Advice for Developers: Make Common Usage Easy

Ken Hendrickson writes "Thomas Sowell has some fantastic common-sense advice for software developers from the viewpoint of an ordinary user: Make it easy to do what almost everybody wants to do. I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!"

Comments

It's just too difficult!

[Lord Grey]

From the bottom of the article's page:

Want to take action about what you have just read?
Then write a letter to your Members of Congress or your local newspapers, who you can find by entering your ZIP code in the boxes below. Also make sure to tell your newspaper editors that they should carry your favorite conservative columnists!
NOTE: Columns will not be automatically attached to the emails you send through this tool.

Sheesh. All I wanted to do was to forward this to my congressman! Now I have enter my zip code, wait for the next page to load, figure out who my congresscritter really is, rub my eyes after looking at god-awful red-on-green text, accidentally click the picture -- which shows a bio and is not what I wanted to do at all -- ....

<WHINE> All I want to do is forward an email! </WHINE>

Yeah

[Anonymous Coward]

Easier said than done. UI design sounds easy but it's not.

Re:Yeah

[happyfrogcow]

Easier said than done. UI design sounds easy but it's not.

that's what you UI people have been saying for years. It's like us developers saying "Writing safe, bugfree code is impossible." Bah!, i say. You're just saying it to demand a premium pay and more flexible schedule. UI's are simple! just look at Emacs.

(just busting your balls, i'm completely kidding)

Re:Yeah

[AKAImBatman]

IMHO, the reason why OS GUIs tend to suck is that there's no one to argue with. When the developer sits down, he thinks about it for awhile and develops what he thinks is best.

When a commercial developer works on a GUI, he first has to sit down with his peers, the art department, marketing, and eventually focus groups to yell, scream, and throw things. Out of these heated arguments tends to evolve a product that has a better balance between functionality, looks, and ease of use then what the developer could have produced by himself.

Of course, different companies have different focuses. Microsoft's focus is to pack features like crazy, then try to find a way to make it usable. Apple's focus is to make a product that does the core job first, then evaluate how necessary the extra features are.

Re:Yeah

[maxwell demon]

I disagree. UI design is very easy. For example, take MS Windows. Much too complicated interface. They should concentrate on their main function, and simplify the interface accordingly. Why all those clicks just to get a blue screen? Simply show the blue screen directly after booting! Damned easy UI, and it's not at all hard to write.

Now, of course there are some experts who want to do something else besides looking at blue screens, like infecting their computer with worms and viruses. This could be solved by just adding a button "Infect my computer" to the blue screen, and as soon as you press it, it connects to the internet (through a pre-configured, expensive phone number, so you don't have to go through the hassle of installing those dialers to increase your phone bill) and downloads a collection of the newest worms and viruses.

So, you see it's not hard to design an easy, intuitive UI. The UI I've described is much simpler than the Windows UI, without missing any important functionality.

Re:Yeah

[at_kernel_99]

Yeah, but he's not necessarily talking about the UI. His primary examples are a scrabble game that blasts music while he's playing and a chess game that takes a 'computer expert' to install. His issue is that all the fancy extras are in the way. He's not proposing that they be eliminated, but that the average joe who's not going to use them 90% of the time doesn't have to fight past them.

For example, my digital camera has all sorts of options for saving in different file types, different visual effects, etc. But if I hand it to someone - asking them to take a picture of me - they just point it at me & press the button. They don't have to struggle through the rather clumsy menuing system in order to do that. But if I want to learn all about that stuff its available to me. Thats the difference.

Re:Yeah

[Squishy Eyeball Jeff]

Apple has a veritable bible of UI design in its HIG, and Joel Splosky (of Joel on Software fame) has created his own tome on the subject.

Both are excellent reading for those interested in the art (science?) of good UI/usability design.

Re:Yeah

[SatanicPuppy]

LOL.

I'd love to see that guy try and use EMACS. Ctrl-X Ctrl-S is "Save/Quit" and Ctrl-Meta-PowerButton-Esc is "Go back to your fucking scrabble"

Re:Yeah

[zoeith]

You haven't spent much time on large opensource projects have you? OSS developers debate just like commercial software studios. It's called IRC.

Re:Yeah

[argent]

When a commercial developer works on a GUI, he first has to sit down with his peers, the art department, marketing, and eventually focus groups to yell, scream, and throw things.

You wish.

When a commercial developer works on a GUI, there's two ways it usually goes.

One: the art department and marketing come up with something that looks cool, and he has to do his best to implement it whether it makes sense or not. The focus groups don't get involved with the actual software until the 11th hour, and shipping fever is upon the company... up until then they're working with webpages and powerpoints that sort of do what the art department thinks is cool.

Two: the developer implements a prototype, and management tells them to ship it because they heard that Microsoft is thinking of making a similar product.

Re:Yeah

[cmacb]

"Oh no. It's the Monty Python Argument sketch all over again.."

No it isn't!

Re:Yeah

[Idarubicin]

IMHO, the reason why OS GUIs tend to suck is that there's no one to argue with. When the developer sits down, he thinks about it for awhile and develops what he thinks is best.

I don't know--if an OSS programmer writes something that is hard to use, two things will happen:

1. People will complain. Long and loud.

2. People will write patches or offer constructive criticism.

Some developers will design solid UIs from the start, requiring only minor tweaks. Some will create freakish monstrosities requiring many iterations and possibly a fork to fix. Some programs will remain unspeakably awful, probably because nobody needs their functionality enough to use them and demand repairs.

Remember, this sort of thing happens with closed-source proprietary products, too. IBM and Microsoft are well represented in the Interface Hall of Shame alongside many smaller developers.

It's tough....

[Mz6]

"There are cameras with features that you will never use -- and that will never get in the way of your taking a picture. Some of these are complex computerized cameras that have a "program" button you can press, so that you can take a picture without having to slog your way through innumerable options."

Since it seems he's used this camera analogy throughout the article I'll comment on this little blurb. I'm not so sure it's a very good analogy to use either. The fact is that if you want better pictures, you NEED to go through all of those "useless" features and change them. All of those values will change depending on the conditions, the lighting, and the activity your photographing. If there are those people that DONT care about those features, get the one-use ones. Hell, they even have digital one-time use camera now.

"Your automobile may have a global positioning system and other high-tech stuff, but you don't have to work your way through it before you can turn the key in the ignition and drive away. You don't have to work your way through all the options on your television set before you can turn it on and watch a ballgame."

Using the GPS and the automobile are not really related in that way. However, before using the GPS (not the car) you do need to setup a few features before using it. For example, adding the location for your house, the area in which you live or will be searching for addresses to. Now, if you want to complain about GPS and features/setup, let's talk about how they need to ALL be voice activated or touch screens capable...

As far as Televisions go.. this really isn't the case anymore. With more and more high-end TVs taking over the market and as they continue to do so in the future, thanks in part to HDTV, there will be a brutal setup process just to turn it on and start watching any kind of TV.

The point is that these devices/programs are being made for just about everyone they need to adapt to everyone's skill level. In the case of software development, it doesn't make sense to create several different versions of software with different default options turned on or off. A lot of the times this software has to be scaled to many different types of users on both ends of the spectrum. As a software developer myself, I try to make things as easy as possible that once the program is loaded they can begin their intended task. However, this may not always be possible all of the time.

I do agree with the following though... Stupid bundled software.

"It seemed like a simple thing to buy some new software with these reference works to put into a new laptop. But so much audiovisual stuff had been added, to make looking something up in an encyclopedia seem like a trip through Disneyland, that just installing it took so much time that it made my computer guru late getting home for dinner."

Re:It's tough....

[rjstanford]

Using the GPS and the automobile are not really related in that way. However, before using the GPS (not the car) you do need to setup a few features before using it. For example, adding the location for your house, the area in which you live or will be searching for addresses to. Now, if you want to complain about GPS and features/setup, let's talk about how they need to ALL be voice activated or touch screens capable...

What? Oh, please. This is exactly the kind of problem that we have in a lot of software, especially smaller projects. First of all, why should I tell the car where I live just because I want to find the nearest Taco Bell? A perfect (although unintended) example.

As far as Televisions go.. this really isn't the case anymore. With more and more high-end TVs taking over the market and as they continue to do so in the future, thanks in part to HDTV, there will be a brutal setup process just to turn it on and start watching any kind of TV.

My TV has an annoying tendency to go into a reconfigure-me mode if its been without power for too long. Oddly enough, it never loses its settings (weird). Anyway, you know what I do? Its the equiv. of Next->Next->Next->Next->OK but more annoying.

Even if it did lose its settings, instead of prompting me to check the convergence it could just power on with a set of defaults. Probe to see if a coax is attached. If it is, check to see if there are channels with signals on them. Check to see if there are powered devices on the line-in and component-in ports. All of that. Then it could stick a little note up in the main menu that says, "You have not performed advanced configuration yet. Doing so will result in a superior picture. Press (X) to configure your TV." Or something.

Re:It's tough....

[bob]

I'm not certain I agree with this. Few one-use or even just low-end point-and-shoot cameras have good lenses, shutters with wide speed ranges, or sophisticated metering. It is generally the case that, as these features are added into a camera, the manufacturers give the user the ability to configure them as needed. However, the program modes are also designed to take advantage of these features in optimizing the settings for the shot it is being asked to make. As a result, you in fact can take a much better picture with a high-end camera in program mode than you can in a low-end camera in its single mode. I've been using cameras -- focusing, setting shutter speeds & apetures, reading internal and external meters, etc. -- for over thirty years and have a pretty good idea what all those settings mean. But I find that for 90% of the pictures I take, the program modes in my Canon S1 IS do a better job, faster, than I can do, despite being able to configure any of dozens of parameters myself. I do not get anywhere near the same results out of a single-use or low-end point and shoot camera. There is value in providing those features, there is value in exposing those features to user configuration, and there is value in giving the user an "autopilot" to take computerized control over all those features when appropriate.

