ryan.freebern.org

Since the advent of the commoditized personal computer in the mid-1970s (with the release of the MITS Altair 8800), things haven’t really changed much. Sure, computers have become significantly smaller and faster, and the number of things we can do with them has ballooned to the point where many of us can’t, or don’t want to, imagine life without them. But when you look at the nitty-gritty details of computer usability design, both the hardware and the software, it’s still largely based on assumptions made by a bunch of scientists who were flying by the seats of their pants decades ago, just making things up as they went along. Things don’t change, because we, as a global society, are stuck in familiar ruts. I contend that these ruts are not just suboptimal, they’re actively deleterious to our culture.

Here we are, brains the size of planets, and yet we’re using flaky computers that sort of do some of what we want most of the time. They require constant attention to keep them running smoothly, as if they’re made of sticks and twine rather than solid mechanical parts. We each spend countless hours fixing problems, upgrading, configuring, maintaining our systems instead of just using them. Throughout the entire computer-using world, that’s an unimaginable amount of potential productivity, education, enlightenment, or entertainment time lost to frustrating and annoying endeavors. And it’s even worse if you’re using a Microsoft OS!

How can you tell we’re stuck in ruts? Just look at the vestigial anachronisms that abound in the design of computers. When Microsoft releases a new version of Windows and the most apparent difference is that the windows are shiny and partially transparent, you know we’re not even attempting to seriously innovate beyond the WIMP GUI developed by Xerox in 1973. This sort of “upgrade” is like putting lipstick on a cow. Sure, the end product is superficially better, but the starting point wasn’t all that great to begin with.

Scientists and engineers have done hundreds of studies and spent millions of hours working on improving computer usability, but very little has actually been done. If we’re going to truly move into the next age of computer development, we have to be willing to take a giant leap in how we design our computers, including operating systems and software, from the ground up. This is what I want to explore.

Why does the PC need to be reimagined?

While computers have addressed many of our daily needs and have simplified many tasks, I believe they do so despite their design, not because of it. People are constantly fighting their computers, and that detracts from the potential of a computer as a tool to enhance life. Computers fall short in three major areas: stability, usability, and simplicity.

Stability: Computers are notoriously temperamental. The instant a single part fails, the rest of the system becomes frustrating and unpredictable at best, unusable at worst. Software usability only matters if the underlying platform can be trusted to be stable. My proposed redesign should correct this behavior, so that when something fails, the system smoothly and efficiently corrects for it, and gets it taken care of quickly.

Usability: Computers work in a way orthogonal to human brains–they’re excellent at the precise, mundane tasks that humans find difficult, like arithmetic and remembering data. Humans, on the other hand, are good at finding general patterns, making guesses and predictions based on observations, and generally being imprecise. Unfortunately, many of the ways computers are designed don’t take this fully into account, and still require humans to worry about precision, whether it’s accurately remembering a long file path or URL or aiming a pointer at a 16-by-16 pixel square to close a window. My proposed changes should allow humans and computers to each focus on what they excel at.

Simplicity: Computers are designed to be a tool, and the main purpose of any good tool is to simplify a task. Unfortunately, due to the way computers are created, the task is constantly redefined by competing groups with their own financial interests in mind. For instance, my laptop has 11 different types of ports in it, each designed to take a different plug. Computers (and this includes peripherals) need to simplify, so that when I want to plug something in, I don’t have to know the acronym used to describe the plug type and hunt through the myriad holes in the sides of my computer to find the right one.

What are my goals here?

I don’t think for a moment that my ideas and recommendations written here are going to, by themselves, jumpstart a revolution in computer usability design. I do hope, however, that by laying out my thoughts on all of these issues as clearly as possible and exploring the various problems and solutions, that I can start a discussion that might, at some point, lead to people making changes for the better. My goal is simply to highlight the problems that we all face (perhaps even unknowingly) but consider “just how computers work,” make it clear why these problems shouldn’t exist and can be fixed, and provide outlines for potential solutions.

I’m not an expert in computer hardware design by any means. My degree is in computer science, not engineering, and what I know I only know because I’ve spent two decades using, building, maintaining, and reading and thinking about computers. I’m open to hearing suggestions and corrections here as I write more on this topic; if you have any particularly good insights, I’ll gladly consider them and try to work them into my thoughts, so please, feel free to speak your mind. And just to make things clear: by “usability design,” I’m talking about the ways the design of the computer affects the way we use it.

In my next post, I’ll vivisect the beast and take at look at its innards, to see what could be designed a little more intelligently.

This entry was posted on Monday, Apr 16, 2007 at 6:52 pm and is tagged design, technology, usability. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.