Information Technology Reference
In-Depth Information
Content
The fundamental challenging of developing a talk is to design a spoken tale. This tale
should inform the audience, and have a structure that keeps their attention and answers
likely questions. Such tales are often based on recent research or on a particular paper,
and can be thought of as an introduction to the work that is the talk's topic.
Many researchers seem to think of the task of developing a talk as consisting
of writing a deck of slides.
1
But this is only one element in the real task: creation
of an engaging, flowing presentation. Slides are a significant component, but there
are others; as well as artefacts such as webpages or videos, for example, there is the
need for structure, narrative, and balance in the delivery. While a slide-authoring tool
might be used to assist in assembly of the talk, it isn't essential in the early stages
of drafting, and the focus should always be on the flow of information, regardless of
how the information is to be communicated.
The first step in preparation of a research talk is deciding what to cover. Such
talks are usually based on a paper or thesis, or on work in preparation, but most
papers have far more detail than can be conveyed in a talk. Problems of this kind are
highlighted in the experiences of research students. The initial reaction of a typical
student preparing a talk is concern that there isn't enough to say, but the initial
reaction of an advisor is, often, that the student's draft talk is far too long. Thus the
content must be selected carefully.
What and how much to select depends not only on the time available but also on
the expertise of the audience. A workshop attended by specialists in a narrow topic
would suggest a different talk to one to be given to researchers in your department.
Papers are usually specialized, but a diverse audience may be unfamiliar with even
the area of your research, so it may be necessary to introduce fundamentals before
proceeding to the results. For any audience, there is no need for a talk to just be an
overview of the paper; it is an introduction to the ideas and research results described
in the paper, and many paths can lead to that same outcome of teaching the audience
about your work.
When constructing a talk, begin by choosing the single main goal, that is, the
particular idea or result the audience should learn. Then work out what informa-
tion is required before the result can be understood. Often this information is in
effect a tree whose branches are chains of concepts leading to the result at the root.
Much of the hard work of assembling the talk is pruning the tree, both to suit the
audience and to strip the talk down to essential points that listeners should remember.
An approach to gathering material for a talk is “uncritical brainstorming, critical
selection” (which should also be used for writing). In the first phase, jot down every
idea or point that might be of value to the audience, that is, list the topics you might
conceivably have to cover. Imagine yourself chatting with someone about your work,
and note down the things you might say. During this first phase it is helpful to not
1
Or set of slides. Or pack. Or, sometimes, of a set of overheads. To me,
deck
suggests the action of
displaying the material in turn, following some sort of order or structure, and that is the terminology
I use here.
