Hacker News Reader

Agreed.

And another good example are the horrible 'mobile' styles for forum scripts and wikis.

Look at the SMF one for example. Or the similar one used on Proboards. On mobile devices, you literally can't tell simple stats like what topics have been recently posted in, which members are online or what the last post was in a section. In other words, all the basic things that make a community usable.

But hey, it looks clean, right?

Media Wiki is even worse. On mobile, you can't see any talk page links, any edit histories or any category lists/infoboxes for articles. So if you don't want to search everything or follow random links in articles, then you have no simple way to browse the site or keep up with what's going on in the community.

Taking away useful information and features like this doesn't make the site easier to use, it makes it more annoying. It makes it simpler for random 'guests' while making it more difficult for anyone who actually uses a site or service on a regular basis.

Ideally these extra features could be put at the bottom or tucked away under menus. But I think the assumption is that power users have a desktop.

> But I think the assumption is that power users have a desktop.

Which would also be an assumption that power users don't want to do things on the move using their smartphones. Sounds like a pretty bad assumption to me.

My personal favorite...

The Google keyboard on Android offers text recommendations and let's you tap backspace after typing a word with the swipe-style keyboard to delete it in every other app except the native Google search app interface.

Searching on Google with the Google keyboard is painful. This is all on Android 5.1.1 at least. Searches for this bug turn up lots of equally frustrated people.

Yeah the Google search in android is painful. Also in the subway with bad connectivity it's still possible to load facebook with gifs, pictures, texts etc. But a simple Google search is not possible.

There's countless times when I'm waiting for something and think "how crazy can this api possibly be?!"

I imagine some convoluted thing where they're passing authorization nonces around and getting normalized data so they have to make separate requests for each object id. (Eg: /business_name/:id, /descriptions/:id, /ratings/:id ... ) and bring it together client side... This isn't a request, it's a long conversation.

That's Facebook being clever and shuffling around posts it already had. That's not possible with a search query.

Really? Does this work in airplane mode?

I think this comes about because of that thing, you know, too many cooks spoil the broth.

Too many UX designers fuck everything up.

I just use my smartphone now to sent SMS, make and receive calls, and read HN. Because the user experience of doing anything else on it makes me want to throw it in the river.

I actually think it's the opposite. I've met many high-falutin designers who seem to care more about an abstract vision and theory then practical terms.

I've seen designers coddled and defended against avalanches of negative feedback under dubious auspices that users don't know what they want or that it's a natural thing to complain...

I envision whenever I see these redesigns that there's some Im-so-smart person on staff from like mit media lab that was given dictatorial power over the interface and features, had no actual idea what they were doing, but had too much of an ego to ever admit to it.

The real "most clever" thing you can do is be useful and get out of the way.

I've seen designers coddled and defended against avalanches of negative feedback under dubious auspices that users don't know what they want or that it's a natural thing to complain...

Then those aren't designers, those are artists. Artists have the option of creating a singular vision and defending it against all comers. Designers have to create a vision that matches the discovered desires of the users, and then ruthlessly road-test that vision through prototyping and structured user testing. If your "designers" are artists (or, just as common, you rely on your programmers implicitly being designers), then, yeah, you've got a problem, because they're focused on the product and not the process.

Incidentally, the sort of UX concepts you see on Dribbble are usually art, not design; most of them are created in isolation and outside of actual design methodologies, which is why they look great but are completely impractical for daily use.

I'm not a big fan of assigning identities to others. I've got disagreements on the culture.

If we didn't have this current culture and the older style continued, this mobile phone would have a physical slide out keyboard.

There may have been a physical scroll wheel on the side of the phone that could go through programs, scroll this page And do volume - not just that.

Maybe even buttons on the back of the phone like on the PlayStation controller to take selfies, dim the screen... If the phone was dropped, the replaceable industry standard battery would eject to absorb the impact. Cracked screens would be even more rare on top of that because the device would be a durable absorbent plastic.

