Preparing transparencies

I have been writing up my transparencies for a talk on my work I will give next week at the Eleventh Marcel Grossmann Meeting. You can read the abstract here; you will see it is the same material covered in my recent paper, plus an extension to de Sitter space I have been working in.

I always use hand-written transparencies instead of printed transparencies or PowerPoint or similar presentation programs. The reason is simple laziness -writting transparencies is much faster, Unfortunately, my handwriting is not only atrocious but also very small. Added to this, my natural tendency is to write complete, grammatical sentences instead of just bulletpoints and the like. So my transparencies end up being full of ugly, densely-packed lines of text and equations. My speech sticks closely to what is written so it is not that difficult to follow -but after seeing my last talk my supervisor suggested that I should try to make less-cluttered transparencies.

The problem is I can't. I may try to write larger -but unconsciously I always revert fast to tiny letters. (Amateur psychologists please refrain from commenting). And a transparency with just three or four lines of text and math in very small handwritting looks ridiculous, so the temptation to go on writing and fill it is large.

So what I did at the end is to use superposed transparencies. For those transparencies that could be "broken" in a natural way I wrote the upper half of one transparency, and the lower half of the next one. In the talk I will show first the first transparency, talk about it, and then place the second one on top of it. Hopefully, the audience will have time for reading each part without being intimidated by a transparency with so many lines of text.

I borrowed the trick from my supervisor, who uses it mostly for the kind of flashy effects you can use PowerPoint to do; for example once he had a transparency with the classical Poisson brackets of a theory, said "And now to quantization", and put on top of it a transparency with "hats" ands "i hbar" in the right places to create quantum conmutation relations. But it's not something I've seen people do often; I guess those who like that kind of effect go directly for a presentation software.

I am interested in hearing, has anyone met the same problems I have? More generally, what are your difficulties and strategies when preparing talks?