Tuesday, August 31, 2004

Talk styles

So I just finished my talk, and it appears to have been well received. Not being a DB regular, I took a somewhat more general tone with my slides, and listening to the other talks in my session, I realized that some notions I could have assumed audience knowledge for.

It brings up a duality in talk styles: does one:

* go into lots of details, explaining how specific things were done.
* present a general overview, glossing over details ?

As with all dualities, the answer is "it depends". On the one hand, if you give a talk where you say at some point, "And now we shall solve subproblem B with this trick that I won't get into in this paper", the people who came to your talk precisely because they had been banging their heads over subproblem B to no avail will be rather disappointed.

On the other hand, a lively and spirited discussion among 3 people about the solution to subproblem B might be lost on the 100 or so other people who have no clue as to why it was such a big deal.

My personal preference is what I'll call 'the big tent'. Keep things at a fairly high level, assuming that someone interested in the problem can read the paper with the roadmap one provides.

But (and here is where good judgement and the art of presentation comes in), make good choices about what details to present and what not to present. Some details are illuminating; some are not. Knowing which is which differentiates a good speaker from a not-so-good one.

No comments:

Post a Comment

Disqus for The Geomblog