Bad UX Roundup #7 — The unique episode

When I was in grade school, the teachers always reminded us that it’s ok to be different. In fact, different is good. Embrace your differentness. Even as a 3rd grader, I was skeptical of such a blanket statement. I recognized that there is “different” like the kid who could read three grade levels ahead of the class, and then there is “different” like the kid who eats their own boogers and makes explosion sounds with little finger explosion gestures.

Fast forward a few decades, and now everyone wants to show off what a special snowflake they are. The mustachioed hipsters at tech companies love to show off how “creative”, “unique”, and (gag!) “quirky” they are through their “innovative” designs. Who wants to just copy the competition? I want to be different! My teacher told me different is good. Stop triggering me.

I have a saying that it’s better to be generically good than uniquely bad. This works out well for me because there are many unique ways of being crappy, and it gives me lots of bad UX to write about.

And here are five examples.

On Articulate.com, the login is at the bottom of the page.

Articulate is an online learning course design platform. Truth be told, their biggest UX blunder is not fully supporting Mac, but that’s not what I’m talking about today. This is a very simple blunder. Look where they put the login link.

As you may have guessed from the legal boilerplate, that link is at the very bottom of the page. The copyright line must be the single most ignored part of any webpage. 99.9% of people probably mentally block off this space as a place that their eyes don’t even need to touch. And yet, Articulate decided to hide their login link there.

It’s bad enough when companies make the login significantly less visible than the sign-up link. This is a whole new level of obnoxious.

Important lessons

• Don’t deviate from established patterns without a good reason.
• Don’t hide any important functions at the foot of the page, especially on the copyright line.

On Google search results, the content types switch order seemingly randomly.

I cannot stress enough this blunder’s takeaway. Before you can strive for greatness, you must rise above mediocrity. The hallmark of UX noobs is the vainglorious quest for creativity and innovation that produces spectacular works of inanity. The vast majority of the time, a tried-and-true design is what works best and yet, all too often some wet-behind-the-ears hipster decides to swim out of their depth in a bid to be featured on Co.Design. Hopefully these clowns will learn from their mistakes. At very least, you can learn.

Let’s say I decide to look up the New Hampshire stoner metal band Scissorfight on Google. When the results load, the different types of content will be separated into tabs, such as videos, news, and images.

If you were naïve, you might assume that these tabs would remain the same no matter what your search was. You might want to program these locations into your muscle memory so you could quickly toggle between the different content types. You might be annoyed with the results if you were to do that, though. Look what happens when I look up Memphis hot chicken (there’s a new place in Seattle that serves it).

And if I ask why dogs eat grass, suddenly I get a link to books on the subject to the right of “all”.

Someone at Google probably thinks they were being cute and helpful by conveniently moving the most correlated content tabs to where we would be most likely to click. That someone doesn’t understand how people work.

When someone uses a product as regularly as they do Google, they begin to construct a map of where all the interface elements reside. Over time, that map gets burned into muscle memory, and use of the site becomes second nature. Think about all of your favorite sites and apps. You probably don’t even think about where you click any more than you think about the street numbers on your drive to work.

Google decided to ruin this happy arrangement by shuffling the interface elements for no reason at all. It isn’t saving us a click, since it doesn’t automatically load the most likely content type. Even if it successfully guesses what kind of content we will probably click on, there’s nothing in our nature or our conditioning that tells us to reflexively click the link immediately to the right of “All”. It almost certainly gets wrong what type of content we’re after frequently enough that we would never form such a habit.

I’d say it was worth experimenting except that common sense tells us what a bad idea this was.

Important lessons

• Don’t try to be innovative for innovation’s sake.
• If users develop usage habits, don’t interfere with them for no reason.
• Interface customization does not mean randomly shifting around elements.
• Don’t hide stuff behind a “More” link. That applies to ellipses and hamburgers too.

The insane UX of Google Inbox’s trash

I won’t lie. This might be some of the worst UX I have ever encountered. And I encounter a lot of bad UX, if you haven’t noticed.

