What Comes After User-Friendly Design? (fastcodesign.com)

Kelsey Campbell-Dollaghan, writing for FastCoDesign: "User-friendly" was coined in the late 1970s, when software developers were first designing interfaces that amateurs could use. In those early days, a friendly machine might mean one you could use without having to code. Forty years later, technology is hyper-optimized to increase the amount of time you spend with it, to collect data about how you use it, and to adapt to engage you even more. [...] The discussion around privacy, security, and transparency underscores a broader transformation in the typical role of the designer, as Khoi Vinh, principal designer at Adobe and frequent design writer on his own site, Subtraction, points out. So what does it mean to be friendly to users-er, people-today? Do we need a new way to talk about design that isn't necessarily friendly, but respectful? I talked to a range of designers about how we got here, and what comes next.

Comments

GetRidOfSubject

[Anonymous Coward]

Khoi Vinh, principal designer at Adobe

lost interest right there

Re:GetRidOfSubject

[Anonymous Coward]

who?

Re:GetRidOfSubject

[Anonymous Coward]

another Dieter Rams wannabe like Jony Ive...

Re:GetRidOfSubject

[Anonymous Coward]

Yep, especially if he is the mastermind behind Adobe reader DC, the biggest downgrade in software usability since Microsoft introduced ribbon. Previously the reader was just doing its job, now the screen is filled with whitespace and random icons and one needs to use google to find the functions which previously were easily available in menus.

Re:GetRidOfSubject

[thegarbz]

lost interest right there

Indeed. GIMP has a much better user interface...es

Re:GetRidOfSubject

[Anonymous Coward]

Well, kinda.

A lot of what makes Photoshops user interface usable is that people are used to it.
If you jump into it as a new user it isn't really better than GIMPs.

OTOH change for the sake of change should be considered bad user interface design.

Re:GetRidOfSubject

[Anonymous Coward]

Indeed. GIMP has a much better user interface...es

Sarcasm? Honest question, your statement could be interpretted two very different ways.

Personally I find gimp featureful but awkward to use. I want to like it, but I seem to spend most of my time googling how to do things, whereas I can usually just figure it out myself with photoshop.

Re:GetRidOfSubject

[Mattcelt]

As someone who does not regularly use Photoshop or GIMP, I find them equally counter-intuitive and confusing. I think they both require a bit of syncing on the user's part with the mindset of the developers/UI designers. Once you're locked in to their way of thinking, it generally becomes much more obvious why things are arranged as they are.

Apps!

[Anonymous Coward]

Only apps can app apps, NOT LUDDITE users!

Apps!

Re:Apps!

[Anonymous Coward]

From now on I am modding up Appy App Guy +1 Funny in all discussions when I have mod points. Why? Because the glorious womynjewnïgger never posts in the threads where we so desperately needs AppFishMan UNTIL NOW.

Re:SJW design

[Anonymous Coward]

I would suggest monotonous command line editing of config files. I respect people who do that shit.

SANITY!

[Anonymous Coward]

Prepare to be dazzled by ads from now on. YOU heard it HEAR. Tell all your friends. Tell all your bullies. Shout it from the rooftops. Then buy more shit you don't need, won't use.

Good equals simple

[petes_PoV]

The most "user friendly" design is one that does the right thing every time.

It is not one that gives you half a dozen options (all equally badly described:: telling the user what they do, not what their effect will be) for half a dozen more operators. It will have intelligent defaults - possibly ones that vary, depending on circumstances. It will provide a clear workflow: top - bottom, left-right, corner to corner -- whatever, it will be CLEAR what to do first, next and to finish.

Having said that, it will still be possible for users to make their own decisions. A good design will not railroad a user into one single path, one single process or one single methodology. If it did, there would be no point providing a user interface.

I look forward to the day - at the rate of progress, in the dim and distant future, when user interfaces work like this. Without any "I have just crashed and wiped out all your work. OK" style messages,

Re:Good equals simple

[K. S. Kyosuke]

In those early days, a friendly machine might mean one you could use without having to code. Today, to me, a friendly machine means something you can fix by coding since it inevitably starts out broken (and not doing the right thing for me).

Re:Good equals simple

[goose-incarnated]

The most "user friendly" design is one that does the right thing every time.

The Right Thing!(tm) differs from person to person, and may even change for a single person as circumstances change.

For example, a menu-driven program in domain $FOO is great for a novice in that domain but as that novice turns into an expert in domain $FOO they will prefer using shortcuts and muscle memory for common tasks.

For novices an exploratory interface is great - it allows them to learn the limits of what can be done. For experts a command interface is better - they already know what can be done and the command interface allows them to apply muscle memory to get things done.

Anyway, this statement:

It will provide a clear workflow: top - bottom, left-right, corner to corner -- whatever, it will be CLEAR what to do first, next and to finish.

contradicts this statement

A good design will not railroad a user into one single path, one single process or one single methodology.

Finally...

I look forward to the day - at the rate of progress, in the dim and distant future, when user interfaces work like this. Without any "I have just crashed and wiped out all your work. OK" style messages,

I look forward to the day that user interface designers read "The design of everyday things" by Donald E. Norman and use it as a checklist against their designs before unleashing crap like Metro and Gnome 3 on the general public with the poor attitude of "You All Are Too Stupid To See The Greatness Of Our Design"(tm).

Re:Good equals simple

[Krishnoid]

The most "user friendly" design is one that does the right thing every time.

... and the premise for most cheap dystopian AI sci-fi.

Re:Good equals simple

[Kjella]

The Right Thing!(tm) differs from person to person, and may even change for a single person as circumstances change.

Agree, most complex tools have lots of paths and it's entirely unclear where to begin but then again the application has no clue what you're trying to accomplish. I'm thinking of applications like Photoshop, Visual Studio, Excel, Notepad++, Resolve and a whole lot of others. Many, many layers of menus, toolbars, dialogs, tabs, window areas, settings, options and so on. I think that past a certain complexity there's no such thing as a particularly great one-fits-all design. So my pet wishes:

1. Let me easily move things around. Like if I want to re-dock the windows, resize them etc. I can do that.
2. Let me easily collapse/remove things I don't need. Or better yet, hide the less used options with an expander/under an advanced button.
3. Give me a usable way to search for functionality instead of digging through menus and reading tooltips
4. Offer some kind of preview/sample functionality where relevant. I'm not always sure exactly what to do.
5. Proper undo/redo history or at least be explicitly clear on what can't be undone.
6. Auto-complete/suggestions from past entry where possible/relevant, but please no assistants.
7. Control over what changes/defaults are saved, like do I want the file dialog to start in the most recently used directory or the one configured.
8. Try being consistent about how things work, avoid unexpected side effects, be clear in naming.
9. If it's in the nature to be scripted, I love GUIs that build a command line I can copy and save.
10. Don't make change for change's sake. At the very least offer a "classic" interface, don't force people.

That would be a good start.

Re:Good equals simple

[Anonymous Coward]

A railroad doesn't let you choose what path you are going to take -- you can only go in the direction that the rails allow. There are ways to branch from the current track, but that branch must be selected before you get there.

Contrast that with a design which guides you in the intended direction but does not require you to take a specific path.

One of the guiding principles of the C# programming language is that it should be easy for its users to "fall into the pit of success". It's not that you can't do hard/bad/complex things, but the right thing should be the easiest thing. Contrast creating an array in C++ vs. C#:

In C++ you can quite trivially create a pointer to an array of ints, but it's also trivial to manage the memory improperly and get all kinds of bugs (security and otherwise). If you want a safe array you have to import a vector library (Boost? STL?) and learn how to use it; the library you choose may not be compatible with that used by a 3rd-party library.

In C#, the default array mechanism is safe. You never have to worry about managing its memory or overrunning the buffer. You can have an unsafe pointer to an array of arbitrary memory, but it's somewhat harder to do and most C# programmers probably don't even know about it. That's the "pit of success" because the easiest thing to do is the right way to do it.

A good design will make the most likely intended result the easiest, while still allowing unintended uses to be possible.

dom

Re:Good equals simple

[justthinkit]

Mostly I think you are on the right track Kjella. However

"3. Give me a usable way to search for functionality instead of digging through menus and reading tooltips"

You say you want to "search" instead. What the [] does that mean? Your idea of "searching" is going to be different from everyone else's.

Whereas a menu system is the perfect way to nest functionality. Not unlike a dictionary. Use a word/label to identify -- often with already understood language -- what something will do if you click on it.