Re:It's tough....

[dasmegabyte]

In the case of software development, it doesn't make sense to create several different versions of software with different default options turned on or off.

I'll bite. Eckel, in his book "Thinking in Patterns," has a list of 9 design principles that good software follows. One of them reads "Simplicity before Generality." The idea is simple: if you sacrifice even the slightest bit of simplicity to make a program generally accessible, you have not solved the problem.

I've seen this first hand -- I worked with a guy who realized that the problem he was supposed to solve broke down into three separate interrelated variables. So he designed the program with three screens, to allow for maximum flexibility.

Unfortunately, nobody wants a three step process to replace what, for them, was previously a one step process. He's created more work to maintain a superior data model. And now we're having trouble selling it...people are asking wheter tracking and searchability and reporting are worth adding fifteen minutes onto each job.

I was working on a project at the same time that has a nearly UNLIMITED number of variables. But I took heed to the customer complaints, and delivered an interface that has a single form with a single required field: "notes." You can, from buttons on that form, change the date, grouping, result, step in the process, methodology of delivery, contact information and overall status of the project -- but all you HAVE to do is leave notes on what happened. Customers love it...and the reports I produce can be just as detailed.

OSS is often guilty of over complicating things in the name of control. Thing is, control is a BAD THING. Having to control something usually means that the defaults are wrong. Managing a program should be like managing a highly effective person...you should be able to say, "server this as a web page," and it should figure out how this needs to be done in the most secure and efficient way possible. After all, the computer knows WAY more about itself and its bottlenecks than you do.

Oh, and simplicity is not the same as brevity. It's not the same as talking down to somebody. When I drag an image in OS X, the Core Framework automatically converts it into the most effective on the drop target. That's not stupid, it's not underpowered, it's not idiotic...but it is simple. And that makes it powerful.

Re:It's tough....

[forrestt]

My car must be psycopathic. I've found my way home thousands of times, and I never told my car where I live.

The main thing I have learned from the slashdot moderation system is that the difference between funny and flamebait is the degree of clueness of the moderator.

Re:It's tough....

[rocketjesus]

What? Oh, please. This is exactly the kind of problem that we have in a lot of software, especially smaller projects. First of all, why should I tell the car where I live just because I want to find the nearest Taco Bell? A perfect (although unintended) example.

Ok. So we make the car have a "find the nearest Taco Bell" button. Because everyone wants to find the nearest Taco Bell. All those people that prefer some other fast food joint are SOL.

Of course, 75% of the population can't eat Taco Bell without getting the runs 15 mintues later, so we need to add a "find me the nearest public toilet" button.

Then when you get there you find the toilet is stopped up with about 30 pounds of mostly digested Nachos Del Grande, so you need to add a "find me the next nearest toilet, prefereably one not stuffed with other people's crap."

That proves a bit more difficult, and you wind up desperatley looking for someplace to go because of your high potty standards. You've also added a "nope, that toilet is covered in pee, find me another one" button and a "I'm not really sure what took place in there, but I bet the Pope wouldn't approve" button.

By that time you're at a gas station in a part of town you never even knew existed, there's a strange man in a fuzzy purple hat asking you if you wanna be his bitch, you're out of gas, spent all your money at Taco Bell, and have no idea of how to get home.

It sure would've been nice if your car knew where your house was. It's really too bad, since now you're turning tricks for the cash to buy gas and a map to get out of there.

Re:It's tough....

[iabervon]

A good camera will come preset to the normal conditions. If you want to take a picture focused at the center of the frame of something at a moderate distance in either good light or with the flash, you aim and push the button. The more your desires differ from these conditions, the more you have to fiddle with settings. That's what he means by making common usage easy. The special settings won't help at all for 75% of your pictures, because they're already set right. The other 25% could be improved with different settings, due to the lighting conditions and activity being different and uncommon.

No interest in doing

[danormsby]

"... zillion options to do things you have no interest in doing."

Like clicking on any of the advert around this article!

Excuse me, but...

[AKAImBatman]

Duh. Apple has made an entire market out of creating "easy to use" software and hardware. The trick is that it actually has to be designed and re-factored a few times before you have a cohesive product rather than a collection of features.

Case in point: iTunes vs. MusicMatch

No one is hearing.

[TuringTest]

Sad thing that software developers tend to follow the opposite advise: "Make it easy to do what is easy to program". It's the biggest mistake in interface design, bar none.

Re:No one is hearing.

[Anonymous Coward]

Sad thing that software developers tend to follow the opposite advise: "Make it easy to do what is easy to program".

I take it you neither are, nor have worked with software developers. I am and have, and I can tell you that is rubbish. More often than not problem is that (we) developers tend to want to try out even more ambitious interfaces that take MORE time to implement, than not. Only with some prototyping we settle for something that is minimum that "just works".

Not that this is really all that relevant -- missing thing in any case is time spent on UI design, gathering feedback, testing what works. Problem is not so much with developers -- we are perfectly capable and willing to do those tasks for techincal implementations (tune performance, refactor, learn new tricks) -- but with resources. Too often we don't have people working with us that are experts on UI or graphics design. It's not more time we have spend on implementing UIs, it's more time we have to spend on (RE-)DESIGNING and testing them.

Logic proves free software is the best

[OrangeTide]

* I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!"

I don't really follow the logic in that statement. Someone help me out, why would Microsoft not satifying their customer base suddenly make free software easy to use? (and how come as a long time open source user I never noticed this?)

Oh this is slashdot, I'm just supposed to assume that Free Software is better in all respects.

Re:Logic proves free software is the best

[Anonymous Coward]

I think the poster is saying that Microsoft has left an opening in this field by not satisfying their customer base. This is a field that open source can attempt to fill.

Re:Logic proves free software is the best

[LordNightwalker]

I think the poster is crapping out of his neck. From the description given in the article it is reasonable to assume that Thomas Sowell, the article author, does not use free software... In that he's probably correct... However, that still doesn't mean he uses Windows; he doesn't specify what OS and what software he uses. But yes, it's even reasonable to assume he uses MS Windows, given the description of the applications he tried to install on it. But... Does he complain about the OS? Hell no! He complains about third party software, presumably written by the $10-15 a program companies that flood the low budget end of the market. So, how exactly does this reflect bad on Microsoft?

BTW, if this article is representative of the kind of stuff Thomas Sowell usualy writes, I wonder why the hell the poster even bothered to post this tripe on slashdot. It's just one big uninformed piece of rant with the vaguest of problem descriptions, filled with bad analogies, and not even a hint as to how the problem might be remedied. The same type of senseless and meritless rant I have to endure every weekend when I go visit the old folks, when my old man goes on and on about how all politicians are greedy bastards who only want to fill their own pockets. I'd like to see him give running the country a go, behold how Belgium will go bankrupt in less than a month! ;)

Bottom line: if you percieve a problem, try to be part of the solution. If you can't be bothered to acquire the insight and knowledge to be part of the solution, then shut the fuck up and let more knowledgeable people deal with the problem. And perhaps this might come as a shocker to all those usability whiners, but did it ever occur to them to actually put some research into the products they buy? If you don't want a chess program with tons of bells and whistles, why the hell don't you check the back of the box to see what you're actually buying before carelessly dropping the damn thing into your shopping cart?

Then again, how serious are we to take someone who admits needing a computer guru to do a simple basic installation of a chess game, and who's so tech-savvy that he would refrain from playing a game of scrabble on his laptop in the middle of the night, lest he wake up the neighbours? If you can't even grasp a concept as simple as a volume button, you have no right to own an electronic apparatus of any design, let alone a laptop! And this guy is supposedly a senior fellow at the Hoover Institute in Stanford, Calif.???

Re:Logic proves free software is the best

[SpaceLifeForm]

No, *nix is not weak in this regard, it is actually the power of *nix!

You have the power to do it how *you* want to do it, for example:

ls | more
ls > /tmp/ls.out ; cat /tmp/ls.out | less
ls > /tmp/ls.out ; vi /tmp/ls.out
ls | grep "something"

You don't ever want software to guess what the user is thinking, or to pre-suppose that you know what the user really wants. Software that tries to be cute, fancy, or a mind-reader is almost always wrong for someone. In fact, software like that usually gets in the way of doing what the user is trying to accomplish.

M$ has that now

[millahtime]

M$ products are really easy to use for the people that use them.

To make OS products more widely used they have to be easy and intuitive for your common non-geek user to use. This is an area we have failed in before. The products that are easy and intuitive to use from OS do well.

Note to developers... this is a very very very big deal if you want your product picked up. It's not just how good your product is at doing the technicalities but how easy you can do them with.

Re:M$ has that now

[justforaday]

M$ products are really easy to use for the people that use them.

That must explain why I have to go through the AutoCorrect setup and uncheck at least 20 boxes anytime I want to sit down and use Word on a new machine...

forget the article, read this book

[studboy]

read "The Inmates Are Running the Asylum : Why High Tech Products Drive Us Crazy and How To Restore The Sanity" by Alan Cooper.

