Hacker News Reader

Comment by ryanSrich | Hacker News Reader

I’m not even that old, but I despise the entire UX industry for what they did to quality, fun UI work. They stole every ounce of soul from interfaces in the name of “research”. I’m glad the reckoning is happening. I’ve not met a single UX researcher that could actually design anything. They’re glorified business analysts and they clearly can’t even do that right because at least business analysts can create requirements from real customer problems. UX research exists purely to only serve itself.

View on HN

Yeah, I'm with you on this one. The author describes the past 15 years as a "golden age for UX Research". It's been the golden age of dark UX patterns, huge widgets for touch-inadequate applications running on platforms without touch screens, cartoon figures in professional applications (please explain to me how Octocat is different from Clippit), monochrome or bland icons you can't tell apart, surprising interactions through label-looking text that turn out to be buttons, and metrics-driven positive feedback loops. All of which were overwhelmingly validated by the UX research community, which gradually moved away from reproducible, quantifiable experiments (which the author lumps under "micro-research") and embraced a kind of number-backed hoodoo that makes 1980s psychology research look good in comparison.

>please explain to me how Octocat is different from Clippit

* Octocat is the mascot of Github, while Clippit was not only not the mascot of MS Office, it wasn't even the only option of assistant, just the default one.

* Octocat is not a functional part of the UI. That is to say, you don't interact with it to do anything. It's just part of the Github logo and it shows up on certain screens, notably the 404 one. Clippit could be used as an additional interface for the application to accomplish tasks.

* Octocat isn't animated, Clippit was.

It seems to me that the only thing they have in common is that they're cartoon characters that appear somewhere on the interface of software products. You can complain about it on aesthetic grounds, but certainly not on UX grounds. It doesn't make Github any easier or harder to use.

> You can complain about it on aesthetic grounds, but certainly not on UX grounds. It doesn't make Github any easier or harder to use.

Neither did Clipit -- every function it offered was also accessible through the UI. It was a completely optional element that degraded the user experience exactly the same way Octocat does: you got to look at a goofy, childish character whether you wanted to write a cute check from the tooth fairy for your niee, a report on inappropriate use of company resources, or a death notice for your late father's friends.

You could at least turn it off, whereas Octocat is in the top-left corner of literally every Github page, and prominently featured wearing an astronaut's costume (!?) on the homepage.

The reason Clippit was so hated was not because it was a cartoon character, it was because it was enabled by default and it would pester you about what you were doing (thus distracting you) until you turned it off, and because the assistance it offered was neither required nor useful. It was a net negative on your usability of the application until you took active steps to turn it off.

>you got to look at a goofy, childish character

Like I said, aesthetic grounds.

>whether you wanted to write a cute check from the tooth fairy for your niee, a report on inappropriate use of company resources, or a death notice for your late father's friends.

>You could at least turn it off, whereas Octocat is in the top-left corner of literally every Github page, and prominently featured wearing an astronaut's costume (!?) on the homepage.

This is quite an inane complaint. Plenty of serious products and companies have cartoon characters as mascots or logos. LLVM has a dragon in the logo; the Go language has a gopher; the Rust language has a crab; the FreeBSD and OpenBSD OSs have a devil and a blowfish respectively. All of these have varying levels of visibility when you visit their web sites.

I'm sorry but I really don't understand the point you're trying to make. Yes, there's a difference in features, but my original point was

> cartoon figures in professional applications (please explain to me how Octocat is different from Clippit)

Is Octocat not a cartoon figure? Is Github not a professional application?

I'm not aware of any authoritative definition of user experience that excludes "aesthetic grounds" from the field of UX. How an interface looks and the feeling it conveys is a part of the user experience. The UX industry doesn't operate with this distinction -- see e.g. NN Group's oft-quoted definition (https://www.nngroup.com/articles/definition-user-experience/), which considers aesthetics, and UI design in general a component of UX, or the ISO 9241 definition.

A tongue-in-cheek logo is one thing, but a goofy interface, from the ever-present logo to the overall widget design, is way out there in kindergarten land. The comparison between Octocat's use and, say, that of the Rust crab, is itself unwarranted: you can (in fact, I do) write Rust all the time and not see the crab for weeks, whereas that goofy cat critter is literally on every Github page.

As for Clippit specifically, while its graphical design was obviously not the primary reason why it was hated, it was definitely one component of it -- in fact, it's one of the reasons why it was specifically considered distracting. Contemporary studies, like Swartz's notorious dissertation about why everyone hated Clippit, obviously focused on the functional aspect, but it's telling that the quantitative framework in use at the time put "fun" and "useful" in separate categories.

You asked what's different between the two and I answered your question. I don't think there's anything else to discuss.