Microsoft moved us away from menus...but the move (like the move to thumbprinted touch screens) made no sense.

Also, FWIW, a lot of what you are listing here (e.g. 7.) is best handled in an extensive -- and well designed -- option system. I've never seen a program do it better than Eudora.

Re:Good equals simple

[Anonymous Coward]

Great Recommendation. You deserve the upvotes.

Re:Good equals simple

[jandersen]

For example, a menu-driven program in domain $FOO is great for a novice in that domain but as that novice turns into an expert in domain $FOO they will prefer using shortcuts and muscle memory for common tasks.

For novices an exploratory interface is great - it allows them to learn the limits of what can be done. For experts a command interface is better - they already know what can be done and the command interface allows them to apply muscle memory to get things done.

Exactly - which is why it makes a lot of sense to design GUIs as thin(-ish) front-ends to a collection of commands, at least as a guiding principle; the real functionality should be in the back-end - the system commands, the relational database etc - not the GUI. Also, the term "user friendly" is too vague, and should be replaced by something like "functional".

Re:Good equals simple

[Waccoon]

It is not one that gives you half a dozen options (all equally badly described:: telling the user what they do, not what their effect will be) for half a dozen more operators.

Unless that's a program requirement.

User friendliness depends on the tasks you're trying to perform and the type and skill level of the user (including developers, who generally need a million options). Sensible defaults are a must, but you can't just make a single UI, built for the lowest common denominator and based on marketing telemetry, and call it a day.

Re:Good equals simple

[AmiMoJo]

11. Let me export those settings, so that when I install the app on another machine/VM I don't have to spend half and hour reconfiguring it.

Re:Good equals simple

[AmiMoJo]

These days "user friendly" means "doesn't artificially limit what you can do with DRM".

Re:Good equals simple

[bickerdyke]

It will provide a clear workflow: top - bottom, left-right, corner to corner -- whatever, it will be CLEAR what to do first, next and to finish.

How does it know what the user WANTS to do next? It should be the user telling the machine what to do next, not the other way round, or you're switching who is the "tool" here.

A computer should automate a workflow, to free users from repetitive tasks and NOT guide a user through a fixed workflow over and over again.

Who Cares?

[Anonymous Coward]

As long as it's different! Those people that do all the training on all this new bullshit, and the people who write all the shit articles about them - they need to eat too, don't they? If we stay with the same stuff that works, those people would be out of a job, and we can't have that!

Re:Who Cares?

[Anonymous Coward]

You're blaming the wrong people. Those people wouldn't have a job if the programmers didn't feel the need to slap another layer of abstraction onto what was perfectly functional and clear before, opening new and heretofore undreamed of security holes.

design-friendly user

[turkeydance]

designing friends for the user?

Re:SJW design

[Anonymous Coward]

I believe the 'respect' that's being referred to is the type of 'respect' where you don't track the user, force them to do things they don't want to do right away (looking at you, windows update), or otherwise coerce the user from the position of 'we, the designers, know better than our users, who are dumb'.

You know. Actual respect.

We Aren't to the Friendly Part Yet

[Puls4r]

Computer Programmers (I'm looking at you microsoft) need to learn that these things are important:

1. Appropriate icons
2. Minimizing clicks to task completion
3. Common control placement
4. Self-explanatory menu trees
5. Consistent menu trees


I'll give you a good idea how NOT to do it. Windows 10 is a mashup of numerous operating systems. You'll find control panels from the original 95, and new 'tile' or web-page-like looks woven together. You'll find some with buttons you push, and others with highlighted words you need to click. You'll find important features like configuring the lock-screen not under right-click display like you would expect, but buried deep inside the user-accounts system. And clicking to find what you want has gotten so counter-intuitive that most people utilize the typing in the search box to pull things up now.

I could keep going on, but Windows 10 is a prime example of how non-user-friendly programs from tier 1 vendors have become. Photoshop is a right up there too.

Re:We Aren't to the Friendly Part Yet

[Puls4r]

Oh, and as a follow up. You know the auto-industry has huge focus groups (usually done by email) where they present icons to people, like a trunk opening button. But they don't tell the people what it does initially - they ASK them what they think it does.

Because if you have efficient button and icon design, you don't need menu trees and your dependence on language (and all the misunderstandings it creates) is decreased by an order of magnitude. Just have a British Person ask an American where a boot is if you want a good example.

Re:We Aren't to the Friendly Part Yet

[viperidaenz]

heh. you picked a bad example.

You'll find important features like configuring the lock-screen not under right-click display like you would expect, but buried deep inside the user-accounts system

Yes, you'll find a security setting inside the user-accounts system.
Sure, it's semi-related to a password protected screen saver. It has nothing to do with a screen saver though, or display resolution, orientation, multiple screens, display adapters, or anything else to do with the display settings.

I agree with what you're saying though, Windows is like every other large piece of software. It's been changed a lot over time. It's full of new things that don't quite match up with the old things.

Re:We Aren't to the Friendly Part Yet

[Tablizer]

I would kind of forgive Windows for bad UI design IF they stop moving shit around for each release. When they shift stuff around it doesn't appear to solve anything, just be a different kind of random. Same with the damned ribbon: it's still half-hazard, just a different half-hazard. Didn't improve my productivity over the old tool-bars (except where they fixed bugs).

Consistently bad is better than inconsistently bad. Don't move my moldy cheese, for I've memorized the mold pattern.

Re:We Aren't to the Friendly Part Yet

[JohnFen]

Didn't improve my productivity over the old tool-bars (except where they fixed bugs).

The ribbon is a great example of truly terrible UI design. It didn't improve my productivity, it decreased it. To this day, it remains a speed bump.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Haphazard, dingus.

Re:We Aren't to the Friendly Part Yet

[Tablizer]

^ user-unfriendly grammar suggestion

Re:We Aren't to the Friendly Part Yet

[140Mandak262Jamuna]

And clicking to find what you want has gotten so counter-intuitive that most people utilize the typing in the search box to pull things up now.

And all one has to do is to stick in a c:\> and make the search box 24 lines by 80 columns and you are back to DOS!

And most X11 window managers had a tiny "mini consoles" way back in 1990s!

Re:We Aren't to the Friendly Part Yet

[tlhIngan]

You'll find control panels from the original 95, and new 'tile' or web-page-like looks woven together.

No, what makes Windows 10 especially bad is having TWO control panels, each doing a different thing.

You have the old style Control Panel that's been around, and the new style Settings control panel. And each seems to have a counterpart in the other, but each does not do what the other does. It's like Microsoft began converting the Control Panel to Metro style, ran out of time and we're left with a mish-mash of both, neither of which does everything, and each having their own set of settings.

It's like a car with two steering wheels, two dashboards, two sets of pedals, etc. Except the left steering wheel only goes left, the right steering wheel turns right only, the left dashboard has the speedometer, the right dashboard has the tach and other gauges. The left pedals have the brake, the right pedas the accelerator, etc.

It's an exercise in confusion and frustration.

Hell, Apple is no better. On iOS, some app settings are under, well, Settings, while other app settings are under the app's options page.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

It's on the wheel.

Re:We Aren't to the Friendly Part Yet

[Solandri]

You know the auto-industry has huge focus groups (usually done by email) where they present icons to people, like a trunk opening button. But they don't tell the people what it does initially - they ASK them what they think it does.

That's very interesting to hear. A friend called me for help saying "the naked butt light" on her car had turned on. (It's a low tire pressure warning.)

It also took me a while to figure out what the inductor light was. (It has nothing to do with inductors. It's a light to tell you the glow plugs in a diesel engine are still warming up.)

Re:We Aren't to the Friendly Part Yet

[angel'o'sphere]

I hate the ribbon.
I never used tool bars, I use the menus or shortcuts.
Toolbars simply need to long to show the tooltip to indicate what an icon is doing ... takes sometimes minutes to get the functionality you(I) need.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Better than Hulu's new interface.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

There is actually a reason behind the old control panel still existing that has nothing to do with laziness or schizophrenic design. Install a Synaptics, Dell, or Elantech touchpad driver and the Mouse control panel has an added tab that has touchpad-specific configuration. Windows 10 has some of the touchpad settings but stuff like Synaptics ChiralMotion or dead zones or PalmCheck level are not present, so it's necessary to keep the old framework around (even if only by some convoluted click path) to keep those configuration tools accessible. Yes, it's incredibly fucking stupid (no excuse for not having standard stuff like "pointer options" in the Settings panel somewhere instead) but there is a perfectly legitimate reason to keep it anyway.