This book clearly and succintly states the difference between how programmers and engineers design (for the edge case), and how people really want things to work (make the common cases easy.) An excellent book, it could be used as a textbook but it's too short. Go read it.

Re:forget the article, read this book

[to_kallon]

programmers and engineers design (for the edge case), and how people really want things to work (make the common cases easy.)

Think about any programming/engineering class you've ever taken. How did your professor test your project? He/she ran all the edge cases against it so see where/if it failed. People learn from their mistakes, some of us anyway, and so they learn that they HAVE to anticipate and make provisions for the edge cases. Yes, the majority of users will never run up against those cases and yes it would make more sense to write a program that catered to the majority of those using it. But if you do that what happens to those of your users who run that edge case?
Summary: The reason programmers and engineers are taught to design against the fail cases is so that if and when they arise after production it is not a disaster. The reason programmers and engineers continue to do things this way is because they've been taught to.

On a side note; it is a good book, i agree.

Message To Developers??

[wbav]

Make common easy??


isn't that like:
Message to blonds.... Breathe in, breathe out?

The two demons...

[rah1420]

... commercialization and commoditization, strike again.

Mr. Sowell complains that computer programs aren't as easy to use as an automobile. Well, the first person to design the steering wheel probably didn't think to patent it; nor did the first person who put the accelerator pedal to the right of the brake pedal and make them thus and so. The auto UI "jest grewed" and became standard through market forces. It became a commodity such that it can't be patented, yet nobody dares to go against it lest they not sell a car.

On the other hand, the designers of software are careful to put a lock on every little feature that they come up with, ensuring that they wring the maximum value from its implementation. Are we ever going to see a ubiquitous interface? Not while the Patent Office lives. (tongue planted semi-firmly in cheek)

And all the bells and whistles? That's simply more commercialization -- let's get more out of it by climbing into bed with the people whose offers we bundle. And make it glitzy, and make it shiny, and make it loud.

I despair of ever seeing an end to this in commercial software.

Re:The two demons...

[kfg]

Well, the first person to design the steering wheel probably didn't think to patent it. . .

Especially since he was a ship builder before the founding of the British patent system. Before the adoption of the steering wheel automobiles used tillers. Philology recapitualates ontology.

Engineering evolution, like biological, is often additive rather than innovative. Sometimes this is a Good Thing.

KFG

Re:The two demons...

[dasmegabyte]

Actually, non-commercial software is usually MORE guilty of poor interface design than commercial software.

The solution? Why, free commercial software! Apple's free apps are some of the best programs I've used in terms of doing what they're supposed to and nothing more. AIM has a pretty good interface.

Incidentally, I don't find automobile interfaces all that easy. The pedals and wheel I can find, sure. But everything else is wherever the designer thought they'd look nice. For example: you're in a car you've never driven before. How do you shift it into reverse? Do you just shift it? Do you need to have your foot on the brake as well as the clutch? Is it over to the left and back, or over to the right? Do you have to push down on the clutch or press a button? Furthermore, where's the headlight switch? Where's the dashboard dimmer? How do you put the cruise on, or the A/C? Shit -- how do you turn on the wipers?

Umm

[Gentoo Fan]

that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!

When "Free Software" has a sizable amount of the desktop market then I think we can say that. Until then, how many years has it been "this year for desktop linux!"???

Joel has a little bit about this idea too

[BayBlade]

I found it a good read here

Re:Joel has a little bit about this idea too

[KevinKnSC]

The whole book is good, you should read it from the start.

Ditto for library developers

[tcopeland]

For example, if you're doing a charting library, make it easy to create a simple line graph - no legend, no colors, no logarithmic scales, etc. Just something like:

Chart c = SimpleChartFactory.create({5,12,8}, {"2002","2003","2004"});
c.renderToPNG("foo.png") ;

Don't force me to wade through a dozen classes which must be carefully assembled to make a chart - just make a simple facade that I can use in a few lines. You've done the hard work of creating the library - do the easy work of making a few classes to shield client apps from the complexity.

The FreeTTS guys does a good job in this regard - just a few lines of code gets some words going.

Re:Ditto for library developers

[danharan]

We could say "well, if you're not happy with the libraries, make the damn facade yourself", but that misses the point.

A lot of F/OSS project devs don't seem to care much if anyone adopts their tools. Those that do often count on consulting- and so what's the point of making it easy and including clear documentation?

Coding is only a small part of the work. To make things public domain in a way that really enriches the commons, we have to make them easy to use.

THis has been said before

[millahtime]

This whole discussion has been said many time on /. but for all this time most of the Open Source products I have used have had no improvements in this area. Either they aren't listening, are ignoring or don't know how to do this.

give me a toaster

[yagu]

I think Sowell "gets it" more than the technical community does. I think the camera analogy is apt -- make whatever product as sexy and complex as you want (WORD?), but make it transparent to do the most common tasks. Kind of like an appliance. Like a toaster... if you just want toast, you drop the bread in the slots, press the lever and you get toast. Any other "cool" features should be accessible but not overlaying the 90% use functionality.

I do technology for a living, and I STILL pull what's left of my hair out just trying to figure out how to make word stop putting bullets and numbers in front of my "paragraphs" every time I indent (please, no advice -- I haven't used WORD for years -- it's an illustration).

Re:give me a toaster

[thenextpresident]

It also depends on what software he is using. Not all software is goign to be easy to use, just as not all cameras are simple.

I have a simple camera. Turn it on, point, and shoot. Now, it's a digital camera, and gives me other options, but I don't have to use them.

He complains about software that he purchases. Fine, that's all good, but that doesn't make the software or product bad because he doesnt' like it. It just means he selected a product he didn't want.

Some people want the Encyclopedia on their computer that has the audio/visual aspect. And yes, the installation might take long, but that's a one and done thing.

He uses a car as an example: If he want's to use his car, he has to just turn the key. If it was only that simple: he has to fill it with gas, fill it with oil and other fluids, get it inspected, and what happens when their is a flat? He has to change it.

A toaster is a single purpose item: it has one purcpose, to toast. A computer is a multi-purpose item. It's used for many different things. You can just magically make a button on the computer that says "Do what I want you to do" (Well, you can, but will it actually do what you want it to do?).

While I agree that software is not always easy, his generalized statements are nothing new. Basically, all he is saying is software developers should try and make their software easy to use.

You know what? No matter how hard you try, it won't always be easy to use for everyone and have it do what they want. Heck, even Apple fails with the iPod if you really want to address issues.

Use Cases

[arudloff]

Design your software from use case perspectives to get a clear idea of what the user is actually doing with the system. Seems to me programmer's don't tend to spend a lot of time getting a strong idea of how the client is actually using the product. Focus your energy on the paths that 80% of the product use follows.

has he ever written a program?

[netdoode]

I'm not a programmer, but my job required me to learn some php/mysql to re-do a web enabled database because the people working on it couldn't get it done. So, I had a clear cut set of goals laid out. However, halfway through, someone called and needed another feature/field. Then another call. And another.

I ended up with an easy to use web GUI, but I had to fight to get the people to understand what they wanted wasn't parallel with what the database was designed for.

I'm not saying that all programs are great... there are a lot of junk ones out there. But I'm sure some people can agree, with changing goals and deadlines, finished products are often not what they started off being.

What is the saying, "You can have two of three... Cheap, Fast, Good..."?

What an inane comment

[EnglishTim]

"I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!"

Per-lease.
So now we've basically got a world of free software and Microsoft software and that's it? What the hell has this got to do with the freeness or otherwise of the software? Microsoft is not even mentioned in the article!

Re:What an inane comment

[aardvarkjoe]

Do you think that this article would have had a snowball's chance in hell of being posted if it wasn't tied into the Free Software Religion in some way? Blame the editors who won't take it upon themselves to edit.

The important word is can

[AndroidCat]

Free Software can perform better than Microsoft even in the ease of use area!

Convert that can to does, and you've got something. Fortunately Microsoft has been helping by shoveling new features into their Office products for many years. (Have to justify those updates prices somehow.) The only way they could help more would be to add a stupid animated paperclip to explain all those new features and changes to how to do simple things, but that would be stupid.

Um - what does this have to do with OSS vs. MSoft?

[Virtual PC Guy]

'I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!'

Yet the article is him complaining about a new chess and scrabble game that he bought. No mention of Microsoft - or even Windows.

For all we know - he is running a Mac. Sheesh people - get over yourselves.

Clearly Apple

[shylock0]

This man clearly uses a PC. A Windows-based PC.

And Steve Jobs is clearly the George Eastman he talks about.

I'm not a Mac fanatic (I don't own one, but I do work with them regularly), but it seems to me that this guy is clearly elucidating what is Apple's strategy: make stuff easy to use. For everybody. Without any pain.

I mean, this guy would *love* the free chess app that comes with OS X.

-Shylock

Re:oh dear god.

[slaad]

Although I couldn't agree with you more, it's very commonly overlooked. It's an engineering problem in general. How many times have you picked up a product and after only using it for a short while found ways that it could have been made better. Sometimes it really feels as though the designers never bothered to even test the dumb thing out. Of course you see this in software all the time too. Users are made to input data when the program could just as easily do most or all of it for them. Maybe the user ends up navigating through menus just to do a frequent task. It all seems like these things are common sense, and most of the time they are, but an amazing number of products make it to consumers with stupid problems.

You missed the point completely

[gr8_phk]

"If there are those people that DONT care about those features, get the one-use ones. Hell, they even have digital one-time use camera now."

