First of all, when preparing to "give a paper" at a scientific meeting, I actually write a paper --- not just a bulleted outline of the talk, but the actual narrative that I seek to present. Then I typeset the paper and store it on my personal webpage as a pdf. Prior to the actual presentation at the meeting, I make a short summary of the salient points of my talk and put these into a document that will fit onto a two-sided 3-by-5 card. (I have the cards printed up in some fun color that might hold a connection to the meeting: Bronco orange in Denver, Carolina blue in Chapel Hill, etc.) I also include on the cards my email address and the URL of the paper that I am presenting. The data density of one 3-by-5 card and an 8-point font probably exceeds 3 or 4 ppt slides --- a good deal of information can be put onto those little guys.