I do not understand why designers are so insistent on making tools so difficult and unpleasant to use.
The modern web disgusts and enrages me more often than not.
talking to the user instead of taking the decision made by some investor/suit in a larger, less efficient firm.
I understand this 100%. Youtube is the same exact thing - the home page is total garbage unless you enjoy watching the same 8 videos you just watched. But we're in a thread about bad UI/UX, so I'm going to complain that the "default" being bad is a bad thing.
[1]: FF: https://addons.mozilla.org/en-US/firefox/addon/youtube-recom... Chrome: https://chrome.google.com/webstore/detail/unhook-remove-yout...
Now, if they could just make the windshield wiper controls all work the same direction...
I've never tried to test it cause I don't like horribly expensive mechanical noises, but yeah.
FWIW, 25 years ago, I had some generic assignment when I was student, I don't remember the exact details, but I made it on auto-windshield wiping tech, and what I found was that multiple patents already existed back then (one from Mercedes or BMW, I don't remember which ; it might have been both). I'm mystified that we aren't seeing those everywhere yet.
The new tech these days is to spray fluid out of the wipers themselves. The blades have a lot of holes laser cut into them, as the blades move across the windshield they spray fluid just an inch ahead of the blade. It's starting to become very common in the luxury car brands, even the lower tier Lexueses and BMWs are getting them as the newer gen cars are released.
It doesn't sound particularly cool, but when you try using them it is a very nice feature to have.
It was interesting but I didn't find it especially better. They moved away from this and the later models just have normal sprayers. Maybe the very latest have that again, I don't know.
Isn’t this a solved problem and completely standard on most new cars for a few years now?
Some work better than others, but most of the time when they don't work well it's because the wipers are worn and need to be replaced.
Automakers also equip automatic transmission cars with paddle shifters, which really give you no tactile feedback to let you know what gear you're in. They also make it difficult to shift while turning. A traditional manual with the HH shift pattern gives you that feedback that allows you to know what gear you're in and doesn't limit your ability to shift while turning.
Pros of touchscreens:
- You can make the UI look like anything you want.
Cons of touchscreens:
- You can make the UI look like anything you want.
i.e. I don't mind touchscreen A/C controls or something, provided the driver's wheel also has those controls (which in my old Toyota Prius it does).
IMO the big innovation is making sure that everything I would want to do while driving is literally at my fingertips, which should be on the steering wheel.
being able to turn the heater or defrosters on without looking is a pretty big upside when driving.
tesla changing where that is in updates and making you look for it in a new design just makes a bad problem worse
There was extensive discussion about that on here a short while ago:
Actually being able to customise it yourself (and have the changes persist long-term) wouldn't be that bad.
I don’t like that on my computer and definitely don’t want it in my car. It’s bad enough when you buy a new car and things have been “improved”.
Aftermarket steering wheels are widely available for road cars. Most of them are wheel shaped.
https://www.consumerreports.org/video/view/cars/news/6272879...
The steering wheel is already motion controlled.
It was basically this.
Because they need something to pad their resume. In fact, most of the things that are wrong with software today are for the same reason.
Executives want to make bold changes, so the Start button must be "re-imagined." For the same reason, Product Managers "re-imagine" the menus that have been in place for years and Product Designers "re-imagine" the design language to remove all cues that a button clicks and a panel scrolls.
In my routine trips to the auto parts stores over the past 20 years, I have seen their throughput drop as more and more of these modern ideas seeped in. Their text menus had been the same for decades, with as many options for a next action as there were keys on the keyboard. If my intuition is correct, they were remote sessions to a server/mainframe. How about that for keeping the computation close to the data. Now that has all been replaced with a single button and a cursor.
I'm not a die-hard CLI user by any stretch, but I understand the impulse after watching usability circle the drain year after year. There's a healthy middle somewhere and I am hoping to find it.
Let's remove all indications that buttons are buttons!
I've actually had to use a site which had all four combinations: static text that looks like static text, static text that looks like a button, buttons that look like static text, and buttons that look like buttons.
This means that the app will work amazing and without thought ("it just works!") for 80% of the population but will be non-functional for the 20% of the population which includes power users and users who would have been willing to go through the initial learning curve for a more powerful app.
Just to note: I'm not deriding the "lowest common denominators" as I am definitely one for some domains. For example, I chose the app with the simplest deisgn for doing my taxes as I didn't want to spend time learning about taxes.