So don't buy a camera with features if you're not going to use the features? His point is just to make a camera with features that I don't have to worry about if I don't want to use them. If that means a lower quality picture fine - it should be at least the same quality as the disposable without the features though. It should not be complex to not use the complex features. That's all.

Easy to use, yes... but features important also!!

[linuxrunner]

The problem with software is that we try and make it do everything so we can satisfy everyone!

This is where modular development comes in and where Firefox excels..

I started an open source portal, with simplicity in mind. It was great for what "I" wanted, but not everyone else...

Some wanted a membership only, some wanted to sell items, some wanted this that, etc etc... You get the picture. Different web site, different needs.

Of course no one wanted to program something for themselves, so I tried to accomodate them as the versions went up. Well, by the end of the year, I had this bloated / complicated portal.

Now I'm on the modular path... I really no longer have time for it, so others have taken over.

---------

Sorry to make a short story long, but the point of this one should be:

Make it simple - Fast - Easy to Use, and then allow modular capabilities to add everything else under the sun.

Don't try and make one program do it all.... Not everyone needs it ALL.

Cost.

[455]

Sometimes, the cost of developing incredible interfaces is not something a company wants to pay for. I'm a developer for my own company and I find that most of my clients want to get the task done (read "functionality"), with all options available, and would rather inform their staff and/or clients on how to use the software rather than spend double the cost to add ultra friendly interfaces.

That being said, of course I always try to develop the most user-friendly screens as possible, but sometimes business functions are just complex. Period.

Geek factor

[Quirk]

"Those who design some computerized products or computer software seem to have no interest in making it easy to do simple things, and will seldom tell you what to do in plain English if they can coin some new jargon instead. They keep adding features in such a way that even programs that were once easy to use become a struggle to deal with, even if you only want to do the same things you have always done."

This goes directly to the geek factor. Certain types of people like to interact with technology, whether it be primitive or in front of the curve. Learning enough about the ins and outs of the technology and production leads to epihanies, eureka moments and generally groking the thing at hand.

Waaaaahh

[AuraBorealis]

"This software is toooo complicated! It's too noisy! There are too many colors!" Geez... what is he, 90 years old? If he is smart enough to write a column online, he shouldn't need his own "computer guru" to install Encyclopedia Britannica.

For that matter, if it took his guru that long to install, maybe he needs a replacement guru.

Now I'm for simplified interfaces as much as... say 80% of the other respondents here, but there's nothing wrong with juicing up the user experience at the same time with some eye and ear candy.

It sounds to me like what Thomas Sowell really needs to do is learn how to use the VOL and MUTE buttons on his laptop. If he's unlucky, it requires some FN-key combination. If not, he's too dumb to keep on living. -B

Re:Waaaaahh

[yagu]

This is exactly the attitude that perpetuates (and widens) the gap between the geek community and their "clientelle". There are many highly intelligent people out there who don't know "our" language, and don't care to -- they just want to do work that was promised to be easier and faster by using a computer -- a promise rarely delivered.

It sounds to me like what Thomas Sowell really needs to do is learn how to use the VOL and MUTE buttons on his laptop. If he's unlucky, it requires some FN-key combination.

If he's "unlucky"?!? Then that would be about 90% of the experience -- My experience has been that I must figure out for each machine what it takes to turn down the volume or mute the sound. It's seemingly different (and obtuse) on each machine. And, you've only addressed ONE of his problems/complaints. He isn't asking how to turn down the volume on his computer -- he's pointing out how "noisy" the world of computer software is; noisy in the sense of interference.

I used to blame the idiocy of the users, their laziness and reluctance to learn until I bought a computer for my parents. One is a concert violinist, the other is a Doctor -- both IMHO of genius caliber intelligence (familial bias of course). It has been an ongoing struggle to introduce them to the metaphors I use daily -- our world of techdom is just that. OUR world... We need to overcome our hubris and live the experience through users' eyes.

Graceful scaling complexity

[plsander]

I think the point Mr. Sowell was trying to make with the camera analogy was that for 90% of the pictures I take, the "automatic" defaults produce a good serviceable photo. The advanced features (Program and Manual modes) are available and easily accessable when conditions or desire call for them.

I think Mr. Sowell would compare various programs that he complains about to the camera I learned to shoot on - my father's old Nikkromat. Manual everything, with a SLR light meter. Every shot required evaluating the shutter speed, film speed, f-stop, focus, depth of field, flash/no flash, etc. This was not a camera I could hand to a novice and tell them to "just shoot".

Developers and designers have to make reasonable decisions about default settings, and make those settings easy to change.

They also need to resist the urge to add every feature into the product. Does a chess or scrabble game really need to play music?

Re:Graceful scaling complexity

[pilgrim23]

Absolutely true. A programmer may be the most creative genius on earth but do NOT let him write the manual for his product! Mr. Sowell is a writer. He could have used an even better example: Word Processing. I can write a letter on a Fred writer Session on an Apple II. the program takes up about 48k on disk, has a CPU footprint of not at all, and creates a basically WYSIWYG image of a letter that when printed on 8.5X11 paper looks NO DIFFERENT then a modern letter produced in M$ WORD from a system that occupies gigabytes of space, takes triple digit mega-bytes of memory, and has more features then any sane man could want/use need. but hey its NEW and IMPROVED (tm). Seems to me, its the same old sheet of paper that comes out in the end...

Isn't this PERL philosophy in a nutshell?

[gblues]

"Make the easy things easy, and the hard things possible." This applies to a lot more than just Perl scripts.

However, the "easy" thing is not always so cut-and-dried. Maybe he wants to remove red-eye from his digital photographs, and maybe he's using Adobe Photoshop. Photoshop can remove red-eye, but that's not its primary purpose. Removing red-eye in Photoshop is going to be a bit more complicated than a program dedicated to red-eye removal, but that is not a fault in Photoshop. In fact, an experienced Photoshop user could probably remove red-eye faster than an inexperienced user could remove red-eye in a dedicated program.

This is where usability testing is key--why spend time on a feature that only a tiny fraction of your user base is trying to do? Which would you rather see happen to The Gimp: a red-eye wizard, or a Windows version that doesn't spawn a new taskbar item for each new window?

It seems like a "duh" comment to say "make it easy to do common things!" but you have to know what the common things are, first!

Nathan

Re:Isn't this PERL philosophy in a nutshell?

[ThisIsFred]

Great point! The reason why things are the way they are in the computer field is that lots of things are general purpose, along the lines of a general purpose desktop computer. So lack of familiarity with the software, and the previous versions of that software are quite an obstacle. There can't be a direct comparison with a camera or an automobile, because those things have a very specific purpose, and are not engineered for expandability. Think about how much more complex it would be to configure and maintain a vehicle that:

Ran on electrical, biodiesel or petrol, with the ability to swap powerplants when more power is needed.

Completely reconfigurable passanger space or cargo space.

Completely reconfigurable suspension.

Reconfigurable dashboard.

Could be reconfigured as a lawn tractor

Could be reconfigured as a camper

Could be reconfigured as a generator

I would imagine that the majority of the population would use this vehicle the same way they use a computer, learn the most basic functions and let the rest go to waste. I like your example of Photoshop, and I'd like to add to it: Photo Elements and Photoshop LE were Adobe's attempt at making the most common things the easiest to do. Unfortunately, I hate those programs with a passion, because I couldn't find the features that I wanted most of the time. You see, my first experience with PS was around version 4.0, and along those lines, I'm fairly comfortable with 7.0 as well. My lack of familiarity was my biggest handicap, while friends who had started doing image retouching using PE have no problems doing all the basics.

The most common mistake that non-techies make is the assumption that the computer understands the context the same way you do. It doesn't. The computer is a stupid machine that needs you to provide all of that during the course of the task. Think about your red-eye example. A human can look at the picture as a whole and immediately spot the problem. The computer needs to be pointed to the specific area where the problem exists, and needs your input to determine whether or not the filter routine has fixed the problem in a satisfactory manner.

No menus

[12357bd]

What about trying to learn from games, and for example; stop using menus? Those small labels on the upper part of the windows, there are a lot, but we seldom use a few of them.

Ergonomic interfaces don't present more than a few options at a time, if my memory es corerct there were studies about using more than 7 options as being confusing. If few options are presented, you don't need menus.

You completely missed his point...

[alexhmit01]

Since it seems he's used this camera analogy throughout the article I'll comment on this little blurb. I'm not so sure it's a very good analogy to use either. The fact is that if you want better pictures, you NEED to go through all of those "useless" features and change them. All of those values will change depending on the conditions, the lighting, and the activity your photographing. If there are those people that DONT care about those features, get the one-use ones. Hell, they even have digital one-time use camera now.

See, this is where you missed his point.

I have a fancy camera (analog), and a less fancy digital. With the fancy analog camera, if I want to take a family photo, I press the button. At most, I need to hit the clearly marked flash button to turn it on. Of course, if I'm feeling artistic, I may want to adjust the exposure, shutter speed, etc., and those features are all there. However, to simply to the most common operation, take a picture, I don't need to do anything.

Your attitude is elitist, "if you don't want the fancy features, get a disposable camera." Beyond the fact that disposables get expensive real fast, what if I want to have a single camera and be able to take real photos AND snapshots?

The point of the article is that the simple should be simple. If I want to take a picture, I press a button. When I install a dictionary program, instead of being interviewed by the program, let me quickly look up words.

The most common use for the references is a lookup mode, and the application vendor could certainly include a dictionary application AND a multi-media application.