Indeed, there isn't -- you've provided pseudo-answers that miss the point completely.

I hate to be that guy, but my silently boiling rage spills over every once in a while. UX designers are so passive aggressive and self-satisfied. They foist change on us dumb users with impunity because they are a better class.

The sad truth is that UIs have gotten gradually prettier but less usable since their peak in the mid to late 1990s. I honestly don't know how an entire UXR department can be justified if they just keep making things worse. Frankly I am sick of the utter wasteland that touchscreens have created, the giant crater they blasted in people's imaginations. And I'm sick of this endless cycle of redesign.

I do not give two shits about drop shadows or rounded corners or animations. I want to use the frickin software tool to do frickin work. I am not looking to play tiddly winks with an animated brochure.

Bring back the damn drop down menus written in my native language with keyboard shortcuts that are right there in the damn menu. I'm sick to death of iconese. I don't speak it. It's not a language, it's alien hieroglyphics. And stop telling me it's for my own good. It's for their good, so the crap fits and they don't have to go to a translation team. Nothing could be lazier or more hostile to users.

Which brings me to the final sin: if you are a UXR and you are not polling actual users, talking to actual people, sitting down and actually watching them use the thing, listening to actual words coming out of their mouths when they do so, which include an *&#$ of profanity, and you are instead relying on indirect means, then congrats, you are a trying to scale a process that doesn't. fricking. scale.

I agree with most of what you said, except this

> The sad truth is that UIs have gotten gradually prettier

I don't think they've become prettier at all. What they've become is bland and uniform. All basically the same look, apparently trying to be inoffensive rather than beautiful.

I think people here romanticize the old days of every website and piece of software having its own unique and quirky UI. Sure, it can be fun at first, but learning a million unique UIs that all look and behave differently is exhausting as a user.

If "predictability" is a key principle of good User Experience™, then UI uniformity is the best way to achieve it. Consistent interface patterns translate across applications and make software easier to use, on the whole. Users learn the pattern once and can use it everywhere with lower mental overhead.

I would take a bland-but-consistent UI across all software over beautiful-but-unique. Modern UI design has taken this stance too for good reason.

Also worth pointing out that a "bland" UI of today (let's say macOS or iOS since they are so pervasive) would be stunningly, mind-blowingly beautiful to a user 30 years ago. It's all relative, we've just gotten spoiled by beautiful UIs to the point they are considered bland now.

We aren't actually disagreeing here, but you're addressing aspects other than what I was talking about.

> Sure, it can be fun at first, but learning a million unique UIs that all look and behave differently is exhausting as a user.

Absolutely. I wasn't calling for anything like that.

Uniformity in behavior is highly desirable. Uniformity in the general location of controls and general workflow is also highly desirable. Also highly desirable (but not in fashion anymore, so isn't really done) is the ability to easily tell how to accomplish things.

But you can have that without being visually bland and without looking essentially indistinguishable from everything else.

> Also worth pointing out that a "bland" UI of today (let's say macOS or iOS since they are so pervasive) would be stunningly, mind-blowingly beautiful to a user 30 years ago.

I'm old enough to remember being a user 30 years ago, and you do have a point. But the most beautiful painting in the world looks a lot less beautiful when every other painting looks just like it.

Also, I would argue that the "beautiful" days of UI design are over. What is in fashion now wouldn't really have been considered mind-blowingly beautiful 30 years ago. Technically impressive, sure, but aesthetically neutral at best.

> we've just gotten spoiled by beautiful UIs

Not spoiled. Habituated. It's the way people are. If you eat nothing but your favorite food every day, you will eventually reach a point where you dislike it.

Variety is an important design goal, too. You can have variety without sacrificing beauty or consistency. You jsut have to break out of using UI frameworks or slavishly following guidelines like "Material Design", or whatever the flavor of the day is.

Regardless, I'll take the UIs of 30 years ago over those of today even though they were uglier. They were also more powerful and generally easier to use.

Being honest, I don't know where the UX industry even came from. They just appeared one day in the early to mid 2010s and proceeded to ruin every single piece of good and usable UI known to mankind.

The computing world was at peace until then, we had UI designers and engineers who through much research and trial-and-error produced very nice UIs for us to use. The UX industry came in and destroyed all that for shits and profits; literal decades of experience and precedence thrown out the window in the name of the latest fashion trend.

> The computing world was at peace until then

Really? They just came from a war elsewhere. Standing up from XML trenches and walking joyfully into green JSON fields. JSON to hydrate UI's.

JSON and XML are not UIs ( in the same way the bits scattered on a hard drive platter is not UI).

if the user is removed from an user interface by means of abstraction, does that change the nature of the interface?

> we had UI designers and engineers who through much research and trial-and-error produced very nice UIs for us to use.

