While I don't disagree with the general points given here, I feel this is how to give a great talk, not necessarily a great scientific talk.
There's an interesting point buried here (emphasis mine):
> Presenters often fail because they try to deliver too much complex information. Language and content, normally, has to be designed with the non-specialist SCIENTIST in mind. “You have to think about the least knowledgeable person in your audience THAT YOU CARE ABOUT REACHING” says Rubenson.
If you are presenting a breakthrough in compilers in a compilers workshop, maybe you don't want to start your talk explaining what a compiler is. Step 0 of every presentation should be "know your audience", and if your audience is expecting complex information you will have to get into details at some point.
That doesn't mean that elitism is a good thing. I think both beginners and professionals should walk our of your talk understanding something new. And a good scientific presentation should be also a good presentation. But with limited time (major conferences implemented 20 minutes talks this year), you need to set your baseline knowledge somewhere.
>I feel this is how to give a great talk, not necessarily a great scientific talk.
I come from a Biology background so maybe its different, but I have been to so many talks where 90% of the audience is grad students and looking around the room you can tell none of them are getting anything out of the talk because they got lost in the first five minutes. Sure it would be silly for every Neuroscience talk to waste five minutes explaining what a brain is, but even seasoned scientists seem to take for granted that not everyone studies their same corner of the field. Giving sufficient background info is severely underrated in talks, as is very explicitly stating ideas that will be necessary to appreciate the core thesis of the talk is. I'm sympathetic to the fact that timing is tight and scientists want to save a significant amount of time for their actual research, but if most of the room comes away from the talk with absolutely nothing then it was a wasted effort. You're right that the presenter should know their audience and that would fix things, its just that in my experience, the presenter is almost always overestimating their audience.
From (the mathematician) Gian-Carlo Rota's Ten Lessons I Wish I Had Been Taught:
"...I remembered an MIT colloquium that took place in the late fifties; it was one of the first I attended at MIT. The speaker was Eugenio Calabi. Sitting in the front row of the audience were Norbert Wiener, asleep as usual until the time came to applaud, and Dirk Struik who had been one of Calabi's teachers when Calabi was an undergraduate at MIT in the forties. The subject of the lecture was beyond my competence. After the first five minutes I was completely lost. At the end of the lecture, an arcane dialogue took place between the speaker and some members of the audience, Ambrose and Singer if I remember correctly. There followed a period of tense silence. Professor Struik broke the ice. He raised his hand and said: "Give us something to take home!" Calabi obliged, and in the next five minutes he explained in beautiful simple terms the gist of his lecture. Everybody filed out with a feeling of satisfaction.
Dirk Struik was right: a speaker should try to give his audience something they can take home. But what? ..."
This is because (deliberately or not) what the speaker often wants to communicate mainly is that they are smart. Getting everyone lost in the first five minutes is not a big problem, quite the opposite. I have seen young people who gave brilliant lucid and comprehensible talks who were criticised by the experts in their narrow sub-field as "popularisers". I think many people know how to give a great talk, but choose not to.
I can see your point, but, at the same time, most talks, like most research in general, are incremental improvements or variations of X. You could spend first 5 minutes of the talk giving introduction to X, but if the audience does not know about it already, why would they be interested in your minor contribution?
Say, I'm computer vision researcher specializing in (non-deep learning) 3D reconstruction. Why would I be listening to talks where someone goes into some esoteric tweaks they made to the deep learning stuff? I feel like there almost should be separate conferences for people doing feature recognition/tracking, 3d reconstruction, object recognition etc., as all those fields are vast in themselves. And, once you're on such conference, it's just the experts there and you don't need the 5 minute intros any more.
>most talks, like most research in general, are incremental improvements or variations of X
Again, I don't know if this is something that is only true in Biological science and less so in Computer Science, but while I agree most research is a minor incremental improvement, its often on a topic so specialized that most outside of your research group aren't very familiar with the intricacies.