I have an HDTV. Yes the DirecTV box required some settings (which are supposed to be done by the installer)... it asked for my zipcode for the guide. HOWEVER, if I just wanted to watch TV, I could have plugged in the box, turned on the TV, and let it auto-scan the antenna (this should happen on first use, instead of via menu, but it wasn't too bad).

I can adapt the colors, I can go into the service menu and tweak further, etc., for a reasonable picture I needed to calibrate the convergence, etc. However, if I just bought the TV and the HDTV box, on Sunday and set them up 15 minutes before kick-off, I could have been watching the game without problem.

The SIMPLE operation: watch a football game, is easy (could be easier, but pretty easy).

The COMLICATED operation: calibrate colors to the Avia disc, adjust convergence, etc., was complicated.

With MOST computer software, it wants me to go through a process to use the application. That is unacceptable.

Most SOFTWARE SHOULD run off the CD, or like MOST Mac software and be a draggable install (drag into Applications). Installers are bad (make them for unusual use), better search order for applications to it can be one Folder/Bundle is better.

If you have features that require libraries to be installed at boot-time, make them optional. If the library isn't there, no feature unless you run the installer.

Wouldn't it be great if simply RUNNING a computer program/game was as easy as playing a PS2/XBox/Gamecube game?

Sure the powerful functionality can be there for power users, but most people should be able to use your program without help. That sadly isn't the case.

Re:You completely missed his point...

[DarkMantle]

You missed the point of computers....

Wouldn't it be great if simply RUNNING a computer program/game was as easy as playing a PS2/XBox/Gamecube game?

With a PS2/XBox/Gamecube game they all have the same hardware so any settings that are hardware reliant can be pre-set. With computers we cannot assume the hardware platform, or Operating system to be run on. For example for a game or high graphics use program. What video card is used? how much memory does it have? And then there are things that we need to know for some other features.

Unfortunately, it's the 80/20 rule, 80% of people use 20% of the program, and 20% of people use 80% of the program. The big problem with this is... What 20% do they use? It varies from person to person... Do you use the Document Map in Microsoft word? I do, every time I'm working with a document more then a couple pages. Do you use Excel "If" functions? I do, how about VLookups? Do you use Nero's CD Text feature? I have (not often) but it saved me when I needed it.

The big problem is... as programmers we can't assume what features the customers are going to use 90% of the time unless the software is for one specific company, and only one use in that company. Take an Inventory system for example.. do the Cashiers at the store care how the stuff is entered to the Database? No, they just wanna be able to scan the customer product, and tell them how much it costs.... 80/20 rule... remember that

Re:You completely missed his point...

[siriuskase]

But this brings up the point, what should be simple? The author and yourself keep doing comparisons to what was previously done. My other camera did this, now my new camera does this and this, obviously the previous camera was SIMPLER. Or, my new cell phone gets text msgs, my previous phone did not. Obviously, my old phone was SIMPLER. Or, in the case of a dictionary, what if it automatically came up with synonyms of the word you looked up. But your previous dictionary didn't do that.

That's because for most people the replacement market is more important than the gee whiz! Nothing like this has ever been done market. Just because something has more features doesn't mean it must be harder to use. When doing the popular, age-old lookup task with the new dictionary, it shouldn't be any harder than doing it with the old dictionary. New features shouldn't get in the way of the old. Complexity doesn't equal good design. Every engineer has been taught that whether they practice it or not.

Scientific method

[TuringTest]

And most important... don't forget to test your cases with real people performing the task. Testing with real users is the best possible advice to follow when designing user interaction.

Also don't forget that, most of the time, the way that an user thinks about the application (the user "mental model") is really different than the designer's mental model.

This is hardly news

[khendron]

This is hardly news. It's been said many times before, and will probably be said many time again.

A lot of his rant has to do with all the unnecessary glitz and flash that has been added to what used to be simple software. One of the problems with technology today is that it has become too easy to add stupid unrelated glitz to basic information. This simply obscures the information.

For example, many (most probably) DVDs have these complete stupid animations that have to play when moving from one menu to another. I recently rented a movie (can't remember which) where you had to sit through 15 seconds of animation before the Special Features menu was displayed. It wasn't impressive, it was just annoying.

There is more and more of this every day. It seems that media and product producers do not have any really new features to add to new releases, so they just add some unnecessary glitz and animations and sell it as a new version.

The producers of Scrabble should take a hint. The Scrabble board game hasn't changed in 50 years, and it's still popular. Some things just don't need new features.

Re:This is hardly news

[DrVomact]

The problem arises from the fact that you can't make money selling "classics". If software company XYZ publishes a simple, usable program that does something many people want to do, then XYZ can make some good money...for a year or two. But then what? Everyone who needs the program has a copy (or a pirated version). Sales drop to near zero, and everyone who used to work for XYZ is out looking for a new job.

There are two ways XYZ could stay in business: they could exercise their imagination, pay the development costs, and bring out a great piece of software that does something totally different than their first hit. Or they could just add new features to the "classic" and sell it as NEW! IMPROVED! MULTI-MEDIA Version 2.0!

Guess which approach most companies take. Sure, some of the additional features might actually be beneficial to (usually) a minority of users, but they also make the program more difficult to use and resource-hungry. After a while, the "classic" becomes so encrusted with "bells and whistles" that doing simple things makes your head explode. But that's the price of "progress" in the software world.

OSX useability overrated and degrading fast

[Ars-Fartsica]

Once upon a time Apple had rules for presenting interfaces. All was well. People followed the rules. Then eye candy came to town. It started with the quicktime player departing from Apple standards. Now the dashboard crap. And why is it some apps use Aqua and some use Brushed Metal presentation?

At this point I don't think you can strongly claim that OSX is any more useable than Windows or GNOME or KDE.

Re:OSX useability overrated and degrading fast

[GlassHeart]

One, in theory, brushed metal is supposed to be reserved for applications that mimic real world objects. For example, iTunes mimics a stereo, so it has a brushed metal face. Both Apple and third parties have violated this guideline.

Two, usability is not a static thing. We have a computer userbase that is significantly savvier than the guy who walked into a computer store demanding a "VisiCalc" not knowing he needed a computer for it. This means they recognize common widgets like scrollbars, drop-down menus, buttons, checkboxes, and others, no matter what shape they're in. With the popularity of $200 game consoles, a lot of people have prior experience dealing with customized (that is, "weird") computer UI.

Three, there's another aspect of usability that isn't static. Users learn. While you can compare the efficiency of first-time computer users introduced to a Mac or to Windows or to KDE, that's a pretty pointless benchmark. Experienced users in each platform probably perform just as well as each other. The most significant difference among them is probably the amount of time spent in maintenance.

Bill Gates = George Eastman

[Da VinMan]

I read the article, and I was unimpressed with the author's pain. Perhaps he doesn't know what it USED to take to use a computer? Does he have any idea?! This is what Bill Gates really succeeded at: making computers easy enough for the masses to use. Why else would 90%+ of all PCs be running Windows? Why else does at least 50% of America have a computer? Personal computing as we know it today would not be nearly as ubiquitous as it is if Bill Gates had failed.

Maybe Mr. Sowell and his "guru" are right in that there is definitely room for improvement. But, I would prefer thankyouverymuch that ALL computers not be dumbed down any more than they are so a "guru" like the one Mr. Sowell employed can go home for dinner on time. How does he know that I do not use the features he considers useless? Why should I be stuck with a crippled product because other users get easily intimidated by the product's other features? If Mr. Sowell truly wants a simplified experience he should just go use Apple products, which are already dumbed down and streamlined. But he should be prepared to shell out the extra cash to support Apple's entirely proprietary architectures.

Troll Warning: Isn't this just a technophobe rant?

[mykepredko]

I've seen articles like this for years and it seems to always take issue with the things the writer doesn't know how to do compared with the things they do and seem intuitive to them.

I would pick holes in just about all his arguments - he seems to ignore the initial training and years of condition on how a car works. The same with a TV set. I'm sure I could find somebody that has horror stories trying to figure them out for the first time and could write an article on how counter-intuitive these items are (like, why do you need a key for a car ignition when you've unlocked the door?).

I've seen many articles like this on the VCR, not to mention ones complaining about more sophisticated cars, kitchen appliances, telephones, heating/air-conditioning systems, all of them wanting the systems to be simpler (and most, like this one, wanting to return to simpler times).

One of the things that infuriates me the most about this article is that the writer doesn't try to do anything himself; his "computer guru" doesn't seem bright enough to be able to load software without getting his mom angry because he is late for dinner.

In any case, if he really wants to play scrabble simply, why doesn't he drop twenty bucks (probably less than he paid for the CD) and buy a hand held scrabble game?

Sorry for the Rant - I would be a lot kinder if the writer had tried to load an application, got a GPF and ended up in phone support hell between the ISV and Microsoft with each blaming each other and the theme of the article is that he just wanted it to work.

myke

It would become easier

[2names]

if the developers would actually take into consideration how the interface is going to be used in the Real World, not the world where everyone is a techno-whiz-bang-'puter-gu-ru-genius.

This is serious, people. I do not know of any other product where the designers/developers are so far removed from the end user. Something that makes perfect sense to a highly trained, technically capable person will make absolutely no sense to a person who has trouble remembering 2 passwords. Really.

the Seldon Patent

[blitz487]

Actually, they did patent it. See the Seldon Patent, which was a huge problem for early car makers. Similar patents encumbered early aircraft makers.

