Slashdot Mirror
Is There Such a Thing as "Too User Friendly"?
rtphokie asks: "The story about the TiVo get-together along with some recent trials and tribulations rolling out a knowledge base along with the time I've spent recently helping my 80 year old grandfather with this VCR and TV has gotten me thinking about user interfaces and the elusive "user-friendly" label. When someone who thinks of themselves as 'non computer savvy' works with a gadget like TiVo and compains that it's 'too complicated', how should we react? Why are users immediately forgiven for not even taking the least amount of effort to look for a solution to their confusion in the manual. The tendency has always been to blame the interface and ultimately the engineers who designed it but isn't there a point where users have got to share some of the blame?
Why do today's software and consumer electronics users expect to be able to fire up their new toy and magically have a complete understanding of how to use it?"
Never. The simpler something is to use, the better.
Don't confuse simple to use with basic - just because something is easy to operate it doesn't mean that it's incapable of doing some complicated things.
Many examples spring to mind but the telephone is top of my list. With my phone I can call half way around the world in just a few seconds - heck, even my two year-old nephew can.
"Accept that some days you are the pigeon, and some days you are the statue." - David Brent, Wernham Hogg
...is to just "Wizard" every action the user may need to take. By trying to anticipate what the user wants, a wizard can be provided to allow the user to quickly, and easily, complete their task. Of course, then you end up with a wizard so large and complex that it becomes an OS in itself, and one needs to read the help files associated with each option to successfully progress thorough the wizard's heirarchical structure (refer to Windows XP's default settings for the control panel). You have to know what each option does before you can click it. So eventually, when wizards rule the lands, there will be a manual for the wizards! And, as a "computer guy" I can still say "RTFM!"
The only "intuitive" interface is the nipple. After that, it's all learned.
Bruce Ediger, in comp.os.linux.misc, on X interfaces
"Weapons should be hardy rather than decorative" - Miyamoto Musashi
I think that goes for OS's too
with all respect to your G'father, he has probably not operated enough electronic items to learn the "language" of electronic gadgets. The more he operates, the more likely he would intuitively understand how to use something.
;)
This idea is discussed in Donald Norman's Design of Everyday Things, which is a great book for UI people.
Also, I have never seen the Tivo's UI, so it could be poorly designed...
::.. check out some Cell Phone Reviews
Now go build your system so that someone can use it without knowing anything. Also, make it so that an advanced user can get to the functions she wants without going through some idiotic "wizard."
UI tests with actual users? What a interesting thought!!! Maybe someone should try that, too!
Yeah, right.
You mean I clicked the "1 reply beneath your current threshold" link, and I didn't even get a funny comment about the poster's girlfriend?!
I know quite a few people who can't program their VCRs, and seem proud of their ignorance. These are not (all) stupid people, but it seems that anything even slightly technical is beyond the interest of most of the population. (I'm laughing here thinking of the episode of the Osbornes where Ozzy is trying to use his state-of-the-art entertainment centre: "Why is it you need f*ckin' compuer skills to turn on the f*ckin' telly!?")
When something as simple as setting a start and end time plus a channel is beyond a large proportion of the population, it's going to be impossible to design an interface for TIVO that *anyone* can use. At some point you have to give up...
Code, Hardware, stuff like that.
From a marketing point of view you're dead wrong. If you want to survive in a competitive marketplace you can't be telling your customers to RTFM. It just doesn't work that way. Bash Microsoft and AOL all you want, but part of their success is definately due to ease of use.
There is no such thing as "too user-friendly". If someone buys a surround sound stereo system it's because they want good sound while they watch movies. They really shouldn't be asked to learn the intracacies of stereo system design.
In the end, it should just work. If you don't make a product that's easy to use, somebody else will.
Everyone writes one interface for every skill level. There ought to be different interfaces according to your choice, or according to what level of interface the system thinks you can handle.
That last part's a bit broad, so I'll clear things up. With a normal PC, you've got CPU cycles to spare, and the computer has time to tell if you move deliberately for a menu choice, or if you're hunting for it, or if you keep choosing something, and cancelling out of the choice.
