|
Tips for giving a good talk
While not meant to be comprehensive, the following is a list of a few tips
which can help to improve the quality of a mathematical or other formal talk.
- Prepare a script. Even if you know your subject like the
proverbial
back of your hand, it can be very easy to get nervous, intimidated, or just
confused during your talk. After all, you're under time pressure, and the
audience can include world experts in your field who very likely know much
more about your subject than you do, which can be intimidating. Having a
script ready means you can pick up where you left off if you get mixed up, and
nothing says you have to follow it precisely -- but it's important to have it
in case you need it.
- Make slides. Writing on the board is very time-consuming and
prone to errors, especially when you're under time pressure and nervous.
Preparing slides beforehand not only gives you the chance to edit what you
show the audience in advance, making sure you catch and correct errors, but
saves you the time of having to copy onto the board. Also, it's generally
easier to read typed slides than handwriting. Preparing slides forces
you to organize beforehand what you plan to say, so that your talk has a
better chance of sounding like a coherent talk instead of disjointed
rambling.
I used to use overhead slides (aka transparencies) prepared with LaTeX,
though these days my preferred format is a landscape .pdf file on a USB
stick, prepared with LaTeX -- it's pretty unusual these days to be asked to
give a talk in a room which doesn't have a computer and projector. One can
opt for an old-school look with SliTeX, or for a more modern look use the
Beamer package.
For the LaTeX-impared, another option is to use PowerPoint (or
OpenOffice), though if you opt for
this route I urge you to resist the temptation to fill your presentation with
distracting fades, special effects, and little dancing monkey animations.
Whatever softare you use to prepare your slides, it's a good idea to make
a self-contained .pdf file in case the computer you're presenting on
doesn't have your fonts or software installed. It's also a good idea
to have a printout of your slides for use with camera-type projectors in case
of computer crashes.
- Slides are not your script. Your slides should be the
illustrations of your talk, not the complete talk itself. Too many words
on a slide are distracting and the audience ends up reading your slides
instead of listening to your talk. You should have a title slide listing your
name, the title of your talk, and your academic affiliation as well as any
other relevant information like faculty mentors, coauthors or funding
information. The rest of your slides should have the main points of your
talk, including short statements, diagrams, equations, statements of
theorems and sketches of their proofs, and computational results.
- Practice your talk. Going through your talk in front of actual
live people can help you catch errors or omissions and give you a good idea
of how much time your talk will take. It can also help you to anticipate
questions that your audience will have, including the ubiquitous "What are
the practical applications of your research?" question.
- Appearance. It's not always necessary to dress up formally in a
suit and tie, but it is important to calculate the effect your appearance
will have on your audience. When in doubt, it's best to err on the side of
formality, and in any case one should always try to look good. Dressing up,
even if only slightly, sends the message that you're aware that you're not
in everyday mode but are taking the conference seriously.
- Attend the other talks in your session. If everyone skipped all
the talks but their own, there'd be no point to having a conference. Part
of what's good about going to conferences is getting to see what other people
are doing and how they're approaching their research problems in order to get
new ideas for your own research. This also solves the problem of getting
lost in the building (unless you're scheduled to talk first) and makes
it easier to deal with unanticipated schedule or room changes.
Copyright © 2005-2008 Sam Nelson
| |