Blame it on magazine articles

[mangu]

Whenever you read an article comparing several different softwares you see the same thing: a table comparing features. People chose the one that has more boxes checked. The end result are menus with thousands of commands.

With cars, the situation is different. First, as you mention, the UI for automobiles has stabilized long ago. The last significant modification was the automatic transmission. Before that, the last mass-produced car with a different UI was the Ford model T, which had a separate throtle pedal for reverse. Besides, cars today are compared for marketing purposes with features like style, power and speed, not the raw number of options, like software is.

Design Of Everyday Things

[rjstanford]

That's another great book for UI design. It talks about the UIs that are all around us. Ever encountered a glass door (shopping malls are bad at this) with a bar that goes all the way across it, mirrored on the other side? Do you push, or pull? On the left or on the right? Its not difficult, but it is an example of bad UI design. Contrast that to a door with a flat "Push" panel on one side, backed by a protruding handle on the other. You now whether you're pushing or pulling, and on which side of the door to perform the action. No documentation required, almost zero chance of failure - this door wants to be opened and makes it easy for you. A lot of computer software has exactly the same design mentality as the first door. Or worse, because the door has been "skinned" to look like a slice of pizza as well.

Required Reading

[TreadOnUS]

This book was written in the 80's but the concepts are timeless. It's not software specific but it is an excellent primer for designers and engineers of all types.

The Design of Everyday things

Classic designs, or Software isn't a camera.....?

[blackhedd]

Everybody is talking about how difficult it is to do UI. Yup, it sure is. Just ask the guys who did the classic industrial designs- the Dreyfus telephones, etc etc. Lots of examples in addition to the Kodak Brownie.
The best designs become classics, and really change the way we work and live. And there really aren't all that many classics, vastly fewer than the number of designs that try and fail.
So why aren't there more really classic software designs? Part of it is that all of us programmers have drunk the koolaid about uniform interface designs. They simplify learning by creating references to things previously learned in other contexts. But a real "classic design" is easy to learn because it's *internally* coherent- its reference points are meaningful in terms of its own functionality. If there is complexity, it maps directly to the problem domain and not to the UI design. That makes it far easier to deal with, because it "just makes sense." The iPod is a very interesting recent example, but I can't think of something analogous in the realm of pure software.

Maybe if we break out of the box on UI design, then we might be able to stop complaining about how stupid our users are. After all someone who uses something I wrote is supposed to be *smart* not stupid :-)

OT: But...

[Da VinMan]

Will slashdot ever drag itself into the year 2004 and provide the ability to edit posts?

I sure hope not. Now being able to add an addendum, I could agree with. But, even that is risky.

Consider for a moment that there are always active trolls who repost previously 5 star posts just to get karma from unaware mods. Now take the case of an editable post. You can get the post modded to 5 then swap the contents out with a porn troll. Not pretty.

It's hard to make things easy

[p3d0]

As an ideal, software should make simple things simple, and complex things possible. Both of these require talent, but the former is certainly the less glorious and more thankless. If you are highly skilled, and design your software meticulously with usability in mind, you can make a software task appear so simple that users wonder why it took you so long to write.

Reactionary...

[MojoRilla]

I generally agree with the principle that simple things should be easy to do. When you take your keys out of your car, your headlights should go off. That is what 99.9% of the users intend. It should be harder (require a special button or something) to make them go on if the keys are removed.

Mr. Sowell, however, seems pretty reactionary about software change.

He is upset that his scrabble vendor released their game on CD. He would rather have it on floppy disks, which are more expensive to produce. And, some machines now don't have floppy disks. This complait has no merit.

He is upset, that the scrabble game he has plays music. Probably because his old game didn't. What he doesn't consider is that most users of this product probably want the music. Products should ship exactly like this. Turn on all the options that the majority of the users want. Make those in the minority use the preferences.

Software is going to change, and make more use of increased hardware capabilities. There is no stopping that. Although there is some truth here, there is a lot of pointless ranting.

This is a stupid article

[GoofyBoy]

In terms of UI and usabilty there are alot better ones out there.

>Too many other computerized products and computer programs, however, force you to get bogged down in so many options, functions, and modes that you may just give up before finding the simple thing you want to do.

In a windowed program, there are menus. Don't want the options, don't go hunting for it.

>Today, it takes a CD to hold all the bells and whistles that have been added

No the reason why they use CD is not because its complicated, its because it cheap to mass produce. A program is > 2 megs and If you are awake in the middle of the night in a hotel room and your spouse is asleep, you would never dare to turn on the new Scrabble game.

Its called a volume control. Either built in, on the OS level or the physical speakers has them. What would the user want?

>Since my old computer chess game will not work on the new computers, I had to get a new chess game

But when you bought the old chess game it didn't specify it would work on the new OS? And this is the programmers fault for not making things compatible with technology 10 years in the future?

FireFox gets it right...

[Apathetic1]

This is something I think FireFox has gotten very right. Don't want to mess around with settings? Great. It works right out of the box.

I've installed FireFox for about a dozen people now. So far only two have even bothered to open the Options dialog. They don't care how the options are set, as long as they can browse. The two who have opened the Options dialog think the customizability is great but those two are not the majority of users.

It's all relative

[dg41]

Ease of use is in the eye of the beholder.

Okaaaaay

[SatanicPuppy]

If installing and playing a CD version of "Scrabble" is too much for this guy, there is no way he is ever going to be happy.

And Jesus, write my congressman? "My computer is hard to use, I want you to make it all better."

I've seen poorly designed software, with poorly thought out UI, but its a big step to go from that simple fact to some blanket article which just says what we all already know (User Interfaces should be intuitive and easy to use), and doesn't even address the issue of HOW to make an interface which is intuitive to everyone, including comptuer illiterate scrabble players.

It always pisses me off when someone jumps up and starts complaining because a lifetime spent doing things BESIDES using computers hasn't prepared them to be able to look at a screen and immediately understand what to do. Sorry folks, sometimes, even with the best interface, you have to RTFM.

I've got an even better advice ...

[maxwell demon]

Make easy usage common!

Re:Troll Warning: Isn't this just a technophobe ra

[MSBob]

Agreed. Cars are anything but intuitive.

Case in point. I once taught someone to drive who had absolutely no idea what the function of the clutch was. He knew that he had to press it when shifting gears but didn't have a clue why. In fact he would at times forget to depress the clutch pedal especially when switching to neutral. He probably "discovered" that it was possible to do and seemingly no malfunction ocurred. It was only after I drew him a couple of napkin diagrams and explained that clutch was disengaging his engine to prevent the damage to the gearing system that he fully comprehended the gravity of his driving errors.

Cars (esp with manual gearboxes) are definitely not intuitive.

There is no try

[elmegil]

Free Software can perform better than Microsoft even in the ease of use area!

Can and will are two different things. Given the common OSS attitude of "you haven't read the docs, fuck you", I can't see ease of use being a priority very soon, except in certain niches.

Use Case Driven Design...

[Dana P'Simer]

could be the answer. The problem Sowell points out stems, I think, from the mindset of the average software engineer. Put Simply: "If you want to do something new, add a function to do it once, and then you can use that whenever you need to do that again." This is an extreemly powerful concept that when extrapolated and compined with other ideas results in Component/Object orriented designs, modularity, and code reuse. The problem is that average people don't think that way.

The idea that open source can fill this bill is laughable. Not that I am against open source. And I encourage them to give it a try, which I am sure they are, but I have not seen a single truely innovative UI out of an open source product yet.

The fact is, UI design is not easy and it should not be left up to amatuers. And yes, most software engineers are amatuers when it comes to UI design. In fact most of the top developer's I know admit that they suck at UI design and look forward to input from someone with that particular skill. That is not to say that engineers are incapable of good UI design, they just need a completely different set of skills from what they use in thier normal design and programming duties.

A good UI designer would need programming skills, creative skills, a knowledge of ergonomics, and an understaning of how people want to use the product. This is best discovered through a process at least similar to the Use Case process.

I think a hierarchial approach should be used..

[carlmenezes]

At level one, do what 90% of all users would do.
At the next level, allow the remaining 10% to perform more complex tasks.
Maybe have one more super-tech level where the elite 1 or 2% can delve in and tweak.

Example 1 :
Web browser app like Firefox :
level 1 : It installs itself, all plugins and figures out how to handle almost any mime type.
level 2 : Extensions & more
level 3 : about:config, etc

Example 2 :
Web Design Software like Quanta Plus :
level 1 : WYSIWYG interface that produces nice clean W3C compliant code and maybe buttons called "text effects" for stuff like mouse-overs that allows the user to see what will happen to the text.
level 2 : Code View, CSS Editor, etc
Level 3 : Javascript debugger, PHP debugger, MySQL queries, etc..

Basically, you should never be forced to descend to levels 2 or 3 to be able to accomplish a basic task.

Not Microsoft

[Zebra_X]

How the hell can you come to the conclusion that Thomas Sowell is using Microsoft's products, furthermore, that he thinks that they aren't useable?

Ken Hendrickson says "I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft even in the ease of use area!"

After reading to the article I noticed that Thomas refers to "computer programs", not "Operating systems, office applications, e-mail, or web browsers." The only particular type of program that he specifically mentions are a "dictionary, an atlas, and an encyclopedia". There are a number of manufacturers of these types software. Crap like this is usually found in the $5 bin and comes with a beautiful VB installer, and accompanying VB application used to browse the content on 100 cd roms of uncompressed video (in the case of a multimedia encyclopedia). Though given all of Thomas information on what is rubbing him the wrong way, there is no way to deduce which particular products Thomas has been using given the words in the article.

