Notes on articles about technical communication


OK, here we go:
First of all let me say that I find it extremely difficult to come up any criteria to evaluate a talk. I have attended various talks where the speakers clearly violated many of the 'don'ts' that I read about in some of the online texts. And those were still highly entertaining talks.
To me the two things that can make all the difference are the personality of the speaker, and of course the content of the talk. On top of that, one always has to take into account the level of difficulty: some talks may be boring to me, even though Bryant and Stern and others may find them extremely interesting. But after all, they know a little more differential geometry than I do.
I read the following articles:

Charles Loan, a short talk:

This one is fairly short and hence also my first choice. The author lists a few valuable do's and don'ts, he discusses the importance of short talks and how easy it is to mess it up. On the other hand, there isn't anything new here: of course a short talk is all about advertising your own research. And the conclusion says it all: in the end it takes practise.

The next one was Kolda:

Well, this one seems to be all about power point presentations. Applied mathematicians may find this one quite appealing but to me it seems less suitable for pure people. But again, apart from the latex advise, there isn't really anything exciting in this text either. Kolda points out many obvious things about talks. For instance, would anybody really forget to bring their computer to a computer talk? And would anyone really stand in front of the screen rather than next to it?

The last text was Mark Hill's oral presentation advise:

He has a fairly useful conference talk outline: and that's about it. Because apart from that there is nothing on the page. Except for the ten commandments at the bottom of the page the summarize how to give a bad talk. And to be honest, some of them sound quite convincing. For instance number two: using less transparencies does save you money: around $7 per year is already a lot for poor graduate students like us. That, together with a car insurance from Geico can save you some valuable cash.


So I ventured off the list on the course webpage just to shake things up a bit.

Website #1: Bob Geroch's Suggestions for Giving Talks

This site really stressed the importance of organization in a talk. It must have been geared toward longer talks, because he suggested breaking the talk up into 3 main points that could stand alone as short talks themselves. I liked the suggestion of keeping the audienced informed and constantly reminding them of where you're going in the talk. One debatable point was that equations should be a last resort for expressing an idea. I liked this site, because he emphasizes making a talk accessible to an audience.

Website #2: How to Give a Good Talk by Joseph Gallian (article from Math Horizons)

This article was geared towards undergraduates, and it had a lot of preparation suggestions that I found useful. As far as suggestions for our critique sheet, I think providing context for your talk and relating your topic to other fields are very important if possible. He also suggested showing enthusiasm for the subject because it energizes the audience. This article was useful because it emphasized how a speaker should feed off of the audience and adapt to the reactions they give.

Website #3: Ten Secrets to Giving a Good Scientific Talk

This website had a lot of the same stuff as the previous two, however it had some good suggestions on fielding questions. The main point was to be personable when taking questions. Repeat the question to buy you time and to clarify. Admit you don't know if you don't know the answer. Don't argue with a questioner. Questions are a good sign that you have stimulated the audience and a good indication of the points you have not communicated clearly.
One suggestion that came up in a couple of the websites was to do a self-evaluation after you have given a talk, which seems like a very useful thing to do to polish your public speaking skills.


After reviewing several articles, I have concluded that giving talks is likely a lot like playing chess. Learning the basic moves is easy but the mastery will come from experience.

Article 1:

Giving a Talk : Guidelines for the Preparation and Presentation of Technical Seminars Author : Frank R. Kschischang
This is a short but useful article. It stresses the idea of focusing your talk around a "take-away message" which are the 3 or 4 most important points you are trying to communicate. It goes on to explain how to structure your talk around this idea.
The article also gives many useful slide tips such as having a few blank sides to go over unanticipated material that comes up during questions. One of my favorite lines in the paper is "Don't fiddle with your pointer, as telescoping it in and out really detracts from what you are saying".

Article 2:

The next article is really long but appears to be a nice reference:
There is so much to this article, I will only give a few notes. It gives a few useful ideas of how to deal with a combined group of experts and nonexperts. It has a detailed section on visual aids including very specific guidelines including how best to incorporate color. The statement "do not read your talk from slides" is perhaps the single best piece of advice. Finally, there is useful information for dealing with a Q&A session.

Article 3:

The next website gives a summary of a book entitled "Dazzle 'em with Style: The Art of Oral Scientific Presentation" by Anholt.
Because it is the summary of a book, this "bullet" list is rather dense. I have seen most of the ideas in this article in other sources, and quite a bit of it is common sense. One interesting idea that it contains is to "phrase the basic idea underlying the talk as a question". According to the auther the most imporant rule for a scientific presentation is "to finish on time and on a clear and resonant note". I tend to agree with this as I have witnessed a many a talk fall apart as the speaker attempted to cram an hour worth of material into the 5-10 extra minutes beyond the stopping time.


1.Mark Hill/ David Patterson

I found this article most useful. Its informal and humorous phrases make it actually fun to read. Also I think giving tips on how to give a bad talk is a very innovative idea to drive home the (opposite) main points. It covers contents and organazation of the materials, rhetorical techniques, and common etiquettes to please the audience.

2.Charles Van Loan

This article is more detailed and serious-minded than the previous one. It is a little dry but is full of useful tips which I would have never thought of. Some of them are subtle (e.g. "Do not fixate on a V.I.P. who happens to be in attendance"), and some of them are just common sense.

3.Tamara G. Kolda

Awfully boring to read. Its advices are often too vague and sometimes questionable (e.g. "Use landscape orientation", "don't put slides in plastic"). On the other hand, its usage of Latex is a marvel to behold. I didn't know Latex could handle presentation this well.

Shu Dai

(In a PDF file)


(In a PDF file)