Understanding open source Part 2

Forces for usability engineering

An increasing number of corporations are building for-profit businesses around open source. Some are selling it and providing easier installation, printed manuals and extras that make it worthwhile to buy instead of downloading. Others are providing paid tech support, an even more compelling option given how many commercial software vendors are charging for support in addition to the license fee. Hardware vendors are shipping computers with Linux and other OSS installed. And some companies, like Apple, Sun and Novell, are building commercial software on top of open source projects, often providing development effort, administration or funding to those projects.

In each case, a commercial venture has a stake in an open source project. The revenue of those corporations is dependent on the product's quality and success. So, they have incentives to make the same sort of investment in usability engineering that they would for a closed source product. They bring usability engineering increasingly into play, but OSS brings with it another powerful force for usability.

The arc of usability innovation

Usability innovation is where usability engineering sees its greatest benefit: the creation of new technologies and the development of new and better user experiences. Yet most software user interfaces today are not innovative, based instead on existing standards.

The life cycle of a software technology can be viewed in terms of usability innovation. The standard graphical user interface (GUI) consisting of windows, pulldown menus and a mouse pointer is a good example. It was originally developed at Xerox PARC. Apple copied it, streamlined it and popularized it. Microsoft copied Apple's implementation and popularized it further. Somewhere between Apple and Microsoft, most of the core functionality became a de facto standard, and more recent operating systems like Windows XP, Mac OS X and popular Linux GUIs have faithfully copied it. Someone creating a new operating system today would likely do the same, not because it's the best way but because of its familiarity. Piggybacking on previous research and comfortable paradigms saves time and money. Xerox and Apple were innovators that had to worry far more about the usability of the GUI than anyone does today, as long as they stick with the same basic principles.

That's not to say there isn't room for improvement. And certainly today's GUIs are all a little different. But the potential usability cost of those differences is insignificant compared to the potential usability cost of changing core aspects of the GUI (for example, replacing windows with some other scheme entirely). GUI innovators are still trying to create something better. But in the meantime, it's easy and effective for software developers to follow convention and focus on something else.

This arc of usability innovation--from critical at the beginning of a technology's life cycle to less important as it reaches maturity--can be seen in most established software technologies.

Status quo as platform

As the arc of innovation nears its end, a technology becomes a platform. Developers wishing to enter that space can copy the platform (i.e. the standard user experience) and then build innovations on top of it.

That stage is a natural fit for OSS:

• Building on the user experience of a closed source product involves rewriting it from scratch. For example, if I want to improve on Windows XP, I can't just modify Windows itself--I don't have the source code. So I have to build my own Windows clone first, an extremely time-consuming process that may result in a lawsuit. I can, however, find an existing open source product and get right to work on improving it. I needn't reinvent the wheel.
• Because the de facto standard user experience is well-known, well-tested (one hopes) and familiar to end users, developers can bypass some of the usability engineering, taking advantage of what's gone before.

It's no coincidence, then, that the OSS products most ready for end user consumption--such as the GNOME and KDE Linux GUIs, OpenOffice.org and Firefox--provide user experiences similar or nearly identical to their commercial counterparts (Microsoft Windows, Microsoft Office and Internet Explorer or Netscape for the examples above). Those projects aren't being unoriginal, they're taking advantage of the arc of usability innovation to provide usable products without the cost.

Darwinian usability

Will usability innovation always be confined to commercial, closed source software? Not necessarily. Usability engineering may not always work for OSS, but it's not the only way to usability. I believe a process of usability evolution may accomplish the same end.

It starts at the end of the arc, where a technology's user experience has become a platform, successfully implemented and distributed as open source. OSS developers have the opportunity to tweak and improve upon aspects of the technology, adding or modifying it feature by feature if they wish and distributing a variety of modifications. Few will undergo usability testing or involve usability professionals, and some will be more usable than others. Think of those little changes as the genes of usability evolution.

Instead of one official version, larger OSS products often exist as several distributions. Linux is an excellent example: Popular distributions such as Red Hat, Fedora, Debian, Slackware, Ubuntu and SUSE target slightly different audiences and come with different options and add-ons. Distributions are handy, particularly compound projects like Linux, because they give users several unified options without requiring them to do the research and compilation of individual components. Think of distributions as the organisms of usability evolution.