Giving a good academic talk

1
Giving a good academic talk
COMP80122

• Carole Goble
• School of Computer Science
• University of Manchester

2
Outline

• General comments on presentations
• Preparation
• Structure
• Slides
• Presentation
• Nerves
• Speaking
• Questions
• Take home message

3
Good Presentations

• Are clear and interesting
• So, simply be clear, interesting...
• Remember
• No rule fits everybody
• we need a lot of practice
• In the rest of COMP80122, we practice

Dont be dull
4
Presentation Preparation

• Logistics
• How much time do you have?
• What technical devices do you have/need?
• What is the take home message?
• New result/technique.
• Consequences of result/technique.
• Read my paper.
• Time
• Practice and make changes.
• Have a friend/stranger listen edit. Esp. an
  audience representative
• A good talk takes MUCH longer to write than you
  think 1 minute1 hour rule.

5
Know your audience

• Who is your audience?
• What assumptions can you make about
• Their background knowledge
• Their listening situation

6
(No Transcript)
7
A presentationneeds a strong framework

• Title slide (title, authors, affiliation, etc.)
• Outline or Take Home Message
• Background
• Motivation, start using running example.
• Related work, continue using running example
• Results
• Summary of results.
• Sketch of techniques used to obtain these
  results.
• Further use of the running example.
• Summary/Outlook/Future
• The Take Home Message.
• Acknowledgements (colleagues and funders)

8
Presentation Structure

• Try not to overwhelm with information.
• 3 point talk
• I came, I saw, I conquered
• 3 part structure
• Beginning, middle and end.
• 3 point repeat
• Tell them, and tell them again, and again.
• Narrative and signposting.

http//www.presentationmagazine.com/rule-of-three-
836.htm
9
What is the story arc?
10
Presentation Communication

• Think visual and high bandwidth info delivery.
• Analogies.
• Minimize words and maximize pictures.
• Two ways of saying the same thing.
• Plan for whiteboard blindness.
• Prepare good slides.
• Expect to throw away two thirds of your slides.
• Use slides to deal with complexity.
• Dont clutter with complicated background images

11
Slide preparation

• Nothing in less than 18 point
• Serif fonts fade out on projectors
• Dont write out great blocks of text from your
  paper. A presentation is not the same a document.
  And we will spend lots of time trying to read it
  and not listen. Are you still listening?
• Use nested bullet points
• To make clear points quickly
• Make sure you spell check
• Not too many fonts
• Dont cram with text or too many bullets
• Be careful of bullet builds
• Dont have bullets at the bottom

12
Colour choices

• Avoid loud, garish colorsdark text on light
  background or light on dark.
• Avoid text colors that fade into background, i.e.
  blue and black
• Avoid color-blind combinations
• Red and green
• Blue and yellow

13
Pictures. Worth. 1000 Words.

• Simple
• Confusagrams
• Animations
• Time to go explain
• Cant? simplify
• Big enough
• Exceptions?
• Making a point

14
Animation fundamentals

• Keep it simple.
• The animation must serve a purpose
• Dont mix up styles.
• One transition style throughout
• Distractions send us crazy.
• No cute cartoons please

15
Handling Technical stuff

• Take your time
• Think of another way of presenting
• Simplify examples for the presentation

16
How does a clockwork clock work?

• It tells the time, mechanically

17
Before the presentation.Get Situated.

• Practice in the room.
• Practice with set up.
• Attend other presentations.
• Get there early
• Dress code for your discipline/audience.
• Smart for Medicine
• Scruffy for Maths
• Wear something that can carry a microphone.

18
Speaking

• Be clear
• Speak clearly, good tempo.
• Speak loudly enough.
• Gestures
• Clarify messages
• Animate talk
• Beware cultural taboos
• Practice with friends.
• Acting.

19
Giving the presentation

• Audience
• Speak to your audience, not your laptop or shoes.
• Check faces of audience whether they can follow
• Include your audience
• Communication
• Dont read out slides or notes.
• Listen to yourself.
• Demeanour
• Dont be cool (bored).
• Be enthused.
• Dont apologize.
• Breathe.
• Mark start and end.

20
On the fly adapting

• Previous speaker has made your introduction.
• Another speaker has presented work close to
  yours.
• Its good to acknowledge

21
Being Nervous. What wont help

• Pretending the audience isnt there.
• Holding (and clicking!) a ball-pen.
• Playing with the change in your pocket.
• Using manuscript or cards.
• Getting drunk.
• Learning everything by heart
• you need to know what you want to say, but dont
  recite.
• Things you have seen?

22
Being Nervous. What can help?

• It is OK to be nervous.
• Be well prepared.
• Pre-prepare a good mental picture of yourself in
  the presentation.
• Dont be late.
• Breath/drink water.
• Pretend the audience is really friendly
  interested.
• Things you have seen?

23
Handling Questions

• Repeat the question.
• Rephrase the question.
• Admit if you dont know the answer.
• Ask for suggestions (review interview).
• Beware getting trapped in a dialogue (research
  presentation).
• The intent of the question.
• Other thoughts?

24

• Content questions
• Could you explain to me how this component of
  your approach works in more detail?
• Philosophical questions
• Do you think there is added value to 3D displays
  over 2D ones in all perceptual tasks?
• Show-off questions
• You mentioned rendering, will you be using Path
  Tracing, Bidirectional Path Tracing, or
  Metropolis light transport, but also semi
  realistic methods, like Whitted Style Ray
  Tracing, or hybrids?
• Trolling questions
• Youve just presented work which I have already
  published in 1983 in Nature and in Science, so I
  dont see how your work is relevant and also my
  method was better and more elegant than yours.
  Can you comment on this?

http//noeskasmit.com/scientific-presentation-tips
-part-2-the-presentation/
25
Take Home Message

• What is your Take Home Message
• to the particular audience?
• Your COMP80122 presentation
• 15 minutes, about your research
• including background, motivation, hypothesis,
  aims, etc.
• in small groups, with lots of feedback

26
Useful Reading

• http//www.cs.wisc.edu/markhill/conference-talk.h
  tml
• http//www.to-done.com/2005/07/how-to-give-a-great
  -presentation/
• http//www.cs.berkeley.edu/jrs/speaking.html
• http//njn.valgrind.org/good-talk.html
• http//research.microsoft.com/en-us/um/people/simo
  npj/papers/giving-a-talk/giving-a-talk.htm
• http//noeskasmit.com/scientific-presentation-tips
  -part-one-preparation/
• http//noeskasmit.com/scientific-presentation-tips
  -part-2-the-presentation/