Recommend https://en.wikipedia.org/wiki/The_Design_of_Everyday_Things
EG, you're building a bridge. It needs to do its job, and not collapse. There are many perfectly functional designs, which have appealing form. Yet, if you want to actually be able to use the beast, it:
- needs to be wide enough to pass traffic, but not too wide (waste of money)
- needs to be able to keep the traffic from careening off into the abyss each use
- needs 100 other things such as the above
But once the above "functional" criteria are met, form is fine. Why not make it look good? Great! Yet, don't (as an example) overlay form which then reflects sunlight into driver's eyes, so they can't see and crash on the bridge.
This is all reasonably understood with normal day objects.
But, when someone engineers a bridge, they have to go to a decade of schooling + years on the job, to rise to that position. And the same for people designing beautiful buildings, or homes. They know the restrictions, the load requirements, and then they layer extra on top of that.
Not even sure what I'm fully rambling on here about, except I guess that I think UX/UI should only be attempted after years of using software in a professional capacity.
I was in to a store the other day, and they were looking up stock. It was an ncurses display. And the employee was zooming through it, as typical, with keyboard shortcuts, and FAST.
UX/UI is really the problem. Making things pretty gets in the way of functional first. Keep it function before form, and you're fine.
And if anyone rattles on about Apple and Steve Jobs, from what I've read, heard, and watched? He was 100% about "function first!". He'd go ballistic if simple functionality was hard to get at, or missing, or buggy, or incomprehensible.
Yup, he wanted form to look good. But that came on top of proper, perfect function.
Which is its own beauty.
Having done research into UI/interaction design, I can confirm that all research in this area prioritizes things that (a) can be measured quickly and (b) with a high degree of confidence.
Good luck running a controlled experiment where multiple folks use an interface for an entire week and then become power-users of the system!
I think one interesting way to look at these sort of changes is analogous to the introduction of new features in code. In the short term, new features can be added to the existing codebase without the need for much change, but eventually, the overhead becomes problematic and there is the need to make more structural adjustments.
I think it’s similar for designers, who get to the point where new features don’t have a natural home in the existing interface and want to make changes. Just as a junior developer might be tempted to rewrite from scratch, a junior designer might be tempted to introduce a new mental model and rearrange everything. In some cases this may be OK but generally you need a more nuanced approach. The fact that users are the consumers of the change, not designers, is where the analogy falls apart, and thus the need for a more thoughtful introduction of any change.
Obviously the abrupt changes are those that cause a negative reaction, especially when combined with the exaggerated desire to clean up an interface. However, not all change is for changes sake. Sometimes it’s just a case of trying to respond to constraints and make the best compromise when dealing with changing software.
For personal use, I find 1Password typically handles multiple screen logins, including 2FA quite well, although requires an extra UI interaction.
Underlying this fallacy is the assumption that usage should somehow be uniformly distributed across all features. In reality, a power law distribution is more natural.
1) Visual changes show better than feature lists. I'm not convinced this matters anywhere near as much for customers as marketing folks seem to think it does, but marketing values it regardless, management values it (it'll show better to their bosses, too), and product managers, designers, and front-end developers value it (so they have nice-looking things for portfolios and presentations, both internal and external)
2) Making something that looks good in a screenshot or demo is not the same thing as making something that works well (that is, has "good UX"), and making something that does both well is hard enough that the talent (and environment/budget) to do it may be fairly rare. The former matters to those concerned about point 1, so that's what gets priority (which is why so many redesigns are neutral, at best, as far as UX, and many are worse).
3) If design is done well, it's heavily cyclical. Design guidelines and languages and component libraries and such should be built up such that most design work only happens when there are initial designs, very large feature add-ons that break with what the previous design guidelines can accommodate, or redesigns. So, if you're a designer and want to stay comfortably employed, and you're a manager who doesn't want to surrender head-count or lose your people to other projects, and a large part of the org has various incentives to push initiatives that result in something nice to to look at (see 1, again) what happens? Lots more redesigns than need to happen. LOTS more.
[EDIT] And all this is before we even consider possible business reasons to deliberately make the product worse, which are also A Thing.
This might be true from a bottom-line perspective, but the marketing/product/engagement manager's targets (whether company-mandated or personally-mandated to use them as resume bullet points) may still hinge on increasing "engagement".
what if they're actually designing systems and schemes which use you? (in soviet russia ---and the modern day internet--- computer uses YOU)
I started trying to make a joke but now I'm not laughing :|