For a VCR, the default interface should be as simple as the buttons on the front. If you read the manual a bit, it will tell you how to turn on the intermediate features. If you read a lot, you can turn on the advanced features. If you read waaay too much, you get to turn on the command-line interface that uses reverse-Polish notation, in Aramaic, but displayed approximately by using Turkish for vowels, and Cantonese for consonants.
Everyone's not as comfortable with it as folks like us are, and because computers can do sooo bloody much, we should stop boring them, and give the computers more to do, such as providing different interfaces for different skill levels. We use short command interfaces with our kids and our pets ("Sit! Quiet!"), and much longer command interfaces with our peers ("Dude, nice frag!"). It's a very natural thing to do, and we ought to start allowing computers to do the same.
Funny, we were just talking about this as it related to another post I just made. The thing is, there is no such thing as user friendly, at least the conventional meaning of the phrase. It all boils down to two factors:
The phrase "user friendly" comes about by confusing the two: somehow assuming that by being easy to sit down and learn with no work, something is easier to use. Then it's "user friendly."
Unfortunately, this isn't how it works in the real world, at least usually. A tool can be built that is easy to use---powerful, flexible, suited toward the job; or it can be easy to learn---no training required. Usually the tradeoff for the latter is that functionality is limited, so the user isn't overwhelmed. A balance of sorts must be achieved. Most of the best tools lean toward easy to use, and rightly so: you're only a newbie for a very short time. You may be using the tool for the rest of your life.
However, these aren't necessarily mutually exclusive, either. It is possible, in theory, to build an interface that is both easy to use and easy to learn, as long as one does not equate the two, or think that one somehow implies the other. Doing this is rather tricky though. A good example of such interfaces are those for simple tools which can be applied to a wide variety of uses (a hammer, /bin/ls, etc.). Another example is that some games tend to use: the dynamic interface, which starts with a few key options, and gradually adds more.
Thus, "user friendly" doesn't really exist in the conventional sense, which equates this sense of immediate ease of learning with continued ease of use. Rather, ease-of-learning and ease-of-use must be balanced, and attaining something truly user friendly requires a lot more than having icons and a mouse, or fewer menu entries.
Don't think of it as a flame---it's more like an argument that does 3d6 fire damage
It's been my experience that:
1. 90+% of users are incapable and/or unwilling to think. Regardless of how obvious the UI is, they need to be sat down and trained like monkeys to repeat a series of steps to accomplish whatever they're trying to do. They cannot, or will not, stop, look at the screen, and make an intelligent choice on how to proceed. No matter how plain and simple the UI is, it's like they had a part of their brain removed.
2. About 5% of users can make decisions based on the UI to accomplish their goals.
3. The remaining few percent, which we would call Power Users, have a decent understanding of how computers work, how files work, where they're located, how to find them. They know that if they're trying to open a file, they can usually do this by clicking File, and maneuvering down the menu. They can figure out that if their X: drive isn't opening, it's probably because they aren't logged in to the network. They can take a tip, and make a logical conclusion, like "Oh yeah, okay, then I can do this and this. Thanks." These users are very few and far between.
Windows is great for the few who understand that there are common elements of (most) every application. Still though, it's that 90+% that will suck the life out of you every time.
"Would it kill you to put down the toilet seat?" -- Maya Angelou
On the contrary; I think it's a powerful and much under-rated approach. The biggest hurdle for most people learning a new tool is (arguably) coming to understand the fundamental way it works. After that, the rest is often just details.
For example, if I'm using a new word processor, maybe I learn that its formatting is broken down according to characters, paragraphs, etc. and where to find the dialog for each. Then it's not a big jump to work out how to make something italic (a simple task) or to set up the kerning (a more advanced one). In this case, it would be useful to have a simple UI with common options (open and save files, change the font, run the spelling checker, etc) and a full UI with the whole lot (revision marks, change the number of columns, configure the grammar checker, perform a mail merge).