Personally I'd prefer full access to both styles even if they control many of the same settings. That way I can use the one I prefer and am used to while the "newer is always better" crowd can use the other one.

Re:We Aren't to the Friendly Part Yet

[edtice1559]

I don't know whether they do these email focus groups or not, but if they do, they aren't very effective. My wife's car has a picture of a tire tread with an exclamation mark that indicates low tire pressure. Looks confusingly similar to a thermometer in water so you may think you have an overhead. The check engine light is useless if you have never seen an engine on a test stand. Honestly text would work way better than icons for these interfaces. A 16 character display with an actual text message would take up less space and convey more information. Of course it's not language-neutral which is a disadvantage but it's much more expressive.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

> Same with the damned ribbon

Came here to post that; you beat me to it.

Coincidentally I was searching and reading about the old slash ("/") menu key on Visicalc/Lotus 1-2-3. And there were other examples, like the very excellent xtree gold.

That's a kind of usage lots of people don't understand, halfway between CLI and graphical. An experienced user needs not to think about commands: it's an experience next to driving, much like expert Vim users probably do (not my case... yet!).

Like JohnFen says below:

> The ribbon is a great example of truly terrible UI design. It didn't improve my productivity, it decreased it. To this day, it remains a speed bump.

IMHO, it's by design. It would be a big turn off for some incompetent dude seeing folks working fast in Excel or Word, so I suppose they made it slow for everyone. If it was easy, people would have time to learn about the software from the competition. It is a way to take up all the user's time, much like large files take up the whole disk.

At least, that's the way it seems to me.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Portmanteau of half-assed and haphazard, dweeb.

Re:We Aren't to the Friendly Part Yet

[doom]

I think it's funny so many people here know what Windows 10 is like. (I know, I know, "I have to use it at work", right.)

Re:We Aren't to the Friendly Part Yet

[Voyager529]

Because if you have efficient button and icon design, you don't need menu trees and your dependence on language (and all the misunderstandings it creates) is decreased by an order of magnitude. Just have a British Person ask an American where a boot is if you want a good example.

This comment brought to you by someone who has never had to give directions over the phone, or try to figure out what the person on the other end is looking at as the two of you play an impromptu game of inverse pictionary. Icons are just fine, but icons + labels are better and have very few downsides.

Your example is terrible because most British cars have the driver side and passenger side in a reverse layout from American cars. If you're doing that, you can relabel the error light 'boot open' instead of 'trunk open'.

Re:We Aren't to the Friendly Part Yet

[Tablizer]

[Same with the damned ribbon] Came here to post that; you beat me to it.

Because there's a quick-post option on my Edge ribbon

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Sort of, the riggon is an example of why you need to sanity check the research data you're working with.

The most usable UI designs I've seen are vi, AutoCAD and Vuescan. They handle things a bit differently and the first two have a considerable learning curve to them, but they're immensely efficient once you've figured out the basics.

The ribbon was a stupid idea because you're not really supposed to be opening up the menus constantly, you're supposed to memorize the short cuts for things you do regularly and use those. For things that you don't do regularly, you should be able to find them based upon the menu and submenu with relatively little effort. The first thing I did with Word XP when I was using it was turn off that stupid smart menu system that arbitrarily hid entries.

Re:We Aren't to the Friendly Part Yet

[sanf780]

The ribbon in MS Word is MS way to enforce how to use the tool. The tool is so complex, it needs various modes of operation. The default mode is all about writing text and applying styles (note: I used other tools like Lyx and Adobe Framemaker that favour the use of styles, and I hate the notion of letting the users cherry pick fonts without using styles). The review mode is all about adding Notes and tracking changes.

Going back to the OP, it is hard to cram every function a tool does these days in a GUI. I hope no solution resembles Google Docs: instead of showing GUI elements, the user is left with a search widget. It reminds me of the Tom Hank's movie where he is playing one of those adventure games of the 80's and the protagonist had issues discovering which command to type. My proposal is to have focused workflow based GUIs that are highly specialized. I have seen this implemented in expensive tools, and it works.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Not everyone hates on something before they tried it.

Re:We Aren't to the Friendly Part Yet

[KozmoStevnNaut]

I have Windows 10 on my desktop and on my laptop, by choice for anything I can't run in Linux. Sure, W10 annoys me in various ways, but it's not nearly bad enough to make me switch back to W7 (which also has a bunch of inconsistencies).

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

>> [Same with the damned ribbon] Came here to post that; you beat me to it.

> Because there's a quick-post option on my Edge ribbon

Which nobody else has, as other ribbons aren't configured to have such icon.

Take that!

Re:We Aren't to the Friendly Part Yet

[AmiMoJo]

I must be the only person on Slashdot who likes the ribbon and finds it to be faster than hunting through menus. Maybe my ability to memorize the location and function of a large number of text items in apps I don't use often is below average, but I find visually searching for the thing I want much faster.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

The problem that ribbon UI solves is how to make thousands of commands available to the user. You could argue that nobody needs more than probably 50 of Word's 2,000 commands, and you'd probably be right.

Unfortunately everybody needs a different 50 commands than everybody else, making it impossible to have a simple UI just showing the 50 commands every user needs. Thus, MS needed to make a UI that allows every user to find the command they need without having to search an intractable menu tree.

dom

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

The problem that ribbon UI solves is how to make thousands of commands available to the user. You could argue that nobody needs more than probably 50 of Word's 2,000 commands, and you'd probably be right.

Yeah, that's probably a sign that your app is trying to do too much, and should be split into multiple programs.

Remember the Unix philosophy (or one aspect of it): do one thing, and do it well?

Yeah, nobody else does either.

Re:We Aren't to the Friendly Part Yet

[bickerdyke]

The ribbon in MS Word is MS way to enforce how to use the tool. The tool is so complex, it needs various modes of operation. The default mode is all about writing text and applying styles (note: I used other tools like Lyx and Adobe Framemaker that favour the use of styles, and I hate the notion of letting the users cherry pick fonts without using styles). The review mode is all about adding Notes and tracking changes.

I guess that's the problems most users have with ribbons: Typing and cherrypicking fonts is all they do! Like when was the last time you saw Joe Sixpack use the notes and change tracking?

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

Not sure why you chose Photoshop, I use it daily and for the most part it's quite intuitive. I use shortcuts easily for my most common tasks and lesser used features are generally where I expect them to be. There's a few things that are aggravating so it's definitely not perfect, but no where near the Windows 10 level of annoying. Also between versions the UI doesn't change drastically so even the parts I don't like remain mostly consistent so I don't have to relearn much when I upgrade, again no where near the level of change and moving things around Windows versions have had.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

These lights are more correct than you think.

A glow plug really is shaped like a coil. (and it is an inductor too, because of its shape.)

The tyre pessure light means you need to get a mechanic to fix the tyre. When he bends over to reach it, you see some of his butt.

Re:We Aren't to the Friendly Part Yet

[JohnFen]

The problem that ribbon UI solves is how to make thousands of commands available to the user.

I argue that it doesn't solve that problem. Or at least solves it in way that is worse than more traditional methods.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

The tyre pessure light means you need to get a mechanic to fix the tyre. When he bends over to reach it, you see some of his butt.

Yep, by using an icon, they save themselves the trouble of translating 'buttcrack' into a dozen languages.

Re:We Aren't to the Friendly Part Yet

[narcc]

My primary objection to the ribbon isn't related to using it personally, but to telling others how to perform some action. Helping someone find an little-used feature by walking them through the menu tree seemed to be simpler than doing the same with the ribbon.

Re:We Aren't to the Friendly Part Yet

[JohnFen]

Maybe my ability to memorize the location and function of a large number of text items in apps I don't use often is below average

Actually, this is the main problem I have with the ribbon (aside from it wasting a ton of space) -- I can never remember where things are in the damned thing, so it takes me forever to hunt through and find the thing I need.

Re:We Aren't to the Friendly Part Yet

[AmiMoJo]

I just tell them to ask Clippy for help.

Re:We Aren't to the Friendly Part Yet

[cellocgw]

Same with the damned ribbon: it's still half-hazard,

That wins the "I heard it but never saw it in print" award of the month.

haphazard

Tho' it's a nicely coined new word. Except that the Ribbon is a full hazard.

