I think that today is a very good day to sit back and think about how average people will come to understand the semantic web.
When you were in school, if you were very lucky, you had a math teacher who understood that the best way to motivate students was to humilate them. I remember one that I had who challenged the students on the first day of class by saying, "You don't know the difference between 'and' and 'or'!" When the students, most of whom had been using those words successfully for well over a decade, took up the challenge, she pointed out that if you say, "I packed peanut butter AND chocolate in my lunch today" that you must be referring to a delightful confection that includes both chocolate and peanut butter. If you want to say that you packed, say, both a peanut butter sandwich as well as a chocolate bar, then you should say "I packed peanut butter OR chocolate." The brief state of confusion brought on by this revelation never failed to bring the students back for more, hungering for the orderly and simple truths of mathematical logic. Humiliation is a great teaching tool.
Teaching the Semantic Web is no different. The only way for the semantic web to succeed is if everyone first gains a strong understanding of the mathematical logic behind it. This will have a positive effect on business practice on the whole, since project managers, product managers, marketers, evangelists and change agents have such a strong need for mathematical logic in their daily work. Imagine how much better a human resources department will work, once it realizes that it can treat people like numbers!
Fortunately, the semantic web provides us ample opportunity to communicate the subtleties of mathematical logic. Take the distinction between OWL-DL and OWL-Full. The importance of this distinction rests on the notion of decidability; This topic was unknown to mathematics until the early 20th century; that means that when we teach this to our semantic web students, we are giving them insights into some of the most advanced notions in modern mathematics. Euler, Newton, Leibniz, Euclid and Pythagaurus all managed their mathematics without this notion. We can't let that happen to our students.
We can capitalize on the popularity of high-school philosophy class by using the words that everyone learned there. Since everyone remembers what "necessary" and "sufficient" mean to a philosopher, we should make sure to use these in all or our modeling definitions. You know, "Having two legs is a necessary condition for being a man" and "Arestotal has two legs" so "Arestotal is necessarily a man". Or maybe its the other way around. Either way, it is a great way to get ordinary people involved in the semantic web. I mean, for the regular web, people were willing to type <a href=...>. Necessary and Sufficient are easy after that.
I am looking forward to the day when the semantic web finally brings mathematical sophistication to the masses. After all, the obvious advantages of the semantic web will make it worthwhile for everyone to learn logic!