Personally, I used to like systems that worked that way. You could start simple and learn the big picture, and once you'd got the hang of it, switch everything on and see all the details. Then you knew everything was there and you could see where you stood. These days, everything seems to come with seventeen different ways to do the simple things and an options dialog with 100 different settings, most of which show or hide some feature if the menus aren't already adjusting under your feet before you start anyway (but luckily there are seven different ways to get help). Is this really easier to learn and more user-friendly, or just making a simple tool like a word processor seem far more complicated than it is? (There's an obvious commercial/upgrade angle here, but it's not really relevant to the issue at hand, so I'll gloss over it.)
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
I'd guess that most people reading this, including me, know more about info tech than 99% of the population. It's easy for us to say that anyone who doesn't figure computers out is just not making an effort and respond with a 'RTFM'.
... Wash DC? ... local judges? ...)
But why don't we look at some fields that perhaps are not part of our aptitude. How much time and effort have you spent learning about,
- a recipe?
- fashion and clothing?
- fine art?
- your elected representatives (quick, name the ones in the State capital
- giving your girl/boyfriend a mind-blowing orgasm?
Now, you may say, 'but these things aren't important to me; I don't have time for them.' And then you'll understand why all the 'lusers' don't RTFM.
I wish people would talk about "logically sound" rather than this completely nebulous concept of "user friendly."
Look at Windows. A great deal of the garbage we hate in Bill's operating system was stuffed down our throats under the guise of being "user friendly." For example, changing the name directory to "folders" because directory has unfriendly latin roots. The actual result of this great "user friendly" move was Microsoft now stuffs the end user's data in a bunch of folders that you cannot find...making back ups harder. The goal of an OS should be to concentrate on creating a logically sound, secure foundation on which you can build other applications. But we compromise the foundation for an undefinable user friendliness.
It is so funny. I see it time and again. People love the "user friendliness" of MS word when they log on the first time. A few years later they are pulling out hairs as they find their systems clogged with gigabytes of files, odd templates, virii and other mysterious things that happen with word documents as systems age.
That really crappy registry thing we have to deal with came out with a great deal of hype about a "user friendly" registry replacing unfriendly ini files. Instead of coming up with a logically sound and versatile and extensible mechanism for recording intialization parameters...we have this supposedly user friendly monster that bites our tails when things go wrong. The only way we can deal with problems in the registry is to hope that some programmer somewhere was good enough that their 5 year old win 98 program will fix the registry problem with XP when you reinstall.
The parent of this thread was "Learning Curve." The result of the user friendly movement has been to add a bunch of garbage to programs to get the public to a feel good level, but the garbage ends up blocking them from complete mastery, since you know have a garbage user friendly layer in the way.
Instead of "user friendly", if you aimed at the goal of logically sound...you would find yourself with products that have only a slightly higher initial learning curve, but that people can master and build on. Take the threads about driving. The configuration of the driver seat has a nice logically sound foundation. It is driven by the logic of the vehicle and it works better.
When you really have a sound logical foundation, the actual workings of the product is all but driven from that foundation. A phone is totally un understandable until you know the logical premise that you have to hold it to your ear, and that different phones have numbers that you must dial before calling.
Imagine a car designed by the "user friendly" gurus of MS. A six year old could get it out of the driveway, but it would take a certified MCD (Microsoft Certified Driver) to get it back in.
Now, every once in a while I get asked this question: how is it that a VCR can record a TV show when the TV isn't turned on? Yeah, I can hear the snickers. But I get this from a lot of basically intelligent people. And the frustrating thing is, I've never found an explanation that makes sense to the asker. To me it's obvious, "You see, there's two tuners, the TV has one, the VCR has one...." But the eyes just glaze over.
So the whole idea of Making Systems User Friendly is just plain bogus. It assumes that people can come to terms with any system if you just find the right methaphro for them to use. Doesn't work.
In the real world, there are three solutions to this problem:
- You do a better job of explaining the basic concepts of the system to your users. But only a few really brilliant teachers seem to have much luck with this approach.
- You build systems that do a good job of hiding the unfamiliar paradigm with a simpler paradigm ordinary people can wrap their minds around. But again, this takes a certain brilliance on the part of the designer, who has to be at home with both paradigms.