I don't remember that era of great UI. Sites were still inaccessible except no one cared, multitude of dark patterns still exists, confusing design language, no information organization at all.

The "trial-and-error" approach was much more "error-and-error" based entirely on guesses by the designer or engineer. I'm an engineer and remember, working through that time period, the horrible sites we created. I had a part in implementing them.

Not to mention the idea of "Web Application" was still very much incipiente in the mid-to-late 2000s. Who had great design for that? Go back and look at Youtube, Facebook, Twitter, MySpace back then, hardly a great example to be had, regardless of how bad we might think it is today.

I mostly agree with you, there are some really smart UX people out there but the vast majority out there are garbage who can just use figma and make wrong decisions based on incomplete business understanding or technical expertise or sometimes even both

You seem to be mistaking UX with UI design.

You seem to think there’s a legitimate difference that isn’t entirely contrived.

I don't think playing "my definition vs your definition" is a fruitful discussion in general, but the answer is, it depends.

If you're in the camp who treats UX as a professional title/career, then in principle it has very little to do with UI, and being labelled "UX/UI" is often an "insult" to UX professionals along the same lines as "Computer Scientist / IT support", or "Electrical Engineer/Electrician", "Civil Engineer/Sanitation Engineer", "Pharmacist/Teller at Boots", "Doctor/Medical receptionist". (not to imply that the latter professions are 'lesser', but rather to point out that the former professions in the pairs are typically deemed to require a higher standard of skills/training rather than simply vocational/apprentice-based training).

At the same time, I'm not naive to not understand the fact that, if a large number of people start referring to X as Y, then after a while, insisting that X isn't Y is a lost cause, and you might as well start calling it Z instead.

But, e.g. my wife was a senior UX professional who is now doing her PhD in Experimental Psychology; the original motivation was that this would make her a better UX researcher when conducting controlled experimental studies on behalf of her employer. She specialises in voice assistive tech, and her topic relates to isolating auditory markers of voice trustworthiness. I can tell you if she applied for a UX job and the interviewers started asking her what css colours she prefers on buttons, she'd be pretty pissed.

It might be helpful if you explained the difference between UX and UI design here. I'm legitimately fuzzy on what the difference actually is.

It might be helpful to view this from a scientific research perspective. UX research is more about how things should be done to provide a better usability for the user. Like, general respectively abstract concepts (e.g., how should a chatbot present itself when it first "talks" to the user, which UI components could be involved). It's not about how to actually design those things, i.e., applying CSS to an HTML document if you will. UI design is mostly a dev job, while UX design is more of an actual research discipline.

Thanks!

With that understanding, it seems to me that many, if not most, of the criticisms being levelled here do, in fact, apply to the field of UX research.

But certainly not all of them.

> You seem to be mistaking UX with UI design.

This is often a motte and bailey distinction though, where the "UX" person's background is UI, and they think only about UI. But they can charge more because theoretically UX is everything.

I don't think this is an individual's issue, as if everyone hires UX instead of UI then what do you do? It's a UX consultancy issue.

It used to be something specific, coming from the heritage of HCI and usability engineering. Then every UI designer decided that it sounded cooler on their resumé, and now the concept is completely pointless.

The poor usability of modern interfaces certainly doesn't come from people who care about what it's like to use it in practice. It comes from people whose main skill lies within the field of aesthetics.

What should you call yourself if you're actually interested in working with usability, and happy to leave the visual design to others?

> What should you call yourself if you're actually interested in working with usability, and happy to leave the visual design to others?

UX.

Most serious people in UX who see job postings mentioning "UX/UI" avoid those jobs in the same manner that a C++ coder would avoid "C/C++".

But I agree that doesn't do much to dissuade people who don't quite know the difference (or what it is that they actually need or want) from advertising "UX/UI" or "C/C++" and then people applying to these jobs.

What you're describing isn't UX though. At least not in the way that UX professionals understand it.

There being "impostors" in the field who bluff their way through, while regrettable, does not imply that the whole industry is made up, any more than the existence of people applying to data science jobs with nothing but excel skills says nothing about data science.

And some companies really, really need UX people on board.

Well, I think it's certainly not how UX is sold, but I think it can end up that way, or even worse. I've been in several meetings with expensive UX consultants where we're paying them to run team retros for teams they aren't in, or doing mood board style exercises. Lots of internal work that's invisible but claims to add enough value to pay a large amount of money for.

Now, I'm someone who insists on having a UX/UI person in the team when I'm making an app, as I want it all to go well, and I don't mind spending a chunk to get it, but the number of people I've found who can actually do that is vanishingly small compared to the number of people who are billed out as expensive UX consultants.

Reply on news.ycombinator.com