Companies like AOL and concepts like WebTV get a lot of criticism for "bringing technology to the masses", thus increasing the level of ignorance in online communities and the Internet as a whole, and ultimately causing moronic Usenet posts and helpdesk flooding. While this is true, one must remember that the "newbies" are as much victims as the more experienced users who have to put up with the ignorance of said newcomers.

The real villains are in fact the companies who create so-called user friendly interfaces. The reason, however, why they are villains is not their aim of getting technology into the hands of everyone, but the counterproductive and miserable way that they are trying to achieve this.

There appears to be a belief that if we add talking animations, and ask the users to formulate searches in the way they would ask a person the same thing, we can make computers more similar to people, and thus make the use of personal computers easier and more efficient. While this may seem appealing at first glance, it is neither true nor desirable.

The worst and foremost example of talking animations is (you read my mind) the evil Darth Paperclip of MS Office. The animations steal system resources in all their forms, and the time taken to perform the animation until the requested action (assuming there really is one) is actually performed, not to mention the loss of productivity caused by the distraction, is a huge waste. I would estimate that four out of five users that I would classify as only mildly computer literate have chosen to kill the paperclip or wish to do so. I have yet to meet an experienced user who appreciates the Office Assistant. I am, however, preaching to the choir. The world has already learnt from this mistake.

(I hope.)

Another aspect of the Office Assistant serves well as yet another example of flawed thinking in interface design. Your favorite paperclip will, whenever you cry for help, ask you to formulate a question in plain English, Ask Jeeves style. Let me clarify once and for all for said search engine as well as said paperclip: this is idiocy.

Computers are not designed to interpret natural language, and will not be able to perform any useful interpretation for several years to come. Research is being made, and is making progress, but examples such as those mentioned above will only serve to give this field of research a bad name. Note to all developers: do not include technology that is in its infancy in products intended for home users.

It is much easier for a person to pick out the keywords in a question than it is for a computer. Computers are, instead, very efficient at searching through vast amounts of text for matching strings. Let people do what people are best at, and let computers do what they do best.

There appears to be a belief that if we hide the true nature of the computer in mysterious words relating to everything from astrology to voodoo, new users will feel more at ease using their computer. While this may be true, it certainly does not make them competent users, and as soon as the unexpected happens, they will call upon the same voodoo spirits they were trained to believe in, with little actual result.

It is virtually impossible for a Windows user to install a program without encountering one of the worst misfeatures of today, namely the wizard. The name of this phenomenon is as mysterious as the reasons why it was ever introduced. Do users feel that answering one question at a time, unable to see the next question until they answer the current one and click next is easier than viewing a collection of related options all at once? Imagine doing your taxes the same way!

The word wizard further adds to the mysticism surrounding computers, instead of teaching users about actual processes. It is possible that the phenomenon as such makes certain groups of users more productive, but even so, this does not justify affecting all users. All computer systems should allow experienced users to identify themselves as such, and spare them the humiliation of being treated as though they were clueless.

There appears to be a belief that if we make interfaces colorful and fill them with round or organically shaped buttons, it will not only help to make unique programs that distinguish themselves so clearly that the interfaces themselves in effect become trademarks, but it will also make them appeal more to new users. While this may be true, it certainly is not wise, and new users who are attracted by this hype will inevitably shoot themselves in the foot, and upon realization of this, possibly the head as well.

Three words for you: standardization, standardization, and for those speaking the Queen's English, standardisation. (Now there's a paradox!) A Macintosh or Windows user who has used another program under the same platform will find it relatively easy to relate to another, to navigate the menus and use the general features of the program, and the more experienced the user, the easier this will be. Most users who have been around for more than a week know what to expect under the file menu, the edit menu, and the help menu of a program. They know how to resize a window, regardless of what program they are using.

This is currently the problem with Linux (or more correctly with its programs, mainly X). Users don't know what keys to strike in order to copy selected text in different programs. The same problem can be seen in a few programs in Windows and Mac environments. For example, take Real Audio under Windows. The program icon in the upper left corner of the window is non-standard. A small arrow pointing downwards brings up the menu one would expect from the program icon. Next to this tiny button is a huge program icon, which does not do what its normal-sized cousins would normally do, but instead launches the default Web browser and opens the Real Audio homepage. Shoot me.

Furthermore, the Real Audio interface barely contains a single standard Windows control, but is instead covered with bloated graphics, which gives the program an uncomfortable tendency to crash. What's even worse is that one of the few standard buttons, the close window button, has been altered, so that it does not shut the program down, but merely minimizes it to the systray. This disease has also spread to other programs, such as the terrible MSN Messmaker, pardon me, Messenger. Close means close and nothing else. Halt, cease and desist with this lunacy I say! Usability relies on the very principle of standardization. When you step into an elevator, you expect it to work in roughly the same way all other elevators you have been in. Pressing basement should not take you to the attic.

Bringing technology to everyone is fine, but doing it at the expense of more experienced users is wrong, and will put new users in a situation they will not want to be in. Let the researchers work on what they do, and spare the users from experimental technologies. Until computers are truly intelligent, let me tell my computer what to do, instead of my computer telling me what (it thinks) I want to do. I will know better, at least for another decade. The computer is a tool, a machine, and not an organic creature. Birds flap their wings. Airplanes don't. Perhaps it's better that way. Keep that in mind. is the Web site of the Usability Group, a company that calls itself "Leaders in Customer Experience Consulting". When visiting it with a browser that is not capable of displaying frames, it displays a white page, blank save for the text:

This page uses frames, but your browser doesn't support them.