- You take the Kuhnsian approach. That is, instead of trying to bridge the nerd-mundane gap, you wait for both sides to die off, to be replaced by big-thumbed folks who've grown up with the technolgy and have no trouble coming to terms with it.
Now, you might think that solution number 3 is basically a cop-out. And I'd agree. But I think it's the solution that will be implemented -- by default.This isn't about being interested or not. This is about people who clearly want the result but are unwilling/unable to learn the process.
This isn't about disgust with people who say, "I don't want to program my VCR." it's about those who say, "The VCR is too hard to program, I can't learn it." Usually, this can be translated as, "I am too lazy/frightened to bother trying."
In my experience, if you have authority over these people, you can easily make them figure it out. Without authority over them, they'll make weak excuses why they shouldn't bother trying. If they have authority over you they'll get you to do it over and over again, regardless how much of both your time and theirs this wastes. 90% of what computer class teachers do is say, "You have to try."
It's a truly pathetic phenomenon. I could throw theories at you about why it is, but I'm not sure why most people's minds work that way, they just do.
A quick example... about three years ago, I commented that you should always use a UPS on a Linux box, because the ext2 filesystem was fragile. (there was much more to this, but in the interest of brevity I'll omit it.)
So what did I get in reply? "You're a moron, you should be manually editing your filesystem when it's corrupted and using backups of the superblock." And other posters appeared to agree with him. I don't think I got even a single reply in support of my stance... that I shouldn't have to, that a properly designed fileystem wouldn't have these problems. I'll not repeat the whole argument. Either you will understand why this was a ridiculous thing to say or you won't. But the blame-the-user mindset was firmly in place... it was MY fault because I didn't know enough, not the fault of the designer(s).
Read the book "The Design of Everyday Things". It is a great set of examples of how badly real-life things can be designed... and how a properly designed real-life thing should automatically guide the user into using it correctly. A door that pushes, for example, should NOT have a handle, it should have a push plate... and maybe a handle for the other side, because it pulls on that side.
According to research, there are two basic ways that humans organize data and navigate through the world: "knowledge in the head" and "knowledge in the world". People who use the former are Slashdotters... they use their memory as their primary navigation device. They tend to trust their own memories over things like street signs and maps.
The other type of thinker uses the world around him/herself to keep them organized. WHERE the piece of paper is tells them WHAT it is. They'll trust a street sign over their memory every time. They don't try to store the entire world in their head, and (this is the crucial part) they get confused when input isn't consistently mappable to output.
A car is easy to drive for everyone because inputs translate to outputs in a simple, direct way. There are only a few states and only about five main inputs. Anyone tall enough to see over the dashboard can successfully move a car with an automatic transmission.
For 'in the world' thinkers, however, a computer is a deep mystery. Inputs don't translate into outputs. In a car, if you push the accelerator, the engine revs up, and the car usually goes faster. On a computer, if you click the mouse, a zillion different things could happen, depending on where the pointer was, what mouse button you pressed, what program was running, or what the time of day was, or what have you. This means computers are HARD for 'in the world' types.
That is part of what was so successful about the Macintosh. One button. Short menus. It's still complex, but the inputs map more closely to the outputs, and the onscreen cues make it easier for externally-organized people. The internal states of the machine are more clearly reflected on screen.
Just because something is complex on the inside doesn't mean it has to be complex on the outside, too. A modern car is an exceedingly complex device, and it takes a lot of training to be able to repair one if it breaks... but pretty much any idiot can drive. (and, judging from what I see on the freeway every day, every idiot does. :-) )
Computers can be this way without sacrificing their power. But it's easy to blame the user and ignore the problem when the solution isn't easy. Look at my ext2 experience. Back then, it was my fault. Now that we have journaling filesystems, it's obvious that a well-designed filesystem doesn't need manual editing of the superblock after a power failure.
Likewise, we'll someday look back and realize that gadgets didn't have to be hard, we just made them that way. And it's nobody's fault but ours.
I remember a web-site that covered the worst software UI's. I cant remember (or find) the site. It covered Quicktime, some IBM software and others.
