I think partly it is because UX is measurable (“A is less usable than B and thusly bad”), whereas aesthetics are not (“different tastes”). A kind of means to escape personal emotional accountability.
The problem is also what we do with the ability to measure. Typically, we're making easy things intuitive, and difficult things impossible. This is exactly the wrong way to go about things, if you care about delivering value to users[0]. We should be making the easy things easy, hard things possible, and forget about the whole intuitiveness thing.
There's this widespread belief now that software is only good if a user who never saw it before can become proficient in it in seconds to minutes. I think this is one of the most devastating, dangerous ideas in computing. The only way you can achieve a learning curve like this is by removing almost all functionality from software - make it so dumb that it really takes only a minute to figure it out entirely. Sadly, this is what we see in mobile and web applications these days.
What worries me here is that we've conditioned everyone to assume software is immediately and fully discoverable. Nobody is expected to read the manual these days, and so manuals are not provided, and since manuals are not provided, any feature that cannot be made apparent without explaining it in the manual goes away.
(Even with kitchen appliances, the situation isn't that bad. When a person sees a particular appliance for the first time, they do read the manual, or get someone to show them how to operate it. Maybe it comes with the fact that buying appliances is expensive and overall a hassle, whereas software is too easy to procure?)
--
[0] - I highlight that condition, because it's my belief that most software vendors don't care about delivering value to users. They care about making money off users, and there are many cheaper ways to do that than creating a truly useful and ergonomic product.
> The problem is also what we do with the ability to measure. Typically, we're making easy things intuitive, and difficult things impossible. This is exactly the wrong way to go about things, if you care about delivering value to users.
Right. We're prioritizing learnability over usability.
> We should be making the easy things easy, hard things possible, and forget about the whole intuitiveness thing.
Whoa there... Another way of thinking about 'intuitiveness' is in terms of affordances. We mustn't throw the baby out with the bathwater. While it is too much to ask that every function should be obvious upon first seeing the UI, it is not too much too ask that every function should at least be obvious in retrospect after trying to use it or having it demonstrated, and of course leveraging affordances and interaction patterns the user is likely familiar with from elsewhere should be given priority.
> Whoa there... Another way of thinking about 'intuitiveness' is in terms of affordances.
Yes, of course. Thanks for bringing this up. I apologize, I went a bit too far there - what I meant was just "intuitiveness" in the sense of expecting people to be immediately able to work well with something they see for the very first time, with no explicit learning or training.
I also don't mean to ignore familiarity with UIs in general - yes, unless you have a good reason, it's a good idea to copy design elements users are well familiar with (if they're not completely insane, or dark patterns). This matters particularly on mobile and desktop. On the Web, everyone is used to websites looking different from each other, but there are still higher-level patterns (like footer with company info, "contact" link somewhere on the site, site logo redirecting to home, etc.).
I totally agree about affordances, and mental handles in general. "Obvious in retrospect" is a great way of putting it - once you know a feature exists, or used it briefly, it should be easy to find it again. Once you familiarize yourself with a bunch of features, it should be obvious where to find them, because they should fit a consistent mental model. In a way, it's the job of the UI - to let the user learn the correct mental model of the application, and how it manipulates underlying resources.
For that to happen though, you as a software team need at least to a) have a consistent mental model yourself, and b) design both "backend" and UI around that model. I think this is one part where we fail frequently, but unintentionally - the developers and the designers don't spend enough time ensuring they have a shared mental model. When this happens, you have UI that may be consistent with itself, but feels off when used, and every now and then you see surprising behavior or incomprehensible error messages - that's the "backend" model leaking out.
I *try* to make my software initially intuitive and gradually discoverable. I think a gentle learning curve is better than a completely flat one.
I thought quite a bit recently about the problem that you are talking about (diminishing end user value in software). I think the source is that we have a lot of devs and designers whose only experience is designing for maximized conversion rate or maximized engagement. As opposed to professional software.
They have no habit of designing for value.
And *that’s the culture*. Go to any awwwards gallery - all of the websites or webapps mentioned there are essentially ads with minimal content.
Which isn’t too bad, since it leaves a market untapped through the collective arrogance of the incumbents...