If I go into the Trash in Inbox, each deleted message looks like this:

What do you think that trash icon on the right does? It must permanently delete the message… right? That would be the trash of the trash, after all.

Nope.

Trashing something that is in the trash actually puts it back in your inbox. I can’t make this shit up. I’m not sure in what parallel dimension that logic holds up, but the possibilities are endless. If you want to bring a dead body back to life, just shoot it in the head. When a prisoner commits a crime in prison, they get sentenced to freedom. If you want to enjoy a meal again, you just… ok, you get the point.

And, by the way, that’s the only way to put something back in your inbox. In Gmail, you can use the “Move to” function or you can click and drag. Inbox does not let you do either; in fact, it doesn’t even have click and drag at all (maybe I’ll put that in another episode). What are the odds most people would even guess that trashing the trash is how to salvage a trashed message, especially if they are flustered at having accidentally deleted an important e-mail.

So, if that trashcan icon is how you un-delete something, how do you delete something for good. Like, what if you have a phishing or malware link in your trash that you don’t even want anywhere near your inbox. You can’t. You can empty the trash completely, but you cannot permanently delete individual e-mails. For that, you’ll have to go back to Gmail which, despite being older and uglier, is actually a complete product. Inbox has been on the market for years and is still missing basic features.

I suppose it’s appropriate that the one of the worst pieces of UX design can be found in the trash. That’s right where it belongs.

Important lessons

• Understand the user’s mental model.
• Don’t omit basic functionality.
• Do not attempt to resurrect the dead with firearms. It doesn’t work.

OSX’s bizarro cut-and-paste behavior

I bet you didn’t even know you could cut and paste files in OSX Finder. More likely, you probably were confused about whether the ability was there or not, because “Cut” shows up in the Edit menu.

But what happens if I select a file and ⌘X? Nothing. Absolutely nothing. If I open the Edit menu, I will see that “Copy” is available, but “Cut” is greyed out. What gives?

This UX peculiarity is actually a fractal of weirdness. First off, yes, you can cut and paste in Finder. But it’s ass-backwards. First you select a file, and then you hit ⌘C (or select “Copy”). Then you go to your destination folder and you hold down Option and hit ⌘V. The file you copied will be pasted into the folder and it will disappear from the place you copied it.

See the illustration below:

Yes, adding Option retroactively turns the Copy you already performed into a Cut. Does that make sense to you? No, it doesn’t.

While searching around for answers, I found this desperately fanboyish post on Stack Exchange employing some serious pretzel logic to justify Apple’s design:

Unlike this action based task-flow, Mac has taken the route of cognition based task flow. They do not have separate action keys for copy and cut, rather they are more task based — copy and move. So a single action to copy brings the item to your clipboard (think of it as picking the item from your container). Now while taking the action of putting it into some other container, the user can decide whether he wants to drop that item in new container or simple place a copy of the picked item in new container while putting the original item back in its parent container.

I hope this guy isn’t a UX designer. Who am I kidding? He probably works at Apple.

Let me list everything wrong with this logic

• Apple are “solving” a nonexistent problem. Most of the time, I have already decided what I want to do with the file once I hit ⌘X or ⌘C. On the rare occasion I change my mind, I can go back and paste it there. Or better yet, Apple could add another hotkey to return my cut file to its origin.
• Every other program I use employs the standard cut-and-paste paradigm meaning that even if, in theory, there is some marginal advantage to Finder’s model, it is completely negated by the cognitive load of switching between mental models. In theory, Dvorak is easier to use than QWERTY, but that’s only if you completely ignore the pesky detail of decades of daily habituation.
• With standard cut-and-paste, both the act of cutting and the act of pasting are visible to the user. The user sees the file(s) get cut when they hit ⌘C and they see the file(s) get pasted when they hit ⌘V. With Mac’s copy-then-maybe-cut-and-paste, the user does not see the file get cut because it’s happening in a different folder. This obscures half of the user’s action. So no matter what dubious usability benefits come from letting the user decide what they want to do after the fact, the usability drawbacks outweigh them.
• Also, why even call it “Copy” if the function could potentially turn into “Cut” later on? I guess giving it another name would only compound the confusion.

