Making it Hard is Sometimes Better

Within the usability community we constantly hear the mantra "make it easy", but lately more and more products are made too easy for their own good. Take something like Apple's Remote. It is incredible easy to use, but at the cost of essential functionality. Just a few more buttons, and a few tweaks to the interaction would make it much better.

Easy of use must never come before making something that works - even when that means making it harder to do.

The nail

To illustrate this let's take a look at something non-digital - the nail.

The nail was invented a long time ago and from the start it was a very simply "device" to use. It requires little effort, it is very cheap to make, and its purpose is straight forward. But, it doesn't work very well.

The problem with nails is that they have little grip. If you hammer two pieces of wood together it is likely that they will come apart over time. It was made too easy and too simple.

Everything should be made as simple as possible, but not simpler.
- Albert Einstein

The Screw

To solve the problem a smart person invented the screw. This was a revolution as it solved the problem if lack of grip. But, from a usability point of view it was a disaster. In terms productivity it was even worse, and it costs a lot more money to make.

But, They had to make it right and thus the screw had to be invented. It was not choice - it was matter of getting the job done.

You need to do the same

When you are making websites, web applications or even regular applications you need to get it right too. Solve the goal, and then you can work on making it better, easier and fancier.

Do not mistake this for just making it technical or otherwise complex - those applications does not get the job done either.

Again, look at the screw. Compared to the nail it isn't easy to use, but it is still a very simple tool. And, over time it has been improved considerably. The first type of screws was the slotted screw (above) which was somewhat difficult to use.

Then came the Phillips drive design which solved many of the problems with the slotted screw, but unless you had the 100% correct driver for it you can easily could damage it. Thus someone invented the Prozidriv and lately we have seen the Torx drive design - which is a revolution in itself.

The point is; get the job done, that is always your number one priority. Your second concern is everything else - technical finesse, graphical fanzines, or ease of use.

Comments

Kevin Cannon - Nov. 8, 2006

I agree with your overall statement, making something easy is not always the goal. Making it effective is important, and sometime that means being difficult.

However, I think saying a screw is less usable than a nail is incorrect. Both are trivial to learn to use and if anything, nails often require more technique from the user.

Similarly, the Apple remote is great. It's simplicity means I could hand it to anyone. I'd be curious to hear what additions you'd like to make to it.

Jonathan - Nov. 10, 2006

Ah, good to see you've been reading Raskin.

This usability vs intuitiveness (or accessibility) thing has been rattling around for about 30 years now. Raskin et al went back to first principles to define usability, but things have gone in a different direction since then. For the nadir of this approach, get ready for Windows Vista and its ability to hide functions away from you and ask you if you want to do this or that all the time until you scream with pain.

I would choose an example closer to home to illustrate this issue: the command line. While the UNIX CLUI is at first completely unintuitive, it's beats the hell out of a WIMP interface for getting 90% of the jobs done on a server. It takes time to learn, but once you have that knowledge, there *is* no alt=ernative.

Thomas Baekdal - Nov. 12, 2006

Kevin, Thanks.

The interaction required to use a nail is:

1: Point it at the material

2: Give it a good wack with a hammer

With screws it is:

(1: Prepare the material if needed - in many cases you need to drill small holes before you can start)

2: Point it at the material

3: Rotate the driver a gazilion times while applying pressure (this is both hard and can take a long time compare to simply wacking nails).

As far as the Apple remote, then it looks great until you start using it with more than just a few movies/songs etc.

Jonathan, Thanks.

I must admit that the nail/screw metaphor is much closer to home for me than the UNIX CLUI. But I understand what you mean.

John Rhodes - Nov. 13, 2006

"Then came the Phillips drive design which solved many of the problems with the slotted screw"

What are the problems with the slotted screw? Slippage of the screwdriver? Just curious...

Thomas Baekdal - Nov. 13, 2006

Yes, you can very easially slip out of it - especially with power tools. And when you do there is a huge risk that you will damage the material underneath (because you are applying pressure to the driver).

The Philips design was specifically made to make it easy to use power tools, as it is designed to "cam out" to prevent tightening it too firm. But, the design also meant that the sides of the drive design was vulnerable to damages from the driver (especially if you use the wrong one).

The Prozidriv solved some of this. It is still designed to prevent over tightening, but with its 8 contact points it is less vulnerable than the Philips design - and is easier to use because it is less likely to slip out if used incorrectly

The Torx design is still designed to prevent over tightening, but less so than the two others. It was designed for durability and easy use. You need much less manual pressure than any other design.