Thursday, January 10, 2008

The psychology of user interface (part III no freedom without boundaries)

There is no freedom without boundaries.

I read this from Bruce Tognazzini many years ago. It has always stuck in my head. Our politics is obsessed with the idea that freedom is good. The business community is obsessed with the idea that people need more choices in order to get exactly the thing they want. We have infinite sizes, flavors, colors, textures in almost everything we buy.

Freedom schmeedom

Too much choice is bad. In life, in politics, in products, in relationships, and indeed in user interfaces.

People need guidance. People need to be managed... told what to do. If you just hired a bunch of people and said lets make the best PIM in the world, and gave them no guidance, it would suck and fail. In the same way, when you provide a user interface, there is a balance between giving them enough options for them to feel comfortable, and not so many that they have no idea what to do next.

In the old days, this issue manifested itself in a different way than it does today. But this is not because we are smarter today, but because things generally suck so badly that we have a more limited set of ways to overwhelm the user. But somehow, designers still seem to figure out how to do it.

Photoshop. It sucks.

In prehistoric times, apps would have zillions of tools and no mechanism for understanding the context for using those tools. Photoshop is a good example of one of these prehistoric apps, because, from a UI perspective, Photoshop sucks. I know I know you probably love it. But that is just a manifestation of the Stockholm Syndrome. Trust me, it sucks.

In Photoshop you have a massive array of tools. There are no noobie level affordances. No tools jump out and tell you what you should be doing with them. You are just expected to memorize how stuff works. And damn it, you're gonna like it.

It is the epitome of blank slate design. You must stare at an empty page with a collection of nondescript tools and figure out how use it.

Of course Adobe is in a bit of a jail on this one because the high end users who spend a lot of money on Photoshop at this point know it like the back of their hand. And they *like* that is it hard. Job security. But this means Adobe is vulnerable to someone coming along and making a truly usable image manipulation program with perhaps fewer features.

So now what

Ok, so I know, you say, "my stuff is nothing like Photoshop".

No shit sherlock. If it was you probably would be on a beach somewhere instead of reading my blog. But there is still an important lesson here. I am dumbfounded by the number of websites and web applications that refuse to tell me what to do. If you have a website, I want you to tell me what is important. If you have an application, I want you to tell me what my first steps should be. I NEED HELP.

There is nothing worse than going to a website and spending five minutes and having no idea what the product they are selling is or how to find out. There is nothing worse than going to a web app and finding a sea of command options of seemingly equivalent importance. You see, if you don't tell me what to do or where to go, or what my first steps should be, I am lost to you. And I will just leave your page going, wow, that *really* sucked.

5 comments:

Anonymous said...

Hi,

Part II lacks a link to this part, for those reading from part I ;)

Juan said...

I certainly agree with your point of view, but while I was reading the post series something that didn't quite fit was bothering me. There is another dimension from where you can look at the same problem, efficency. Efficency is a very practical concept in my opinion that leads to linear thinking most of the time, but picture this, take a good ol' mainframe user (not programmer) and give him the best UI you can ever imagine. I think he will stick his terminal and continue mumbling weird mainframe command, and I also think that this is not (entirely) due to his stubborness, it has to do with a clear difference between the entry level and the pro level. When you are a pro you just want a commandline that can keep up with your amazingly fast typing of the most bizzare combination of commands, while when you are a newbie you need the application's guidance and not a 2000 pages manual full of commands and constraints. What a large comment, this will probably have to end up in a blog. But in the end, how do you think the difference between entry level and the pro level fits in your idea? There is an interesting article by Tuomo Valkonen somewhere at http://modeemi.fi/~tuomov/ion/. Cheers, and you are in my favourites list too!

Hank Williams said...

Juan,

First of all thank you for making me one of your favorites. That is awesome!

Regarding the point about experts versus noobs, my view is this. All things relating to the human experience are on a bell curve. The goal in any design is to hit the meaty part of the bell curve for your design. In other words using the 80/20 rule, you want the 80 not the 20. Unless your market *is* the 20. There is nothing wrong with this because narrowly targeting markets should allow you to charge more or more deeply penetrate a segment. But for mainstream products, designing something that works for *most* people should generally be the goal, and my theories are tied to that goal.

Anonymous said...

AMEN! Photoshop does suck. It's great if you have hundreds of hours of experience with it and can do it in your sleep. If not, Photoshop says "F.U, Pal. I ain't telling you how to do this, and you won't be able to figure it out on your own, either." The UI is horrible, the instructions are nonexistent, and the help file is a circular, self-reflexive piece of crap.

Anonymous said...

NEW YORK SUCKS BADLY

Post a Comment