First there was the blackboard, and all talks
were given using one. Then there was the
overhead projector. Now the method of
choice for a talk is PowerPoint. Here we discuss
how to give a talk where the assumption
is the speaker is using PowerPoint. However,
if you are not, most of the advice still
applies. Whether this is your first talk or your
100th, the advice is still worth reading.
Preparation
Ok, so you've decided to give a talk. We will
assume you already have, or are close to
having, all the results you need whether it's
through an REU or some other sponsored
research program, or from independent research
overseen by someone at your institu-
tion.
When giving a talk, you need to have a title
and abstract. Some things to keep in mind:
* A title should be short and informative.
Cute titles are usually poor titles. "On a
Theorem of Erdös" is too vague. "On
Erdös's Prime Number Theorem" is short
and informative.
* Your abstract should be one or two paragraphs.
It is not supposed to be a miniversion
of all your results, but a taste to
whet the audience's appetite.
* Include in your abstract the level of the
talk; e.g. "This talk is intended for a general
audience."
Next comes the actual putting together of the
talk. Of course a faculty advisor can help you,
but here are some key ideas.
* It's almost impossible to overestimate
what the audience knows about your
subject.
* Keep technical terms and unfamiliar
symbols to a minimum.
* Don't try to do too much.
* Use simple examples and concrete special
cases.
* Provide a context for your talk. Explain
how you got interested in the subject.
Mention others who have worked on the
subject of your talk.
* Use a very large font (even for a talk in a
small room). If you reproduce printed material,
enlarge it.
* Restrict each slide to a small number of
lines.
* Avoid filling slides with equations and
formulas.
* DO NOT COMPUTE IN PUBLIC.
* If you need a particular slide more than
once in your talk, reproduce it at the appropriate
places rather than going back
to it.
* Many people find unwarranted use of
special features of PowerPoint annoying.
No flying equations.
* Rehearse your talk, but do not memorize
it.
* Anticipate questions you may be asked
and have a response to them prepared.
Giving a Good Presentation
by
Joseph A. Gallian
University of Minnesota, Duluth
ˇ Practice your talk using a projector and
time it. Don't forget to leave time at the
end for questions.
* Conjectures and open problems add in-
terest.
* Prepare transparencies as a backup.
Equipment can fail. Practice with these,
too.
* Be sure you know what equipment/
software is provided. You may have to
bring your own.
Delivery
You've practiced, but you're still a little nervous.
That's normal. Remember, the audience
wants you to do well. Some key points to
help you do your best:
* Show up early to check out the room and
the equipment.
* Erase chalkboards/whiteboards even if
you don't plan to use them.
* It helps to have a laser pointer and use it
to highlight items.
* Keep some room lights on. The audience
wants to see the speaker.
* Begin by letting your audience know that
you are happy for the opportunity to
speak to them.
* Show enthusiasm for the subject.
* Generally stay off to the side, but MOVE,
MOVE, MOVE.
* Don't read your slides to the audience.
Spend most of the time looking at the
audience.
* MAKE EYE CONTACT. Single out a particular
person in the audience who appears
to be interested in what you are
saying.
* Don't distribute handouts at the beginning
or during your talk.
* Go over big ideas twice.
* Go over big ideas twice.
* Speak loudly. Project to the people in the
back.
* Vary your voice for dramatic effect. Pause
before stating a major result, raising a
question, or showing a complicated fig-
ure.
* It is not necessary to prove anything.
* Don't belittle your own results or downplay
your own knowledge of the topic. It
reduces your credibility.
The End
You're almost there. Just a few things to
keep in mind to with your talk.
* Don't forget to leave time for questions.
If you are slotted for 15 minutes, speak
for 13.
* When you are asked a question, move
towards the person who asked it and repeat
or rephrase the question for the au-
dience.
* Have a grand finale. It could be the main
result, a conjecture, an open problem, or
an application.
* Close by thanking your audience.
* DO NOT EXCEED YOUR ALLOTTED TIME.
* Plan to stay a while after your talk. People
may want to talk with you about it.
* After your talk, do a self-evaluation. Make
notes about what went well and what
could be done better next time.