Re:We Aren't to the Friendly Part Yet

[Anonymous Coward]

A portmanteau should only be used when it's clever or helpful, and not confusing.

Re:We Aren't to the Friendly Part Yet

[datavirtue]

Clippy is behind the ribbon pushing pedals and pulling levers while laughing hysterically.

Re:We Aren't to the Friendly Part Yet

[datavirtue]

If Windows 10 annoys you then you need to get out of the house more.

Re:We Aren't to the Friendly Part Yet

[KozmoStevnNaut]

Every version of Windows has its share of annoyances. We mostly just learn to work around them.

Re:We Aren't to the Friendly Part Yet

[Tablizer]

uh, it was (intended as) a joke

Re:We Aren't to the Friendly Part Yet

[MercTech]

Automotive displays are made to look pretty not be informative. If the on-board computer is going to generate an error code, why not have a lookup table on an eprom in the display circuit so it actually tells you what the code means?
    Oh, wait, that would cut the dealership out of $180 for putting a $29 code reader on the under dash interface plug.

Re:We Aren't to the Friendly Part Yet

[MercTech]

And, with the example given, just like you can change display measurements from miles to kilometers you could shift from "American English" to "UK English" or to Spanish or whatever. I'll be there would even be an after market for Klingon language proms.

What

[GLMDesigns]

We need to keep understanding how people use (and expect to use systems). Nothing has changed. Do you want to change the name from Information Architecture to UX Design. Whatever (even though I think Information Architecture) is more descriptive and ought not be reserved for HCI (Human Computer Interaction) or for people doing Environmental Psychology.

What is the f**king use case that is better covered by "respectful" as opposed to "user-centric." I actually think that user centric is, in today's word, fairly close to an irrelevant qualifier. Who else is one doing design for? To enhance the prestige and psychic well-being of department heads and CEOs?

ooops. better not mention that else we'll have a new discipline called CEO Design.

Re:What

[Tablizer]

You can have all the UI research in the world, but a clueless PHB or marketer will likely override you with some stupid fad or whim. Science doesn't work on idiots and big egos.

Re:What

[JohnFen]

Who else is one doing design for?

In an ever-increasing amount of software, the impression that I get is that the interface was designed for the designer. Certainly not for the user.

Re:What

[GLMDesigns]

That tends to be bad design then.

A good design should allow a domain expert to use the system with little to no training.

Systems are to be designed for the people using them, not to have some shiny example in a portfolio. /rant

Re:What

[bickerdyke]

Ignoring research is either stupid or genius. Apple got most of its mojo from ignoring conventional (or rather: historical) UX guidelines and not from new technology. Like the Henry Ford quote, that if he listened to market research, people would only have asked for a better horse.

Re:What

[Tablizer]

Steve Jobs lied, Apple actually did UI testing.

Re:What

[bickerdyke]

Well I never said they didn't.... they probably rather knew when to ignore the results

Re:What

[Tablizer]

There is a difference between brand new ideas and general UI design. Most PHB's making dumb UI decisions are not on the cutting edge, but rather just making dumb decisions on something fairly ordinary, and not being curious about what real users may think of it.

The touch-screen smart-phone may have been a new idea (or at least a rare one) when the iPhone was designed, but Apple probably still tested beta iPhones on actual users. And Henry Ford probably tested actual cars with actual drivers. In other words, go ahead and leap-frog existing ideas, but still test those leaps with actual users.

There's a new buzzword, complete with acronym

[viperidaenz]

UX: User Experience

Re:There's a new buzzword, complete with acronym

[Tablizer]

"Market-Optimized Synergized Experience Management"

Re:There's a new buzzword, complete with acronym

[Anonymous Coward]

"New buzzword?" The first documented use of the term was in 1986, and it's been used commonly for at least 20 years. "UX" has been used as an abbreviation since at least 2005.

Re:There's a new buzzword, complete with acronym

[Misagon]

Yeah, but the use has shifted. It used to stand for everything around a user and a product: from how you learn about it, how you buy it, how it is packaged, how you install it, how you use it, how you upgrade it, how you get support for it and how you get rid of it.

These days, people use it to mean "user interface design": Just one part of the whole.

Re:There's a new buzzword, complete with acronym

[Anonymous Coward]

UX used to mean all those things, now it is basically concerned how to make a desktop application look like a phone application. The target audience is also some mythical new user, who has never used or seen computers in his life.

In old days applications followed OS' style guides, so if one had used a single application in a OS, he could learn to use another easily as they all used the same metaphors and similar visual clues. Now each application looks like a DOS game where user interface elements are randomly distributed and user has no clue, what each element does.

Give-The-Users-What-They-Deserve-Design

[Dirk Becher]

GTUWTDD!

Next

[Geoffrey.landis]

"What Comes After User-Friendly Design? "

As far as I can tell, user-hostile design does.

Re:Next

[wafflemonger]

When you read the article, there is a video at the bottom that autoplays. If you pause the video and scroll so that it is off the screen then scroll back, the video picks back up and starts playing again.
I guess the author is testing user hostile design on us to see what we will do.

Re:Next

[Megane]

Fortunately, I practice "hostile-user design", and I already had the video player site manually blocked in ABP, for being a repeat offender.

The next step?

[GrumpySteen]

Reduce contrast and move everything around in order to maximize how pretty the interface is without any regard for how it affects users.

You think I'm kidding, but I'm not. Companies left and right are jumping on board with the Internet Of Things idea of using websites and smart phone apps to control everything. Those interfaces are increasingly dictated by design idiots who care only how pretty the interface is. The end result will be a generation of truly shitty interfaces that barely work, but look pretty in the ad copy.

Marketing-agent-embedded software..

[Anonymous Coward]

See: Large-company games being developed/released.

They're increasingly marketing-heavy, while still increasing development budget... but they're using that budget to create pay-walled content INSIDE their full-priced software.

Basically, they're converting the software from a product, into a virtual salesperson, asking for more money.

Now, there's a landslide of jargon you could attach to this, like 'monetizing' and 'empowering', and 'whales' - but it's all basically just again, having the software itself act as a marketing agent, rather than a product.

When I went to a developer conference (PAX Dev) a few years ago, the folks there were already sick of it - but knew it was what they'd be asked to do, even on small teams.

The whole thing is a bit crazy though - like mutually assured destruction.

The audience size in general is increasing over time, the profit potential even for small groups is as good as it has ever been. This whole movement is largely based on fear of losing out - but it was never really needed before, and it's largely management demand rather than real need that drives this trend.

All someone has to do to reverse this trend is release quality software that provides what these products do, without the money-grubbing aspects, and the 'marketing-agent' mentality will die back, by the force of the greatest marketing tool, word-of-mouth (or social media).

For the time being though, the investor demand is, as always: Grab for everything you can, and the large guys comply with everything they have. I've been in the meetings - the software developers will do what they're asked, and the sad results are about what you'd expect.

After?

[JohnFen]

I wouldn't mind more user-friendly design in the current day.

In my book, a perfectly designed UI is one that lets me do what I want to do without me noticing the UI at all.

Note that this isn't an endorsement of the current trend of "minimalism" -- which largely accomplishes the opposite of remaining invisible.

Basic and expert modes

[myid]

Have a "basic mode" for beginners, and also an "expert mode". In the expert mode, you see buttons and menus that let you fine tune the app's behavior. (The expert mode's buttons and menus would confuse a beginner, and clutter up the beginner's screen.)

Also include a button that resets all preferences to their defaults.

Re:Basic and expert modes

[sexconker]

What about retard mode and normal mode?
Or APP mode and LUDDITE mode?

Re:Basic and expert modes

[ChrisMaple]

Also include a button that resets all preferences to their defaults

I just spent 2 hours setting preferences to where I want them. Now I'll press the OK button -- oops! Reset All. It needs at least a warning and an "Are you sure?"

Re:Basic and expert modes

[dgatwood]

Why do we even have that button?

No, seriously. Resetting preferences is a silly concept. It almost never fixes anything. When it does, the misbehavior is still a bug, and will still recur if you somehow manage to get your preferences set back to exactly the way they were. Worse, when it does, you've lost the state information that would help developers figure out what is wrong, which means that if you don't manage to get your preferences set to the same state, the bug will end up biting somebody else down the line.

Why does that button even exist? Nuking your preferences should be a last resort way of getting back up and running, and such last resorts should be as painful as possible to discourage their use as much as possible.

