Part II of the link between physical and logical architecture, at least as they relate to accessibility. Each of the principles listed yesterday can be applied to design on the web, at least indirectly. Let's show some examples of analogous situations.
Principle 1: Equitable use. Easy one here. Exactly the same principle applies in both forms of design, preventing segregation and stigmatization.
Principle 2: Flexibility in use. Ditto.
Principle 3: Simple and intuitive use. An important point that perhaps isn't addressed enough in web design is the last one, "regardless of the user's... current concentration level". We are very aware of prompting and feedback, but the idea that the site the user is on, or even the computer in general, may not be the center of the user's attention, rarely enters into our designs.
This point is particularly important when we consider that a large number of users, particularly power users, surf with multiple windows open, or with the stereo on, or while working in a word processor, making preservation of state and a sense of place even more important than initial impressions might indicate. (Lesson: Stop setting cookies and scripts that time out in a few minutes. We might not want to continue on your site until the next commercial break.)
Principle 4: Perceptible information. The only innovative idea we get here is tactile feedback. While there are currently nascent attempts to introduce that elusive third sense, usually involving a force-feedback mouse, it would appear that we definitely have a need for more interaction between our pointing devices and our content. Why doesn't my mouse slow down over links?
Principle 5: Tolerance for error. A classic usability point. But it bears repeating: "Hazardous elements are eliminated". Instead of making a link that deletes all the user's data and then asking "Are you sure?" 5 times, just take the link out and hide it somewhere where someone could only find it deliberately.
Principle 6: Low physical effort. Another area where the web could make some progress. The key concepts here are minimizing repetitive action, which is largely a result of efficient design, and minimizing sustained physical effort, which is another concept we have neglected. Merely keeping a hand (or eye, or foot, or mouth) in constant contact with a pointing device can present a challenge to many users. Mouse fatigue?
Principle 7: Size and space for approach and use. The best, simplest implementation of this is to bump all your links against an edge of the screen, because the edges are one of the easier parts of the screen to reach with current pointing devices. (Corners are the easiest.)
The other key thing with links is to place sufficient space between them, as we've seen that one of the toughest parts of the toolbar clutter that pervades today's applications is the fact the buttons, icons, and menu choices are all squeezed so close together that it's tough to differentiate and navigate between them.
I'm sure there are a lot more parallels between these two disciplines... more as I encounter them.