Being one of the world’s highest valued companies means you can be brilliant at some things, completely useless at other things and have your head up your ass about most things.
- You don’t need to report unit sales numbers any longer, because you’re above all these piddly details.
- You can frame tax rulings as total political crap, and still value buildings worth $1,000,000,000 dollars as worth $200 for tax reasons.
- You can say you check all apps in the App Store, because otherwise you’d risk shipping apps that could ask you to put your finger on the Touch ID sensor and bamboozle you into verifying a purchase, and then still shipping them.
- You can refer to your stores as “town squares” and instruct our tech support professionals to never admit that something is broken, only to apologize that it being broken makes people upset.
- You can sell a five year old computer model, whose upgrade you started working on last year and didn’t show because “you can’t rush these things”.
- You can up the price for new models of everything you make without anyone talking back.
- You can stop putting in ports for items everyone owns and everyone needs to use regularly and then sell dongles, and then cart out Jony Ive to call it cleaner and simpler. What’s cleaner about needing convertors for everything?
This is a complicated question with a complicated answer. (If it makes things more expedient for you, feel free to just call me a shill, cult member and/or idiot.)
Picking a desktop PC platform right now is a classic case of picking your poison. Apple cares way less about their desktop platform for every passing day, but so apparently does the rest of the world. I use Windows 10 every day, and I wish I didn’t - if Microsoft was still interested in advancing their platform in the way they did between Windows Vista and Windows 7, I might switch tomorrow.
At the risk of disappearing up my own butt for a moment, it’s frighteningly clear that no one at Apple or Microsoft values the way macOS and Windows respectively has worked enough to not see turning parts of it into a tablet OS – irrespective of the fact that tablets have not gone beyond gimmicks for most desktop users – as progress.
The Linux crowd is what’s left, and they cook their own punch, so they seem like an obvious refuge. I’ve been giving them short shrift for reasons that make sense to me, but may not be obvious.
The words “user experience” is thrown around a lot these days in place of “UI” and sadly also in place of “usability”. I’ve always been curious and tried using many platforms. I cut my teeth learning to use computers in the System 6 era of the Macintosh, during the period where there was a clear chasm between the Macintosh OS and Windows 3.11. The Mac was a coherently designed platform, with cohesion, with a sense of nuance and personality. It, and Windows 95 after it, had a culture that became synonymous with the OS, that put clear expectations in your head as a user and that you could fulfill as a developer. You don’t need a single company to do this, but you need a place of authority that’s open to criticism and change, that will adapt and survive, that exudes longevity.
Unix is a platform like this, and it has a coherent and cohesive user experience. It has clear rules and ideals that make sense - they’re all about doing one thing and being able to be reused. The problem isn’t Unix. The problem is that if you don’t want to live your life inside a command prompt (no offense to the people who do), it starts to fall apart. Package managers work great. X is a trainwreck, fixing it is impossible and getting people to move to Wayland is still an uphill battle. The Unix idioms and ideas are being applied to graphical UIs, where they don’t make sense, and what results is a cacaphonic mess, not conducive to a pleasant and effortless user experience; things will look different and work different and be largely programmed, structured and maintained by people who do not care about usability. There are people who do care about it, but in this culture they tend to move on to other things.
All this said, the significant time I’ve spent in various forms of GNOME or KDE over two decades has not once instilled confidence in me. (Again, aside from package managers. Although with
systemd now being a sprawling, hot mess prone to extreme security flaws but still heading to near universal adoption, maybe I just have no idea what’s going on, and all of this can be ignored.)
The previous section was not simply a detour. You may notice that other platforms have their own signature viewed through the platform cohesion prism. Windows 95 and Windows XP were both very cohesive. iOS started out being incredibly cohesive, has gone here and there, and is arriving at a point where I may not agree with all the decisions, but at least the apps in the OS feel reasonably consistent.
The Web is a difficult beast. Web pages, by and large, are easily understood. They have an easier job because they’re not asking for a lot of interaction. They are mostly vehicles for information (text, video) with limited interactivity. Web pages are cohesive, for the same reason reading a magazine is cohesive. There’s no learning curve.
Web apps? It depends on the app itself. The original version of Gmail looked like someone had made a UI toolkit out of Google’s own web site, but it was easy enough to use and solved more problems than it created.
Web apps as a way to implement the range of functionality most often associated with full-on desktop applications? Only very rarely done well, and I can’t think of a good example off the top of my head.
Web apps are part of the problem, but they’re not even the only part of the problem. The other part is the copy-that-floppy road of lazily emergent (read: lack of) user experience design. People wanted to make early mobile apps that put many choices at a user’s fingertips, so they invented the hamburger button, opening a sidebar menu, as a reduced example of a menu bar. Other people wanted to not change user paradigms, so they put the same interface on a tablet app. Yet other people wanted to make desktop apps look modern, so they made them look like tablet apps. And so what we now have are web implementations of 3.5” touch screen assumptions for a desktop platform. If your goal is to make an application that users of desktop applications will find familiar, you couldn’t start from a worse place if you tried.
It’s not that desktop apps should look the same if you leave the planet and come back ten years later. It’s that they should at least not completely give up everything that was put in place to make them understandable and efficient; and that if they do replace those things, that they are replaced with things designed with desktop applications in mind. Shrugging your shoulders and saying “but everyone has a phone and everyone uses the web” is like arguing the door on a microwave should look like an actual house door or the lid of a toilet.