Re:Basic and expert modes

[Anonymous Coward]

Also include a button that resets all preferences to their defaults

I just spent 2 hours setting preferences to where I want them. Now I'll press the OK button -- oops! Reset All. It needs at least a warning and an "Are you sure?"

Spot on. Having a button for this is a bad idea. Having a menu item and dialog for it is much more forgiving.

Re:Basic and expert modes

[doom]

Have a "basic mode" for beginners, and also an "expert mode".

And you're well on the way to re-inventing the Wordstar design.

(By the way, anyone who thinks the kids-offa-my-lawn joke is still funny is clearly a geezer lost in the late-stages of senility.)

Re:Basic and expert modes

[Moof123]

Cute.

We have Luddite/retard mode. It's called the Ribbon interface. It sucks.

A basic problem we have is that there are industry standard programs like Word/Excel that need to cater to too broad of a spectrum of users. Folks who need to do nuanced stuff need more features than the kid doing his homework. Frankly Word peaked in usability around Word 4.0/5.1 on the Mac side, and has been a too complicated mess ever since. The Ribbon is awful answer to this, resulting in a dumbed down interface that is still too complicated for simple work, while burying more complicated stuff in places that are even harder to unravel than before.

Other programs like Ansys EM let you modify menu bars and all, which is a time consuming headache, only to mangle them if you upgrade, or open it on too small a screen (remote desktop for example). I've mostly given up on customizing interfaces because it is work that has to be redone far too often, and makes it a real pain to get support (or to support others).

Re:Basic and expert modes

[mcswell]

Hey, I resemble that remark! About the geezer, I mean. At least early stages...

What comes after user-friendly design?

[jrq]

Boozer-friendly design
An interface that's easy to use when inebriated.

Re:What comes after user-friendly design?

[Anonymous Coward]

You stole my standard for ease of operations

If you need to ask, turn in your geek creds...

[Anonymous Coward]

"Evil Geniuses in a Nutshell" by J. D. "Illiad" Frazer . Now get off my lawn!

Menus obsolete

[Tablizer]

As more options pile up, menus are becoming too big and deep to be useful. It's time to meta-tize options so one can search for them google-esque. Give each a unique ID so that one can bookmark them and even add their favorite option into their own tool-bar and/or menu as they choose. It could be kind of a friendlier version of Firefox's about:config tool.

If there are dependencies, then the "parent" option(s) or group-set can be also displayed. Old-fashioned menus can still be available, but not be the only way to access options.

And make the scope clear: is a given option just for the current document, all documents, all documents of current user only, a given domain, a given sub-domain, etc.

ruff draft schema:

options TABLE:
    id
    title
    descript // longer explanation
    type // string, int, double, datetime, bool, path, etc.
    value
    default // out-of-box value
    scope_type // depends on app
    group_ref // id of optional group, null if no group
    menu_ref // id of menu for the old-style menu position
    keywords // synonyms to aid in searching

Re:Menus obsolete

[Anonymous Coward]

totes relevant XKCD

Re:Menus obsolete

[excelsior_gr]

Give each a unique ID so that one can bookmark them and even add their favorite option into their own tool-bar and/or menu as they choose.

Do you want SAP? Because that's how you get SAP.

Re:Menus obsolete

[hackertourist]

Adobe FrameMaker does this. A search field (new in the 2017 version) allows you to search all menu commands and dialog boxes. You can customize menus, move commands around, make new menus, add shortcuts to any command etc.

Re:Menus obsolete

[Anonymous Coward]

Not to be rude, but that sounds, well, awful. Horrid. Terrible. User-hostile. Shockingly bad. Positively user unfriendly.

Menus are a simple, natural, hierarchical, informally semi-standardised method of storing available options.

Menus spell out the options. They don't make you play "guess what the squiggly thing with a donut and a kebab next to it does?".

Menus don't hide things. Look and you shall find.

Menus don't move stuff around for no reason. If it was there last week, it's there now.

Menus can be navigated easily with any device: keyboard, mouse, touchpad, touchscreen, trackball, whatever.

Menus make it easy to display shortcuts: underline p in print and power users will know that means ctrl-p prints.

Menus are consistent and universal. You don't need to relearn some ux guy's random thoughts from their last hard drinking session every time you get a new app or update.

If you must include toolbars or whatever on top of menus then fine, I'll happily ignore that, but please: leave the menu be.

Re:Menus obsolete

[Moof123]

How about a right-click->hide menu option? Have the first selection in EVERY menu be "show all menu items", allowing you to right-click->unhide menu option.

Also, provide default sets of displayed menu items aimed at new users, experienced users, and "Show everything damnit" mode.

Re:Menus obsolete

[Tablizer]

I did not propose outright getting rid of them. They are just getting too big to be the primary and/or only way to find options.

If you like your menus you can keep your menus. I promise.

A favorite quote

[mfnickster]

"Unix is user-friendly. It's just very selective about who its friends are."

Anyone know who said it first?

Re:A favorite quote

[thinkwaitfast]

One of my favorite quotes. I think I first read it on a bbs in the 80's or 90's.

Re:A favorite quote

[mfnickster]

One of my favorite quotes. I think I first read it on a bbs in the 80's or 90's.

Ah, back when Unix was in her teens... she was so beautiful! :')

Re:A favorite quote

[Anonymous Coward]

Tollef Fog Heen

"Do what I mean"

[Anonymous Coward]

Use AI to eliminate the user interface. Just do what the users want. Yes, you want to buy that, shut up.

After? FIRST time is still a fuck up!

[UnknownSoldier]

Modern UI / UX design is a clusterfuck of bad design.

Everything that was learnt for the past 40 years has been thrown out the window. These morons are so focused on Form over Function that you get stupid shit like this:

* How dare we "clutter up" the UI and show the user a scroll bar so they can gauge spatial proximity. Now we have "endless" scrolling with no scroll bar -- so you have no fucking clue how far along the content you are. Want to QUICKLY scroll to a specific spot? LOL. Waste even more time trying to remember where it was. At least with scroll bars the slider position was a VISUAL MNEMONIC to help you remember roughly where it was.

* We get idiotic error messages that don't:

i) explain WHAT caused the problem in the first place,
ii) nor HOW to resolve it.

I just ran into one this week. I purchased an album off iTunes and only half the album was downloaded. Clicking on a song that was in grey pops up a dialog Item not available. No Shit, Sherlock. HOW do I _fix_ the problem ?! Really, there was no room to say "iTunes > Purchased Music" ???

* Worse, everything is "flat" so you have NO visual cue to tell what can be interacted with and what is purely informational. You are kept playing a stupid guessing game of "Can I press this?" In the past we had 3D shading for objects that you could interact with and flat shading for informational. From the _context_ you could figure out the UI. Now a days? HAHA.

* Gaudy colors are now "in vogue" because they have been smoking Hollywoods Orange and Teal crack pipe.

The only progress is that:

* "Search" has now been added to "Options" because who needs manuals, right?

* At least they are _finally_ starting to get a clue with 120 FPS. Consoles are still stuck on a shitty 30 fps.

Modern UI / UX people are morons. I fight with these people weekly where their latest design is always half-baked. Hell, just getting them to understand "mach banding" and the simple concept of adding noise to reduce it is an uphill struggle.

--
"Those who forgot the past are condemned to repeat it."

Re:After? FIRST time is still a fuck up!

[Krishnoid]

* How dare we "clutter up" the UI and show the user a scroll bar so they can gauge spatial proximity. Now we have "endless" scrolling with no scroll bar -- so you have no fucking clue how far along the content you are. Want to QUICKLY scroll to a specific spot? LOL. Waste even more time trying to remember where it was. At least with scroll bars the slider position was a VISUAL MNEMONIC to help you remember roughly where it was.

I could see this devolving from relational database query results. If you run a query for stories ordered by most recent first, you could render/memoize the first few stories right away, and then have the server retrieve and render more query results in the background.

As you scroll down (hopefully with at least a dynamically growing scrollbar), the server would keep retrieving results from that query and rendering them live. In this way, there wouldn't ever be a 'length' to the content -- only the end of the query results.

Re:After? FIRST time is still a fuck up!

[JohnFen]

But the user-friendly way to do this is simple: you know how many search results there are, and you scale the scrollbar accordingly from the very start.

Then the scrollbar would be able to fulfill one of its primary functions: giving you an indication of how much data there is to scroll through.

Re:After? FIRST time is still a fuck up!