I'm not going to jump to any conclusions as to what he uses, becuase I can't. The point regarding usability is well taken, and should always followed when building an application.

Although, with people with Ken's superior power of reasoning and logic, it's a wonder more products don't turn out better.

Asinine

[DeltaSigma]

Why must people insist that it's just as easy to make a graphical interface as intuitive as a hardware interface?

Let's take a step back here. What's the difference between a computer and a hammer (from an interface perspective)?

They're both tools, used by people, to accomplish various tasks. Why can't a computer be as easy to use as a hammer (or any other elegant physical tool)?

Because we can't touch them.

It's that simple. In the hardware world we have hammers, to cars, to ipods, to... let your imagination go wild. It's entirely easier to make a hardware interface user friendly because a hardware interface designer can use their own intuition, and the intuition of others to make it easy.

Anyone ever ask themselves why hammers have a handle and a head? Anyone ever ask themselves why a walkman's volume is controlled by a dial? Do you commonly wonder why turning the steering wheel on your car turns your car?

Physical things can be made an order of magnitude easier than programs merely because they are physical things. The human mind easily pierces most facets of a physical object within seconds of its observation. The option to handle it makes it even easier.

Computers have a very serious handicap. We can't interact directly with our computer! Under most circumstances we have a keyboard and monitor as standard input and output. So we have our hardware interface. But this interface doesn't directly control the computer. We have to use this hardware interface to work with a software interface.

It's this simple little factor that trips people up: Interfacing with an interface.

To get a real life analogy of operating a program, use a hammer (and only a hammer, no hands) to operate your vehicle. No hands, no feet, use the hammer to accelerate and steer. Hell, I'll make it easy, you can use a hammer in each hand.

This being said, we've still got SO MUCH left to do in the graphical interface world. There's so much experimenting left to do, so many advances we've yet to make. Expecting the relatively young computer industry to produce interfaces that are as easy as interfaces that have been around since the stone-age is insane!

Camera analogy still a poor one...

[Otto]

See, this is where you missed his point.

I have a fancy camera (analog), and a less fancy digital. With the fancy analog camera, if I want to take a family photo, I press the button. At most, I need to hit the clearly marked flash button to turn it on. Of course, if I'm feeling artistic, I may want to adjust the exposure, shutter speed, etc., and those features are all there. However, to simply to the most common operation, take a picture, I don't need to do anything.

Your attitude is elitist, "if you don't want the fancy features, get a disposable camera." Beyond the fact that disposables get expensive real fast, what if I want to have a single camera and be able to take real photos AND snapshots?

The point of the article is that the simple should be simple. If I want to take a picture, I press a button.

Last I checked, one shot picture taking *was* just that simple, on every digital camera. You don't have to set all the options. On my Canon Powershot G2, for example, if you have it in the Auto mode (the default mode) you press the button and it takes a picture. Even the flash is in automatic mode by default. Done and done.

You might get poor pictures if you don't mess with the options, but that's true with even film type cameras (more true with film type cameras, in fact, since digital cameras have a lot more auto-correction capabilities, not to mention all the software auto-correction that exists).

Can't agree more.

[SharpFang]

Append some extra data to a CD...

mkisofs -C `cdrecord -dev=0,0,0 -msinfo` -M -dev=0,0,0 -J -r -o image.iso ./source && cdrecord -dev=0,0,0 -v image.iso && rm image.iso

So exceptionally intuitive...

In Defense of the Complex Machine

[Alzheimers]

In defense of PC, let me point out that he trying to compare machines with a specific purpose to those with a general purpose and is upset that the level of complexity is higher. With all due respect, No Shit Sherlock! Even something as "advanced" as a digital camera has, in essence, one goal in life: to take photos. Toasters, televisions, even cars are designed for very specific tasks -- any extra features are an added "complexity".

Computers and Operating Systems have no such luck. They must act as their own human-machine interpreters to an infinite number of possibile inputs and commands. Some of these commands are simple and can be optimized for -- eg, run program. Most, however, have their own set of additional complexities -- functions such as printing a document, manipulating data, and searching for files all have so many possible outcomes that more specific instructions than one-button "Do This" interfaces are required.

Not to mention that every user will have their own opinion about how their interface should optimized. The "complex" interface is a good thing because it gives the users complete freedom over their interactions with the system.

But what about users who don't want that freedom? They want machines with a big "Do this" button -- cameras have one, cars have one, toasters have one. Computers, by their very nature, can *not* have one -- the set of all correct answers to "Do this" is infinite. How? Where?

The bottom line is that the computer cannot read your mind. It cannot perfectly and accurately translate simplistic commands into complex functions. The best it can do is try to predict what you mean and give feeble human-to-machine translation function to the rest.

Re:In Defense of the Complex Machine

[goon]

Not to mention that every user will have their own opinion about how their interface should optimized. The "complex" interface is a good thing because it gives the users complete freedom over their interactions with the system.

homo logicus, Allan Cooper author of Inmates running the Asylum and creator of Visual Basic calls this. Software developers, coders call them what you like actually are very different to ordinary garden variety software users. Garden variety users run software to achieve goals. Cooper outlines in the book *goal directed* development that prescribes the interaction of the user to achive their goals and alligns user interaction with the software created.

This is not simply interface design ~ the sort normally tacked on the top of a product by graphic designers but a form of user interaction contract between user, user interaction and the underlying software completed before gui design, software engineering, coding etc. Think of it as a form of a user interaction interface (in the coding sense) that allows users to achive goals. Cooper pointed out this part is negotiatable with developers but not to be used as a guideline, but as a specification to be followed.

Having more power does not necessarily allow them to achieve their goals any better. In fact more complexity can actaully increase cognitive friction, a term describing the mental model you must construct to work the software to do something. Think about the cognitive friction the next time you want to create a document in MS Word ~ a product with a complex interface whose goal is essentially to create text documents.

Re:Bill Gates = George Eastman

[poot_rootbeer]

This is what Bill Gates really succeeded at: making computers easy enough for the masses to use.

I'm going to assume that you're talking about Apple BASIC here... only a fool or a Microserf would give Gates and Windows the credit for making computers usable by regular people.

This principle should also apply to APIs.

[exp(pi*sqrt(163))]

If I just want to plot a simple graph using a C++ API I shouldn't have to write dozens of lines of code to specify what icon and cursor I want, to install an event handler, to actually write an event handler, to specify in gory detail exactly what format my pixels should be in and so on. On the other hand, I do want to be able to do all of the things I've just described if the need arises.

depends on your budget.

[twitter]

I'll agree that most users would be better off buying a Mac than a brand name PC. It's always been the case that the brand name PC costs as much or more once you add all the goodies you get with the Mac. Mac, since Jobs came back, has earned it's price especially with their laptops. Those who don't mind having a software master are well ruled by Apple.

Those, like Mr. Stowell, who simply want their old computer to work and do all the things it used to might give free software a spin. Most people are pleasantly surprised to see their old computer come alive again with Knoppix. Windoze PCs that won't boot anymore are great for such demonstrations. Mepis is the easiest of the Debian based free beer distribution I know of to install. Sarge is not much more difficult and is cleaner as well as more free as liberty. People on a budget will be happy that their old PC once again plays games they love with sound they can turn off.

Free Software can out-ease-of-use Microsoft?

[JessLeah]

Yeah... right. I LOVE free software, and I HATE Microsoft (crap, I run Debian and OpenBSD _ONLY_), but let's be freaking realistic here. If free software is so easy to use, then why did I have to send THIS email to the OpenOffice users' mailing list this morning:

Subject: scalc: How do you select WHICH row or column is the X or Y axis???

Yes, I have viewed Help. Yes, I have Googled for more information. Yes, I have searched the FAQ.

But for the life of me, I cannot figure out-- and this sounds really dumb, I know-- how to tell scalc "Use column A, 'Amount', as the Y-axis, and column B, 'Date/Time', as the X-axis'.

I am trying to have OOo do a very SIMPLE line graph here-- how much money is in my account, graphed against time. Very very basic stuff, the kind of thing they teach to fifth graders. I cannot manage to convince scalc to do it.

And God help me if I wanted to keep 'Date' and 'Time' in separate columns, and have the software know to parse a single row's 'Date' and 'Time' cells as one date/time object...

Can someone please, please, please tell me how to do this? In M$ Excel, it's really really easy. You just tell it which column to use as which axis. It's so simple even a 10-year-old could understand it. IN OPENOFFICE.ORG YOU ARE NOT EVEN GIVEN THAT OPTION! You ARE given a billion different choices as to how you want the chart to LOOK, and by diddling with the chart object once it's in the 'sheet, you can give it a spiffly gradient effect, or change the labels on the axes...

But nowhere can you actually tell it what variables (read: columns) the axes should be bound to????

Just as an aside, I am really going nuts here, and pondering just going to Microsloth Office (running in CrossOver, as I run Debian). I appreciate the effort thot everyone has put into OOo, but seeing as how I CAN figure out how to insert a 3D rainbow-coloured torus into my spreadsheet, but I CANNOT figure out how to do a simple line graph, I'm kind of miffed at the moment. It would seem that the people who handle 'useability' had their priorities hat on backwards? ;) I mean, are rainbow-coloured tori really more important than line graphing? ;)

You just don't get it, do you!