The Interface Hall of Shame, most likely.
I think you've hit on an interesting social phenomenon. It's culturally acceptable -- perhaps even desirable in some circles -- to profess ignorance about certain things. I can't count the number of times, for instance, that I've heard people proclaim "Well, I don't really understand math", not with shame but with something approaching pride. (In case math-savant slashdot readers have a hard time relating to this particular example, try replacing it with something more personally salient like "I really don't understand women". In my experience, such a statement is often used as an incentive to bond with other people who feel similarly, not as a shameful admission.)
Then again, there are things that it's not socially acceptable to admit lameness in. Openly admitting lack of knowledge of computers would probably be fatal in a forum like this one. Openly admitting a lack of knowledge about the mechanics of sex (once you're beyond a certain age / experience level) is probably something few people would do. (Though there is a Sex for Dummies book, so who knows -- I figure that's something you buy only as a gag gift, and you make sure that you get it gift-wrapped at the checkout counter!) Or ignorance of how to operate a motor vehicle (unless you're a lifelong Manhattanite, in which case it could be a perverse source of pride)...
"Biped! Good cranial development. Evidently considerable human ancestry."
The 80 year old grandfather's problems with the TiVO can be attributed to the fact that as people age, they experience a decline in Fluid Intelligence (their ability to deal with novel problems that do not draw upon previous experiences). It's not that the grandfather was stupid, or that he didn't read the fine manual. It was that his brain's ability to deal with a new situation that didn't draw on his past experiences was not what it used to be. When you also consider the decline in performance of short-term memory that the average 80 year-old experiences, it is really no surpise the grandfather had so much trouble.
To design something for someone of that age, you have to draw upon their Crystallized Intelligence(the store of knowledge or information that a given society has accumulated over time). You might (if you're *really* a geek) be able to do something like rig up an analog alarm clock to the TiVO and expoit the grandfather's 30 years of experience setting alarm clocks to get him to successfully set the TiVO. Yes, he'll probably still need a TV Guide to look up the time so he can set it in the alarm clock, but the point is that the show will be recorded. It sounds crazy, but older adults often exploit their crystallized intelligence to create strategies that work around deficiencies in fluid intelligence.
If people hack network interface cards into their TiVO's, why not hack Grandpa interface alarm clocks into them as well?
Ergonomica Auctorita Illico!
uberworld.org
That was it. In short, it's a like a MUD, except it's full of people who sit around (mainly students and sysadmins) and chat about whatever they want all day. It's proper name is a "talker" and it used "telnet".
Now this is where the problem lies. I consider the interface to be obvious. You have a bunch of commands and help files called with "help" and it's all very easy.
But the people logging in from Slashdot, just didn't have a clue. And by that, I mean they had no idea what to do. These are people who use UNIX all day long and yet they were lost.
So I looked at the mistakes they made and I added handholding, better information, cleaned up the help files and stuff but STILL and this is the clincher: even then, people just didn't bother reading the information on the screen.
Even when you first log in, there are a couple of pages of information that tell you what to expect. When you actually "arrive" in the main room, you get told of the useful help file to read. Before you register if you type a command wrong, it again points you to that help file!
Most never even found the "say" command. They would log on, scrabble with a few commands, ignore the friendly points on the screen and the automated robot that pointed them to help files and in the end give up.
In the end, I now ask people who want to link, to actually point to a website (see my sig) in an effort to stop people logging on and being rather clueless.
So what am I saying here? Nothing can ever be too user friendly. But it's amazing (and sometimes amusing) to see that even those people who assume that they are cream of the crop when it comes to IT issues get totally and utterly lost using something that we have both 18 and 40 year olds using with little to no IT experience at all.
The problem comes about when there isn't enough testing. We learnt a lot from the confusion of slashdot people, but unfortunately you get to a point where you just cannot do any more but hope that users think for themselves.
(As an aside, if you can read and can handle telnet and some basic commands - you only need 20 odd to get started - then feel free to drop by and chat, website is here)
Avantslash - View Slashdot cleanly on your mobile phone.