The interface itself would have global text and keyword search, the color scheme would change at night, There'd be an sd slot and a dual band gsm/cdma chip etc...

I look at the trajectory we are currently on and I'm envisioning everything to be a toddlers interface in 10 years that's enormously infuriating and useless where you're cattle prodded into doing some very precisely planned flows with intentionally no room for deviation.

Nothing stupid is off the table. Replace the url input box in a web browser with a "share" button to post it on Facebook or Twitter? Google nagging you about new results to some random search from last week? News sites using third party cookies to only present articles on the front page that agree with your biases? A merging of sms and email so you have no idea how things are being sent?

I can see these happening in the Name of Design.

everything to be a toddlers interface in 10 years that's enormously infuriating and useless

No need to wait 10 years, you've just described what we have now. Sure, it'll be worse in 10 years, but what we have now is already so simple that toddlers do use smartphones.

What's infuriating is that there is no pro- or power-user version. I'd pay AU$1200 for the device you described above but not for a flagship smartphone from Gooapple.

don't know if you'll see this ... but I think that part of the reason that we don't have this is because nobody has made it "sexy" enough.

I think we've conflated user desires in things like the iphone as the clever marketing from Apple Inc. I think the effect of their marketing really eclipsed the quality and features of their product line and we've drawn a false equivalency that the methodology of their product line is what consumers actually want.

I have no empirical evidence to support this wacky theory however and I remiss that it's likely 80% wishful thinking.

> Designers have to create a vision that matches the discovered desires of the users, and then ruthlessly road-test that vision through prototyping and structured user testing.

I think this approach is actually part of the problem of the UX today. It ignores the fact that most of the time, users have no "desires" about the product, and that instead what user does and wants is directly shaped by their interaction with the product. It's discovery by looking it the mirror - you don't see what your users want (because they don't want much, UX-wise), you only see reflections of your own preconceptions.

My belief is you should focus on understanding the problems your target users want to solve with your tool, and then design UX around making the product a powerful and convenient tool for solving those problems. And if you think UX is your product, then I don't want to waste my time even checking it out.

My belief is you should focus on understanding the problems your target users want to solve with your tool, and then design UX around making the product a powerful and convenient tool for solving those problems.

I agree! And, when I talk about "design," UI and UX are only one facet (because strategy, content, and technology are their own components in the larger design space).

When I talk about "desire," I'm following Gause and Weinberg in Exploring Requirements, where they say, "[W]e don't know how to figure out what people need, as opposed to what they desire. ... We do observe, however, that by clarifying their desires, people sometimes clarify what they really need and don't need." In that sense, they are in agreement with your approach: determine the desires your users have (what do they want out of the product, what will the product help them do, etc), and then use that to inform their needs (here's what the product should specifically do or afford). Your experience designers then work with you on how to create experiences that effect the needs to address the desires.

What I'm saying is, I think, that you're already thinking at a very sophisticated level in terms of design, it just sounds like your experience with design has been with organizations at a lower stage of process maturity.

Thanks for the response. I haven't heard of Gause and Weinberg's book before, looks interesting. I'll take a look.

It also has a lot to do with how the features are presented. On mobile for example it's very annoying not seeing what you type, as it's easy to slip and make mistakes, so you have to make sure any input box get a lot of screen realty.

It's a tough balance. I guess the right way to is to have UI metaphors that minimize extra controls from scratch (i.e. if you remove 'save as' find a way to remove files as a metaphor; you can just have automatic save, undo, and 'duplicate this'.)

The way we currently have it has worked for at least 30 years and takes at most, a few seconds of explaining and everyone I've ever met understands the file metaphor.

There's this anecdotal story about how people in the early 80s couldn't figure out that the mouse should be intentionally rolled around on the table without a demonstration but once shown it was clear.