[Anonymous Coward]

* How dare we "clutter up" the UI and show the user a scroll bar so they can gauge spatial proximity. Now we have "endless" scrolling with no scroll bar -- so you have no fucking clue how far along the content you are. Want to QUICKLY scroll to a specific spot? LOL. Waste even more time trying to remember where it was. At least with scroll bars the slider position was a VISUAL MNEMONIC to help you remember roughly where it was.

I think the problem here is that software used to be reserved for tools, whereas now it is a platform for "entertainment". That, combined with the metastasization of marketing further into "entertainment". The result is that a lot of the software that people interact with most often is not designed to be their "friend", but rather to manipulate them.

A lot of software doesn't want the user to know where they are, or have the ability to make considered, informed decisions. It simply wants to keep users "engaged" so as to pump up its own statistics or deliver the user's attention to "partners" working their own angle. It is the same attitude as casinos that create an atmosphere of perpetual twilight in order to keep the marks oblivious to the amount of time and money they are losing.

Re: After? FIRST time is still a fuck up!

[Anonymous Coward]

I think you must work with this guy

Re:After? FIRST time is still a fuck up!

[Anonymous Coward]

It is pretty common for a DB to not be able to tell you how many results there are until you finish retrieving them. Some people run a query twice -- once to count (SELECT COUNT(*) FROM ...) and once to get results. Other people run the query once, saving the results in memory, then counting them once finished. Both methods require waiting until the query is complete before being able to tell you the count.

If you're trying to start rendering results before the query finishes running, you can't start with knowing how many results there will be.

The "old fashioned" way to do it was to just have paging. When you wanted to see more results, you'd just click a button to go to the next page. That was great because when you clicked on a link and then went back to your results, you'd end up on the same page you left. With an infinite scrolling page, you end up back at the beginning.

dom

Re:SJW design

[Anonymous Coward]

I believe the 'respect' that's being referred to is the type of 'respect' where you don't ...

Right. The Summary makes no sense; User Friendly and showing that kind of respect are unrelated and certainly not disjoint.

tuned interfaces

[Anonymous Coward]

What comes next are interfaces tuned for levels of experience and intelligence, as well as options for speed vs. simplicity. A great example is the horrendous multi-step mobile friendly interface Paypal now uses to withdraw funds to your bank. It's so, so horrible, very slow, and designed to be used with fat fingers by people who can't concentrate on more than one thing at a time.

UX engineers should be shot

[Jamlad]

Maybe I'm getting old but I'm finding more and more "modern" UXs simply confusing and frustrating. I tried to book a plane ticket last week and found the entire experience undermining. Heaven help my grandmother try the same process.
Modern UX "designers" seem to have confused shiny with usable.

Re:UX engineers should be shot

[Anonymous Coward]

Maybe I'm getting old but I'm finding more and more "modern" UXs simply confusing and frustrating. I tried to book a plane ticket last week and found the entire experience undermining. Heaven help my grandmother try the same process.

Modern UX "designers" seem to have confused shiny with usable.

Yes.. please shoot them all, right after the lawyers.

My first suggestion is that UX designers need to have their eye sight, reduced 20/50 with an added astigmatism. Next they need to try to use their previous interfaces with their vision degraded and report on the experience.

A tool I have to use for work uses a pale brown background with pale brown text ( maybe 2% different from the background) and no way to changed it. More and more apps/tools are doing this sort of bullshit.

Re:UX engineers should be shot

[OneHundredAndTen]

Yes.. please shoot them all, right after the lawyers.

And bankers. Don't forget the bankers. Bankers, lawyers, UX designers - the scourge of the 21st century.

Re:UX engineers should be shot

[doom]

Yes.. please shoot them all, right after the lawyers.

An idea whose time may have arrived, though personally I'm willing to start with dunk tanks and see if they learn better.

Friendlier design

[Anonymous Coward]

first there was interfaces,
then their were user friendly interfaces

now i give you, friendlier interfaces!

What comes next?

[edibobb]

Apparently the crap we have now. I haven't seen a new user-friendly interface for 15 or 20 years. And get off my lawn!

What You See Is What You Get

[Anonymous Coward]

WYSIWYG.

I can teach a UI class

[avandesande]

Whatever thinking went into the ribbon thing do the opposite!

emacs does it all

[Anonymous Coward]

After all, emacs does everything.

Stop being manipulative

[JohnFen]

From TFA:

It’s a balance any designer with a brief to design an effective, engaging experience has to strike: “You want people to spend money on your game and you want them to spend time in it, but there comes a point where that can become detrimental to what’s good for them and what’s healthy for them.”

If you're wondering whether or not your UI is good or bad for the user's mental health, the problem is that your design has already gone off the rails and into unethical territory.

You're not designing a UI to be good or effective, you're designing it to be manipulative. Worrying about whether that manipulation is good or bad for your users is merely distracting you from the root problem.

Re:Stop being manipulative

[shess]

From TFA:

It’s a balance any designer with a brief to design an effective, engaging experience has to strike: “You want people to spend money on your game and you want them to spend time in it, but there comes a point where that can become detrimental to what’s good for them and what’s healthy for them.”

If you're wondering whether or not your UI is good or bad for the user's mental health, the problem is that your design has already gone off the rails and into unethical territory.

You're not designing a UI to be good or effective, you're designing it to be manipulative. Worrying about whether that manipulation is good or bad for your users is merely distracting you from the root problem.

In the 80's and into the 90's, user-interface design was reasonably user-centric because the goal was to get the user's jobs done so that they would stay with your product because it got the job done. "Stay with your product" meant buying new versions. These days, the product isn't paid for directly by the user, a lot of what keeps a user in your system is "network effect", and your goals are to prevent the user from leaving your system, rather than helping the user get a particular job done. Don't get me wrong - you're on Facebook because the people you want to do Facebook things with are on Facebook, so to some extent, having a horrible user experience is irrelevant, as long as the experience isn't bad enough to actively drive people away. But that's a really sad place for UI/UX design to be at, and it sets really bad examples for other companies (who can't really afford to screw the user this way, but don't know any better).

Unfortunately, it won't change as long as the primary question is "How do we convince the user to do what we want them to do?" rather than "How do we make it easy to do what the user wants to do?"

Desired Action Design

[gurps_npc]

Some designs encourage "errors". For example, the practice of making the "Close this window" button small and hard to see on advertisements.

You can of course do the opposite, making it big and red. Part of this involves making major decisions AFTER the product has been tested.

Good design should not be focused on "if the user wants this, they should do that." Instead it should reverse the process, asking "If the user does this, what is it they want?"

A good example is the horrendous, evil "Video that refuses to scroll away." When the user scrolls down to read the article, a well designed video would shut itself OFF not move down to block my view because of your desires. I clearly do not want to hear or see the video, otherwise I would not scroll away.

Re:Desired Action Design

[Anonymous Coward]

Good design should not be focused on "if the user wants this, they should do that." Instead it should reverse the process, asking "If the user does this, what is it they want?"

I think a better question is "What will the user want to do next, now that they're here?". Make the next reasonable action the easiest.

Re:Desired Action Design

[gurps_npc]

While your view is interesting, it has two issues.

1) It assumes there is a reasonable next action or small set of reasonable next actions. Often way too many other actions are reasonable.

2) It has a tendency to paternalistically tell people what their next action should be, with a bias towards what the company wants (hence CNN's evil "keep my video around" technique).

3) It is a lot easier to figure out what people want after they did an action (my technique) instead of trying to predict what they want before they take the action.

Software needs an 'ingredients label'

[DidgetMaster]

All software needs to have the equivalent of that 'List of Ingredients' you find on the side of a soup can. It should tell you exactly what data it collects, what kind of privacy you have, and how to switch it off. That information should be listed in a short, concise manner with a few icons that will make it easy to recognize. You should not have to wade through a 10 page legal document (after clicking though a dozen pages to even find that) in order to find out what it is doing with your information (assuming you understand cryptic legalese). The company should not be able to change the terms at the next update without throwing up a big 'red flag' and tell you exactly what they changed. Maybe we could even get some kind of standards body to come up with a 'Rating' from 1 to 10 about how intrusive a piece of software is (1 = saves your screen name, 10 = records the contents of your medicine cabinet) and make the software display it in the 'about box'. Adherence would be optional, but market pressures could drive out anyone who refuses to show the information.

Re:Software needs an 'ingredients label'

[JohnFen]

I really like this idea.