[lcsjk]

You don't get it do you. Ranting without stopping to understand the problem just gives the world more poorly thought out software. Just because you have expertise in computers or some other area does not mean everyone does.
For instance, in MSWord-97, I type file/new and it asks me if I want a blank document or blank template. Duh!!
Try to load your old Netscape or Mozilla mail files to your new computer so you can continue seamlessly. Same for the bookmarks.
Think ahead, not behind. Programs do not need bells and whistles to be good, but they do need to do the things people need done.

Re:You just don't get it, do you!

[SoSueMe]

I think the point was more to the effect of getting a load of "features" that have nothing to do with the primary intent of the original program.
For the Scrabble example, all he wanted to do was play a nice "quiet" game. Now there is so much cruft installed, he has to remove some of the added "features" that he didn't want anyway and couldn't select just the ones he did want.
For the dictionary, etc. the unwanted multimedia seemed to be the issue.
Some of us are more of the "just the facts" types, but often there is no alternative except to take the force-fed install and tweak out the superflous stuff afterwards. Like some OSes we know.

You will never please all the people but neither should we dismiss their critiques either.
Everyone has their story and opinions. All are valid to some extent.

Huh?

[suwain_2]

I don't believe he uses Free Software; that means that Microsoft is not satisfying their customers, and Free Software can perform better than Microsoft.

What? You don't think they're talking about free software, so it must be anti-Microsoft, so it must be pro-free-software? I think what's being said is that software as a whole is sometimes not laid out with usability in mind. It's not saying Free Software is better. It's not saying it's worse.

Am I completely misinterpreting what was said? Or is this the most ridiculous comment ever made?

I've just started Microsoft Word...

[deischi]

.. and it absolutely was easy to do the most important simple thing:

I cut immediately start typing text.

And I can do the most important tasks by pressing easy to find buttons. From my experiences with beginners: they all could that within 1 minute. Learning how to move around with the mouse usually took more time.

Now try explaing a novice how to use LaTeX.

PS: "that means that Microsoft is not satisfying their customers," - The Scrabble mentioned was probably not produced by Microsoft!

PPS: I would never use anything but LaTeX for a text with more than one formula.

Photocopiers

[AlpineR]

The author sounds a bit cranky, like Andy Rooney on 60 Minutes, but the underlying point is good. I have encountered a particularly bad example of features interfering with simple tasks:

My university library replaced all the photocopiers with some fancy shrink/enlarge/collate/digitize/duplex models. In my mind, the one feature a photocopier should present is the ability to lay down a page, press a button, and get a copy. But no, these machines required you to enter the paper size, number of copies, and cropping options first. And then, once you copied one page, you had to do it all again for the next. I'm sure these machines are very efficient for a person who has some complex copy jobs and is trained properly, but they are inappropriate in a library where most people have simple tasks and will never use the same machine again.

In programming, I try to follow the theme of keeping simple things simple. In my C++ class for a random number generator, you can initialize and seed the generator with no parameters. The code gets a seed from /dev/urandom or time() on its own, since that's what most people would do anyways. If someone wants to be more careful, they can do the seeding themselves, but software should always allow simple tasks to be performed easily.

AlpineR

In other words...

[porter235]

Make Easy Things Easy and Hard Things Possible... sound familiar? Now just apply it to the UI!

Not quite.

[MisterFancypants]

Anyone who thinks Microsoft doesn't spend a lot of time trying to make it easy to do things people want to commonly do is completely out of touch and hasn't heard of "Activity-Based Design". This stuff is hammered into people who work for Microsoft.

Why does Excel have such easy to use list-making functionality when it is supposed to be a spreadsheet? Microsoft did a lot of user testing and found that an awful lot of customers just used it to make lists. So they made it really easy to do so. That's just one example.

In short, the view that you should make the common tasks easy is completely on target. The idea that Microsoft is unaware of this and doesn't follow these ideas themselves is completely wrong and has no basis in reality.

read Sowell's other articles too!

[RussP]

Wow! That makes my day. An article by Thomas Sowell featured on slashdot! He's a great writer, and he happens to be a conservative black too. Please read some of his other articles too while you're at it.

about usability

[IshanCaspian]

Someone once said "unix is very simple, but you have to be a genius to appreciate its simplicity." The point is, sometimes interfaces that are "easy to use" become less functional as you become more and more proficient in using the device that the interface allows access to. I think the entire linux console UI is one of the greatest interfaces ever made. I can quickly configure everything, recompile my kernel, browse the web, and so on using one consistent interface. A truly well-designed interface, no matter how complex it is, is always easy to use if you have the sufficient skills to use it. This is why that dude who designed Metacity is dead wrong in saying that preferences are the root of all evil: interface usefulness is dependent on the user's experience.

Think of it this way. If I don't know what the /etc folder is, and an interface hides that, it's made my life easier. If I do know what it is, it has made it harder. There is no absolute "easiness" in any interface because it's all dependent on the user's skill.

Really, it's really confusing to lump so many things under "ease of use." There are three distinct levels of UI, really:

1) The wizard. This means a user wants to have the computer hold his hand through the whole thing. ...e.g. he won't know what he's looking for even if he's right on top of it. There should be a wizard for every aspect of configuring a computer that a newbie would need to do. Windows knows this, and that's why newbies consider it easier than linux in some situations. Wizards are frustrating to the user that knows what he's doing because this represents a tradeoff...it's easier to accomplish a certain task as the designer conceived it, but it's more time-consuming to tweak things down to the letter. Windows abuses the wizard, forcing me to use them when I'd rather just tweak a text file.

2) The GUI. If you've moved off the wizard, you've progressed to the point where you know what you're looking for when you see it. The problem with the GUI is that most of the time it gets abused, turning into a "go find it yourself" mentality. A good GUI should

3) The Command-line. At this point, you know what you want, and you just need a simple, fast way to tell the computer that. If this is the case, nothing beats a command line. Can you imagine how insanely fast you could get using microsoft word if you could print at different qualities, load files off the web, etc without ever resorting to any kind of gui?

Really, in order to be truly "easy to use" a program has to allow all of these different modes of input. Furthermore, the wizards have to be bulletproof and co-incide exactly with what the user needs to do. GUI's have to be reasonably helpful, but try to avoid the complexities associated with the command line. Command lines need to have good documentation so the user can start to figure out the commands if they want to.

IMO, there's no way to create a successful interface that suits everyone. If you don't give your varying users all of the interfaces, they're going to just look somewhere else.

How simple

[GunFodder]

Of course! How could we all have been so blind? All we need to do to make software better is to make the most popular features easier to use. I feel like I've wasted years of my life writing crappy UI's without this incredibly valuable nugget of wisdom.

Now all we need is someone who can both see into the future and read our user's minds to tell us which features will be the most popular.

Maybe we can ask the jackass who wrote this article to figure this out for us. Or maybe he's only smart enough to whine about the software after it's already done.

Making stuff easy to do works *for* you

[FortranDragon]

Come on folks, saying that making the common cases easy is a dumb thing to do is like saying the Win32 API is better than a clean API because of the cruft. A chunk of you sound like you measure you manhood by the complexity you can handle. That's being counterproductive and wasting your own time.

If you think that all cases are edge (tough) cases then you haven't done enough analysis or you don't understand who you are targeting your app to. A common case made easy for iPhoto won't be the same common case for a Photoshop or power Gimp-user. Let the computer do the simple shit for you so that you can focus your brain power on the tougher cases.

No, it isn't easy to build simplicity into an app to make the common cases easy. It requires the ruthlessness of someone willing to toss out good code/interfaces that almost, but doesn't quite work. It also requires placing your end-user (of whatever skill level you've targeted) ahead of your own desires for the app. Tough to do, but well worth it in the end.

You just have to ask yourself do you really want to take 27 steps (hypothetically) to configure a printer *every* time? Wouldn't you prefer to just have to do 3 steps 98% of the time and save the brain power for that difficult 2%?

geoworks?

[utexaspunk]

anyone remember the old geoworks ensemble/geos windows clone way back in the day? their office suite had an interesting way of dealing with this problem- have multiple user interface levels. one could choose the beginner, intermediate, or advanced level interface, and it would vary the amount of options available accordingly.

i don't know about you, but I don't want any LESS options, just cuz some newbie can't find his way around. let's make the software easier for dumb people without making the software dumb, ok?

hit a nerve

[GunFodder]

Ahh, you must be a Sowell groupie. That would explain your flame-on attitude.

I suggest before you sound off about writing software you try it - it's harder than you think. The users usually don't know what they want, they just know how to complain about what they don't like. And if you don't already have a user community then it's even harder. You have to guess which features will be the most popular and design your GUI around that.

Maybe you were thinking of ancient software domains like email, word processors, etc. It doesn't take a genius to look at existing software and say "this works, this sucks..." Hindsight is 20/20. It is quite a bit more difficult to make the same observations about software domains that are new or don't even exist yet. Or domains where there aren't as many examples to compare.

Even well-understood domains like word processors are still the subject of user debate. Many people prefer very specific versions of different products. I've used a number of word processors and in the end they all boil down to a common set of features. But those features are more or less accessible in different products, hence the difference in opinion.

So the basic problem is that different people think different things are "simple". I personally think changing my browser proxy settings is very simple, and should be an easily accessible browser control. Apparently the various browser development teams disagree with me, because they always hide that control in a sub-sub-menu. Instead they feature buttons for useless features like "Search" (who needs this with Googlebar?). So I end up downloading an extension for my browser to get this feature.

All Sowell did was play the roll of Captain Obvious: GUI design is hard, and most people screw it up. And it will stay screwed up until software learns to read minds, despite your simple assertations.