We shouldn't have some visceral animosity against something if it takes non-zero time to learn; especially if we are talking about under 30 seconds.

Many major common things take months of time to figure out; literacy, basic arithmetic, cooking, any mode of transportation, relationships, temper, finance, and time management; infants even take a significant time to figure out sleeping.

Anyway, what we have is fine - no further simplification required.

If we followed that idea though (keep it as it has been for 30 years) then Google Drive or iOS would have you browsing through hierarchical folders like Windows Explorer.

I think no.

I think they changed because search became affordable enough to become a valid strategy, not because some genius figured out people are confused by folders.

It changed the up-front effort in FILING for a greater but later effort in FINDING (assuming keyword search).

To compensate, of course, users can invest a greater upfront effort in NAMING and TAGGING.

Several decades ago, I made filing mostly go away by creating files in the directories in which they would eventually be filed. This requires the ability to create nested subfolders, without which no substantial filing system is viable.

There's probably an order of magnitude at which hierarchical filing systems become the only rational approach (depending on the material). It might be 1,000 in some cases, or 10,000 in most cases, and maybe 100,000 in a few cases.

At least, that's how it works for me.....

That would be wonderful!

I've had many non technical people ask me how to create folders in those systems...

People can live with a flat file structure; Early versions of DOS were like this. Maybe we should go back to line editors too... So much prettier and simplified! No menus or dialogs! You don't even see a pageful of content at once! Just a blinking cursor and some brief commands. No messy interface...

(I can unfortunately actually see this happening)

It's a definite balance though, is my point. You have to make these determinations all the time while making consumer-facing software.

For example, I'm making a 'Newsfeed for Twitter' right now (top tweets people you follow have liked/shared) and people I show it to want keyword muting so the feed isn't dominated by Trump news. From a UX perspective I really resist adding keyword muting. Instead I added a 'Tailored' mode that shows top items your friends are talking about that aren't being talked about across the network.

Similarly I had "Top -> 12 Hours | Day | Week" and to remove the "12 Hours | Day | Week" submenu, I created a feed that algorithmically balances popularity and recency.

So I'm saying these choices are ever-present when making software. Some moves like Apple removing the 'Escape' key seem like random silliness but some hard choices are important when making a good product.

I admire your effort here but I really think that such concerns were not shared equally.

Stupid and direct interfaces that "any rank amateur could do" seem to catch on pretty well.

Regardless, I sincerely wish you the best in your project. Send me a link when it's ready and I'd love to take it for a spin.

That would be amazing. As it is now, I have to search for stuff and I no longer have a clear picture where stuff is and just how much stuff I have. Therefore, when I'm searching for a document and it does not come up, I'm not sure if it's because it doesn't exist, or just because I did a wrong query.

Yes, an explorer.exe interface for Google Drive would be a huge win. That's also why I avoid using it, and keep everything I can in Dropbox instead.

The github mobile site doesn't have search. If your mobile site doesn't provide all the features your desktop site does, you're asking me to make a choice between features and usability - and I'll usually choose features.

Whenever I use github on mobile I feel like they very intentionally designed the site for someone who isn't me; a bunch of solutions to problems I've never had.

I agree with you 100%, but I think you are conflating two different things. What number 12 talks about is having too many choices for one thing. E.g.,

* Too many themes

* Which of these 10 apps would you like to open the text file with?

* New Word document, what kind of document (lists 40 choices)

What you are really talking about is #34. Designers trying to make it simpler for mobile end up reducing functionality.

> in Google mobile search, you can't do custom date ranges...

In fact you can. Just swipe the menu bar to the left to access Search tools. (I admit it took me a while to discover this)

not on google native app. it's available in web app.

Nope. Still not possible. Here are screen shots showing what I'm talking about in this Twitter account I save for just the proper occasion: https://mobile.twitter.com/FuckYourUI/status/753547054236872...

Ah yes, same here. (Note to self: do not read HN with a hangover)