Re:Software needs an 'ingredients label'

[Anonymous Coward]

All software needs to have the equivalent of that 'List of Ingredients' you find on the side of a soup can.

*This software was packaged in a facility that also processes tree nuts and spyware.

Re:Software needs an 'ingredients label'

[Kjella]

It should tell you exactly what data it collects, what kind of privacy you have, and how to switch it off. That information should be listed in a short, concise manner

All your base are belong to us. How much shorter can it get?

Re:Software needs an 'ingredients label'

[doom]

Actually, all software needs is a list of names of the design team, preferably with a ranking system you can use to prevent them from ever being hired somewhere else.

Software companies could probably make some money on the side by putting the designers in dunk tanks, and have people contribute money until it reaches the "dunk" threshold.

Re: Software needs an 'ingredients label'

[Anonymous Coward]

What you say?

Re:Software needs an 'ingredients label'

[Anonymous Coward]

Adherence would be optional, but market pressures could drive out anyone who refuses to show the information.

Good luck with that. All the 10s would self-report as lower numbers.

(Hint: Locks on doors only keep honest people out, and self-reporting only works for honest companies.)

Touch screens are part of it

[tepples]

Perhaps the "UX" hate comes from the fact that the term took off shortly before smartphones using capacitive touch screens became popular. Capacitive touch screens forced UIs to drop long skinny menu items and dense toolbars in favor of larger square icons that a finger can hit reliably, and large (80x45em to 120x67em) desktop monitors were ousted in favor of 20x30em phone displays that just can't hold as much information without having to scroll or otherwise navigate.

Re:Touch screens are part of it

[JohnFen]

I agree.

But I think there's a trap that developers have fallen into here: different form factors have different UI requirements.

Trying to make a single "UI to rule them all" results, at best, in a user interface that is barely tolerable on any of them.

Re:Touch screens are part of it

[tepples]

The other problem is that a single device may support multiple input modalities. For example, a laptop may have both a touch screen and a trackpad available, and an all-in-one desktop PC may have both a touch screen and a mouse available. If the layout of controls in a native application or web application is supposed to depend on the input device, how is it supposed to know which input device a particular user wants to use at any given time?

Re:SJW design

[amicusNYCL]

That's nonsense. After user-friendly design, everyone know that the thing that comes next is a better idiot. That's the natural progression. Once you come up with the perfect design, you'll find people that have no idea how to use it. Those are the better idiots.

Some one please implement a user friendly design f

[Anonymous Coward]

Then we can talk about what comes next.

The Good Old Unix Way

[next_ghost]

As a long-time Linux user, I vote for going back to the good old Unix way of get-shit-done-friendly design.

Change for change's sake is not friendly

[swb]

Why does it constantly have to change? Most people can learn less than friendly user interfaces (have you seen some of the stuff low-paid clerical staff have to use in 3270 terminals?).

What's totally obnoxious is having the user interface redefined every year because some new crop of developers has decided they need to leave their mark and they have somehow determined that changing everything is necessary.

IMHO, GUI user interface changes not driven by significant changes in functionality really haven't improved ease of use. The original Macintosh or Windows needed user interface enhancements when they went full-on multitasking, but by and large those interfaces were probably as functional as the flat, widget-and-gesture-laden interfaces foisted on us as improvements now.

Touch-driven, small devices like smartphones benefit from UI changes (physical use case and interface technology), but I really don't think desktop PCs have benefited at all from the UI churn.

Re:Change for change's sake is not friendly

[JohnFen]

Yes.

Designers need to internalize the essential truth that all user interface changes are expensive for existing users, even if the changes are clearly for the better.

If the benefit of the change is greater than the cost, users celebrate it. If not, users rightfully despise it.

Re:Change for change's sake is not friendly

[Anonymous Coward]

As the number of features increases, the UI has to change to allow the user to access the ones they need in a direct, discoverable fashion.

It used to be that most computers didn't even have networking, and if they did it was configured by some corporate IT personnel. Now every computer is networked and the configuration constantly changes. For example, on a given day a business user may leave home, go to work, then to the airport, then get on a plane, and then arrive at a hotel -- each of those places having a different WiFi hotspot. If I have to change my network configuration several times a day, it has to be much more accessible than back when it was set in stone on my corporate desktop computer by the IT guy.

dom

Re:Change for change's sake is not friendly

[Anonymous Coward]

Sales. If the UI doesn't change people will stop churning, sales will drop and profits will fall off a cliff. Screwing with the interface props up the illusion of progress and keeps the new and shiny brigade out and buying.

Obvious example: windows xp through to windows 10. Real improvement? Meh. Lots of change to keep people on their toes? You betcha.

Re:Change for change's sake is not friendly

[Anonymous Coward]

What's totally obnoxious is having the user interface redefined every year because some new crop of developers has decided they need to leave their mark and they have somehow determined that changing everything is necessary.

Well, you don't have to go with that. Don't buy (or use) their changed design.

I used the same computer UI from 1997 to 2017. (icewm window manager). In 20 years I didn't feel a need to change the looks of my day-to-tay computers - so I didn't.

Open source is sometimes criticised for having too many options. (Not really a problem, go with defaults if you don't like to choose). But the beauty is that I can stay with an interface I like. There are newer options - osme of them trying to mimick windows or whatever. Those who like that may try - others do not need to waste time learning something 'new' to do the 'same' every year.

Re:Change for change's sake is not friendly

[JohnFen]

access the ones they need in a direct, discoverable fashion.

Judging by the majority of user interfaces I've seen coming out over the past few years, "discoverability" is no longer considered an important thing.

Re:Change for change's sake is not friendly

[JohnFen]

Screwing with the interface props up the illusion of progress and keeps the new and shiny brigade out and buying.

Maybe, but pointless UI changes are the primary reason why I don't let any of my software auto-update anymore.

Re:Change for change's sake is not friendly

[mcswell]

"the UI has to change to allow the user to access the ones they need": That's ok iff I do the changing. When the software's builder does the changing because s/he/it thinks they know what I want, then it's bad. The Ribbon is a case in point; with the menu, it was easy for me to find what I wanted. No more; everything I don't need is in my face, and what I do need is somewhere else.

And Microsoft is not the worst at making useful things hard to find--Adobe is. The recent versions of Adobe Acrobat are simply terrible at finding what I want, and the UI icons take up way too much room. (Microsoft at least gives you the option of making Start menu icons smaller; no such luck with Acrobat.) The only reason I use Acrobat is that I'm forced to at work. At home I use PDF XChange. I won't say its UI is ideal, but it's far better than Acrobat's, and much more easily modified.

DoItForMe Design.

[Anonymous Coward]

See- The Animatrix: The Second Renaissance for theoretical examples of how this might go down. (Is already going down...)

We already know

[johannesg]

After user-friendly design comes aggressive phonisation:

- Simple, clear, concise, discoverable menus are removed, or hidden underneath an utterly unclear icon, or somehow folded out into a ribbon that makes just about zero organisation sense.
- Bloody icons everywhere. And none of them represent a meaningful real-world thing, it's all about as understandable as Chinese. In fact they should go and label all those icons in Chinese; at least some significant fraction of the world's population will understand, and for the rest learning Chinese will not be any harder than learning 'hamburger', 'triple dot', or 'diagonal bar with half circle'. And there's the added benefit of learning Chinese, of course.
- Animation, animation, animation. It was a relief when I finally turned animated windows off on my new laptop (after a few hours); waiting for windows to melt in and out, menus to slide open, dialog boxes to fade in, etc. was really getting on my nerves. If a window opens, I want it as soon as the machine can manage.

We used to have a visual language that did a really good job at hinting what an interface was going to do. Buttons had a certain shape, checkboxes another, text input boxes yet another, etc. Nowadays it's a total crapshoot. A blank area might trigger an unexpected command. A piece of text may turn out to be a button while a neighbouring, visually identical piece of text does nothing.

We could have hoped for true resolution-independence instead of half-baked coordinate scaling, but nooo... Considering this was a solved problem back in the nineties it's a shame we still have to suffer through non-resizable windows and fixed-size fonts.

Re:We already know

[mcswell]

Amen. I did a similar posting about hieroglyphics before I saw yours (which is much more complete than mine).

UX designers need not respond

[Anonymous Coward]

UX is not UI, UX is people who don't know software who get paid to obfuscate software. You are not computer people you are parasites.

After...?

[mattack2]

Too many products still aren't user-friendly, so we can't be 'after' it yet.

