Quite often, when I give a talk, someone records it, often as video, and webs it. That raises a question relevant to what talks I give: Is watching the video a reasonably close substitute for attending the talk?
If it is, then once a talk has been webbed there isn't much reason to give that talk again. Anyone interested can watch the video, and two recordings are no better than one. So I should make more of a point than I do of expanding the range of talks I give. I currently have, at a guess, five or ten that I have given multiple times. But with a little effort, there could be more.
This links to a question that has puzzled me for a long time. One common pattern in schooling is the mass lecture—a professor speaking to an audience in the hundreds with students taking notes. In the fourteenth century, that made a lot of sense as a low cost way of spreading knowledge, but why did it survive the invention of the printing press?
The author of a book can do a much more careful job of presenting information than a lecturer can. A student is lucky to attend a class by the best lecturer at his school—he can choose to read the best book on the subject that has ever been written. Lectures must be attended at a fixed time, books can be read on the reader's schedule. A lecture goes at the same speed for everyone in the audience—when reading a book, you can go quickly over the obvious parts, slowly over the parts you find difficult. A small class permits a substantial amount of interaction between teacher and students, but with a mass lecture that is reduced to at most a few questions followed by responses and the author of a book can include in it responses to the usual questions.
Nonetheless, mass lectures continue to be given. Which suggests to me that there is something I do not understand very well about the realspace interaction between speaker and listener, some reason why, for many people, the lecture works better than the book.
And, perhaps, the live lecture better than the video.