What, exactly, is the cut feature even there for then? Pretty much just copying the text in file names. Yeah, lame.

Oh, and one more thing. OSX used to not have any cut-and-paste for files. At all. You had to either drag the file from one folder to another until you reached your destination, or open up two Finder windows to the respective folders and drag it from one to the other. According to this article, the oddball cut feature only showed up in 2011! This is just one more example of Apple being so obsessed with “thinking different” and showing how innovative they are that they produce something uniquely stupid.

Important lessons

• Show the user what is going on. This is one of Nielsen Norman’s heuristics.
• Don’t break with UI convention if the convention works well and is ingrained into the user’s muscle memory.
• If you are designing a cut-and-paste function, add a hotkey which cancels the cut and sends the cut object back to its point of origin.

The bizarre way of undoing claps on Medium

What better way to show your approval for someone than to give them the clap? I mean, a clap. Since you’re reading this on Medium, you are probably aware that instead of the tried-and-true binary “Like” feature of yore, Medium now employs something called “claps”. You can hit the clap button multiple times, to award up to 50 claps. Yes, that means you can give more than one, which means if you’re going to do it on my article, hit it a lot of times.

While I think it’s a stupid concept in general, the biggest problem with it is, how do you undo it? I discovered this the hard way when I accidentally clapped on some crappy comment, and then made it worse by clapping a second time to make it go away, a la The Sorcerer’s Apprentice. With those old-fashioned, uncool, binary a.f. likes, you can just click them a second time to make them go away. Medium claps are harder to get rid of.

So how do you do it? That depends on how you’re viewing Medium (the Medium medium).

On the desktop web, you hover your mouse over the clap button once you have clapped, and an X will appear after a second. It’s a simple act, but it has lots of problems. While there are other examples of a delete option appearing when you mouse over content, they don’t come with a delay. A user might actually think to try a mouse hover if they accidentally clap, but the delay is just long enough that they might pull the mouse off before the X appears, at which point they will be unlikely to try that approach again. The only time you should hide a feature behind a mouse hover, especially one with a delay, is when it is a highly established pattern, and this just isn’t.

Unfortunately, the hidden hover X is a hell of a lot better than what Medium has you do on the iOS app. For that, you have to click the SHARE icon. You know, that universal box with an arrow poking out of it that NOBODY will ever assume contains a command to undo likes.

So, not only does the iOS app hide the function, but it hides it in a place where people are specifically conditioned to not look. If you want to share something, you probably don’t want to un-clap it. If you want to un-clap something, you probably don’t want to share it. This is so extravagantly stupid I am now convinced Medium lets its interns design the UX.

Ok, but even that steaming raccoon dropping of bad design cannot compare with how Medium handles un-clapping on mobile browsers. You can’t. Simple as that. Well, you can, but you have to go to your profile, select “Claps”, find the article, click a down arrow, and then there will be an option to un-clap. But you certainly can’t do it on the article itself. And, if you’re flustered from having mis-clapped, you probably won’t think to go looking.

Of course, all of this crappy design is the result of the original decision to use the unique claps over the standard binary like model that everyone uses and nobody has complained about in the history of the internet (except when Netflix dumbed down their star rating system, but that’s another story for another day). Supposedly Medium plans to use this inane model as a basis for their profit sharing system, but it comes across to me as suspiciously twee. Oh well. I should just be thankful it wasn’t jazz hands.

Important lessons

• Don’t hide crucial functions
• Don’t put functions in the exact opposite place of where someone would look
• If you’re dead set on having an innovative design, hire damn good designers and not a bunch of clowns.

Want more of me?

After you’re done getting your head checked, you can find me at these places.

LinkedIn:
https://www.linkedin.com/in/jclauss/

More articles like this:
http://blackmonolith.co/publications

When finding the truth makes no difference to you, any way you look at it, you’re gonna get screwed. You lose.