Re:SJW design

[mrclevesque]

I think they mean user friendly used to mean things like ergonomics, efficiency, simplicity, and the like, but more and more 'design' seems to be about things like getting the user to purchase stuff in a 'store' / getting them to install stuff they don't want but now must install or else lose out on some functionality they do want / or getting them to do things like run through a 'design maze' in a way that maximizes things the software owners can track and monetize.

So, if good software design should be about being friendly to the user's wants, and what we often get is the opposite, then maybe what designers are lacking is respect for the users.

So, to the question what comes after today's so called user-friendly designs -- hopefully it's designs that are actually user friendly.

I know the answer

[AndyKron]

What comes after? Stripping it down to the bare minimum and extinguishing any and all quality that it may have originally had.

First...

[Anonymous Coward]

Maybe we should actually make things user friendly before we try to go beyond that

Covered in www.userfriendly.org

[Antique Geekmeister]

That old webcomic covered a surprising number of very real user interface issues. It also covered office politics, and technological egotism, and interactions between complete geeks and other people. It's still surprisingly apt after being in reprints for years.

Re:Covered in www.userfriendly.org

[ray-auch]

Yeah, but it's dead, just stuck on repeat nailed to its perch.
As is the old "user interface hall of shame" - dead but archived in various places.

And yet every damned thing they said about UI design is still correct and relevant - it's just that no one actually gives a s**t anymore. Success in UI is now measured by how long you can keep users on your site / in your app, not how fast or how accurately users can actually get stuff done. Or how many ads you can show per user action.

Eventually the decline in UI will render the user irrelevant - when computers start doing things for the user it won't be because computers are smarter, it'll be because even a dumb computer can do a better job than the user through-the-crap-interface. Then we can all climb back into the trees and watch the computers screw up the planet.

You just don't understand

[Anonymous Coward]

You misunderstand because you assume altruistic goals when it is evil, marketing goals at play.

The value of knowing your relative position within a sea of content is that using that information later to find your way back to where you were.

The places that implemented infinite scrolling did so because they want control over what you see. Infinite scrolling means that when you return to them you cannot jump right to what you want and will instead have to wade through viewing what they think you should be viewing.

The other reason such places use infinite scrolling is because it provides intermittent reinforcement for operant conditioning. Each visit you have to scroll a different amount to reach your destination and that is a powerful way to condition behavior (which is why the same approach is used in slot machines).

So the next time you encounter infinite scrolling ask yourself: does this site/app benefit from making me look at what they are promoting rather than just what I want to see? And does this site/app benefit more when I stay here longer by scrolling more?

Design is not the issue

[edtice1559]

There are some really poor UI designs out there. But generally the UI is somewhat of secondary issue. Many problems require deep domain knowledge to solve and we keep trying to build electronic systems that can be used without the prerequisite skills. Those will continue to fail. Things like Facebook are useful regardless of UI as long as it is even tolerable. So the issue is not UI improvement in most cases.

Make it user hostile

[thinkwaitfast]

from history

Magic

[Anonymous Coward]

User enters information, and poof a new discovery about themself is born. UI makes it easy for systems, not necessarily users.

After user-friendly comes...

[doom]

Advertiser-friendly. In fact that comes before user-friendly.

And asking a bunch of designers about this has fox/henhouse issues, I think.

(I know, let's do a user survey on the subject of javascript popup windows, and if it comes out thumbs-down, that means everyone will stop using them, right? )

Idiot-obvious without being expert-obnoxious

[LesserWeevil]

Back in the day when I was doing UI code, I strove to make interfaces "idiot-obvious without being expert-obnoxious" and figured that if you needed a manual (or man pages) I'd done a poor job of things. These days, I don't know whether to laugh or cry at some of the commercial efforts at UI design as, frequently as not, some ad pops up obscuring some item I'm trying to read and usually can't be made to go away without the click-through dance. Want to make UIs run smoother and look better? Take the freaking ads out. And get off my lawn!

Re:SJW design

[Anonymous Coward]

If it's monotonous, then you're not doing it right. The point of text config files is that you can either edit them by hand or use tools to do it for you. *NIX has great tools for changing text files built into the system and if that's not enough, you can install Perl and do all sorts of crazy stuff.

For some systems, you can even load those files over the network and generate them procedurally so that you're not having to do them over and over again.

I think the next stage is going to be where they lose all pretense of caring about the user doing work and just load random cat videos from the internet.

Menus NOT obsolete

[Anonymous Coward]

So guessing at what a designer named a function is better than being able to read it from a coherent hierarchical structured list?

I see.

FUCKING FUCK this fucking shit. FUCK you and the fucking horse you rode in on you fucking stupid fuck.

I bet you're one of those people who runs a search indexer on their computer because they have no idea where their files are or what they are named. (I keep a working directory structure of over 300K files and I know exactly where they all are and what they contain. No search index here, the disk is active when I want it active and at no other time.)

Trees are natural. Dendritic structures are logical.

Oh, and fuck you and your fucking stupid horse and your stupid idea.

Re:Menus NOT obsolete

[Tablizer]

Please re-read my statement. You missed a key phrase.

We are already out of "user-friendly design"

[Casandro]

GUIs are lacking more and more visual clues to how they work. Programs are getting rid of menu bars or scrollbars.

In the past companies used to have usability testers which tried to perform certain tasks under supervision. This is gone now and, at best, replaced by much more primitive AB tests.

Essentially we are now at the level of 1990s film-gui designs. It doesn't matter if it's efficient to use all that matters is that it looks "pretty",

Re:We are already out of "user-friendly design"

[JohnFen]

This is gone now and, at best, replaced by much more primitive AB tests.

This brings up another thing that I think has had a truly awful effect on user interfaces: telemetry data. It's so much easier and cheaper to just base your decisions on telemetry rather than actually doing user testing, so that's the way things have gone.

The problem is that telemetry data distorts everything (for example, it encourages idiocy such as assuming that reducing the number of clicks to accomplish something always makes things easier for the user) and is very, very easy to misinterpret.

Re:We are already out of "user-friendly design"

[mcswell]

Agreed. Also icons, which are as incomprehensible as hieroglyphics. (There's a reason why Egyptian hieroglyphics were indecipherable for almost two millenia, and Mayan hieroglyphics for centuries.) Give me an alphabetic writing system any day, and a menu.

user friendly?

[SuperDre]

There's a topic that isn't clear cut.. What IS user friendly? there is no ONE user friendly way, as what might be user friendly to one person might be crap to another (believe me, these days a lot of designers have no grasp on what actually is user friendly).

What comes next? Pulling the rug out.

[HalAtWork]

After user friendly UI design comes the Flat UI that starts removing visual cues and functionality

Look at smartphones

[p51d007]

You want to know what comes after so called user friendly? Look no further than your hand, because you are probably clutching one in your hands, afraid to put it down. Smartphones. Instead of being a good user friendly "tool" they have been taken over by the slim/stylish/colorful/fashion conscious types. When a new phone is introduced, they typically tell you or bring out the guy that "designed" the outer shell, who is typically some high end man bun wearing fashion designer. They will tell you how attractive it is, how sleek & stylish it is. Hell, you'd be lucky for them to even bother telling you how well it works AS A PHONE. And, in the past few years since they latched onto this stylish crap, look how the price has skyrocketed into the $1,000.00 range. Most smartphones in the "flagship" range, have build prices of LESS than $300.00, but "command" A THOUSAND dollars or more. Can't blame the manufactures if consumers are stupid enough to line up and pay that for "a phone".

If you are asking this you don't grok UI/UX at all

[Anonymous Coward]

Nothing actually "comes after" user friendly except user friendly. You seriously need to grok UI/UX better because "just, wow!"

The end goal is to reach an appliance level of usability. Do you need to read a manual to use a refrigerator? Do you need to receive special training to use a light switch? That's the end-point.

Once you've achieved "invisibility" of the interface, you are done with that particular interface - there is nothing after that. That is the goal.

Of course, you may or may not have really reached that end-point. So there is refinement. For the majority of designed things, there is no awareness of how people actually use the thing. These tend to be where the UI is the absolutely worst. Either no thought was put into it (no one ever did a "use case" at all - usually there should be a number of these explicitly documented) or the designer didn't give a shit about the user and only wanted to "pad their artistic portfolio with what they imagine is 'pretty' or 'professionally advantageous'".

VIC-20

[mcswell]

The first computer I owned was a Commodore VIC-20. It advertised "user friendly BASIC programming language".