much better than drawing with a mouse.

Leave a Comment