On reliable technology

This weekend, I had to purchase a new iPhone. My iPhone 11 — bought in October 2019 — was becoming unreliable. Its touch screen was acting up, either producing ghost touches or becoming unresponsive in some areas, which could be frustrating when the affected areas were covering the keyboard. In the past couple of weeks, I even had to use a Bluetooth mouse and a Bluetooth keyboard to unlock my phone a few times.

After four and a half years, getting a new iPhone seems expected. I would have preferred to get an updated-for-2024 version of the iPhone SE, but surprisingly, no new iPhones SE were released. Therefore, as fixing the screen was not a viable option, I had to spend 2 or 3 hundred euros more to get the iPhone 15, which is excessive for my modest needs, but future-proof enough as I plan to use it for another four and a half years, at least.

During this process, I thought a lot about what makes the iPhone the iPhone, what makes it my obvious first choice when it comes to smartphones, and what I was expecting the most from the new device. My first — and obvious — answer is the Apple ecosystem: AirPods, iMessage, Apple Photos, iCloud Drive, continuity, available apps, Safari, etc. But disregarding the appeal of that infamous walled garden, I believe that reliability is what makes the iPhone so good and what made my broken iPhone 11 so frustrating to use in its final weeks.

If you ask people on the street what they think is the best quality of the iPhone, I’m pretty sure you’ll get answers like ease of use, design, the choice and quality of apps, camera performance, etc. I don’t think anyone would say reliability.

Using something that is dependable and performs as expected every time you use it is what I want from a phone, a tool, a car, an app, and many other things. Of course, I love good looks, great ergonomics, durable objects, and a satisfying sense of perceived value. But I’ll take “reliable” over any of these qualities.

I believe that this is a quality we all unconsciously prefer. Think about it: which pen do you use when you need to jot something down? Which knife do you select from the kitchen drawer to slice something up? We all tend to use the same 2 or 3 pens or knives for most things, while other pens and knives in the drawer gather dust. The appearance and value of these 2 or 3 reliable pens or knives won’t matter much. When picking them up, their level of performance will obviously matter, but only if there is confidence in the delivery of that performance, only if the objects are known to behave as expected.

This is why Apple Maps had such a rocky start back in 2012. The app looked great from the start, the UI was — and still is — better than the UI of Google Maps, it had interesting features from the beginning, but it was not reliable. Trusting it to go somewhere was a leap of faith, and I’m talking about the Indiana Jones and the Last Crusade kind. A decade later, things are much better, and I now find Apple Maps to be very, very good for directions and navigation. It has become a reliable app.

But to be able to notice that something has gotten better, one needs to try it again. If you know that a knife is not great in the back of the drawer, you’d never pick it up again after the first disappointing experience. Someone would need to tell you “I’ve sharpened your knives, including the shitty ones” and you’d need to be willing to give that knife a second chance.

This is why Siri is so frustrating. Siri has had many occasions over the year to prove its detractors wrong. Because it’s basically the only knife in the drawer, Siri had plenty of second chances. Its problems are not really about the things it can do or not, not about how the synthesised voices sound: the problem is that Siri is unreliable. Sometimes it works, sometimes it doesn’t. Sometimes it will do exactly what you want, sometimes it will ask you to unlock your phone first. Sometimes it will clearly hallucinate what you just asked, and sometimes it will just do nothing.

Reliable tech is tech one will want to use repeatedly. If the tech is great but only works from time to time, people will end up using reluctantly. It doesn’t matter how great the tech is if it can’t be great all the time.

This topic reminds me of the “best versus good enough” idea. In a post on this blog, I quoted Robert Watson-Watt:

Give them the third-best to go on with; the second-best comes too late, the best never comes.

Following this analogy, the iPhone may not be the best phone. If I take the photos it produces as an example, the iPhone may not produce the best photos captured by a smartphone camera. They may not even be the second-best. But the iPhone will consistently produce the third-best photo. Every time. It’s a reliable camera, and that’s why it may arguably be the best camera phone.

In the end, it’s not about the quality of the photos: it’s about the reliability of the camera. And I feel that it’s kind of the same for other features of the phone and how I use it.

As I’m writing this in mid-April, I obviously have to mention the Humane AI Pin, as it fits so well into the topic of quality, expectations, and reliability.

As I expected, the device looks like a complete disaster. After reading and watching a few reviews, I found that the lack of reliability was a recurring theme: battery life? Unreliable. AI? Unreliable. Laser screen thing? Unreliable.

For something meant to be worn every day and eventually replace the smartphone in the daily life of its users, this lack of reliability is a significant issue. How can someone embrace the Humane AI Pin’s philosophy of unintrusive technology if the technology does not perform as expected? Unreliability certainly sounds like an intrusive characteristic of a device, forcing the user to constantly think about how to handle and manage it.

Innovation is great, but without reliability, it scarcely means anything. The more I think about reliability in software and hardware, the more I think it’s the most important quality for the tech I use, as it should be for the proper tools they are.