This begs the question, like you said, then why would you care about hearing that talk? Maybe you don't even care about their thesis and are just interested in adapting a technique they used in your own work. Or maybe their thesis bridges a gap between two fields of study that were previously not very closely linked. Besides, when you're trying to communicate your work to other people, isn't it to your advantage to try and cast as wide of a net as possible? Isn't part of the point of talks and conferences to widen people's spheres a bit? There's no need to leave your home lab if you want to converse with people who understand most of what you're researching already.
>... I have been to so many talks where 90% of the audience is grad students and looking around the room you can tell none of them are getting anything out of the talk ....
Sometimes the students aren't the targets. They're just guests. The speaker is targeting the handful of senior people in the audience.
I guess, as a defence mechanism, people embbed complex (almost impossible to understand during the talk) information in their slides, avoiding hard questioning afterwards.
It was not once or twice that I have witnessed that during conferences and workshops. People are afraid to be ridiculized in public, or seem incompetent in front of a large audience. We won't be seeing that to change in the short run, unfortunately.
> ...as a defence mechanism, people embbed complex (almost impossible to understand during the talk) information in their slides...
Yes, this happens a lot, and it has the effect of alienating those who actually need to hear the talk and coming off as pandering or cloying to those that already know the material.
One thing I suggest to folks, if there is some insecurity about not covering "everything", is to make some slides on detailed sub-topics that aren't included in the main talk but are intended to be used when/if there's a follow-up question.
If the speaker is careful, they can even "bait" their talk to implicitly suggest follow-up questions for which a detailed auxiliary slide has been prepared.
Perhaps. Presenting overly complex & technical information happens a lot, but I think there are other explanations. Wanting to impress one's peers, for example. After all, the last 10% technical detail is often where one spent 90% of the time, and people have a tendency to want to tell that part of the story, whether or not it matters to the audience (and it may or may not, depending on the context).
Or they have a single slide deck for many occasions and sometimes overestimate the audience. I've seen this a lot (as an overestimated student) in my first year of grad school -- these were all budding researchers, right?
Generally in life I tend to overestimate people's technical level -- maybe not of computers, which change a lot, but basic statistics and finance, general knowledge, etc. This is why: "brains" only get you so far in life, but being an intellectual-type person one wants to believe that those more succesful than you are even brainier than you.
Some people work out their arms so much their legs are comparatively small. These people are not about to become star athletes. I wonder if they look at olympic sprinters or soccer players and think "what great arms this guys must have, I wouldn't want to wrestle with them".
> If you are presenting a breakthrough in compilers in a compilers workshop, maybe you don't want to start your talk explaining what a compiler is. Step 0 of every presentation should be "know your audience", and if your audience is expecting complex information you will have to get into details at some point.
My personal pet peeve of nearly every astronomy documentary/talk/youtube video is that they have to explain what black holes are. By aiming at a way to general audience it wastes the time of the (usually) 99.9% who understand it at least as well as the somewhat brief description they include.
For scientific/technical talks specifically, a truly useful technique I learned from The Craft of Scientific Presentations[1] is the assertion-evidence slide.
This type of slide has a complete sentence at the top that makes a substantive assertion. Then the rest of the slide contains evidence for that assertion usually in the form of a diagram, chart or table. This type of slide is the backbone of a scientific presentation.
All good points, but there are a couple mistakes I often see people (not just beginners) make that did not get mentioned:
1. State the problem (and convince the audience it matters) before rushing to present a solution.
2. Know the background of your audience, and figure out how well they need to know the background to get something out of your talk.
And to the grad students: sometimes, when one only has 10-15 minutes to give a talk, it is hard to give one that makes sense to junior members of the audience and still speak to the experts. Don't give up! and don't be shy about asking speakers questions afterwards. Sure, there are some people who think they are too important to talk to grad students (or even faculty) they don't know. The worst you get from asking such people a questions is you get brushed off (don't worry, they're not likely to remember who you are, and it's not likely to have a negative impact on your career). But more often people are happy to go into details, and/or put you in touch with the postdoc / grad student who knows all the gory details, etc. etc.
State the problem (and convince the audience it matters) before rushing to present a solution.
Exactly. A presentation that is about a "what" or "how" often should start with the "why". The audience needs to care about what is being presented, and often the best way to do that is to remind them why they should care.
Tell them what you are going to tell them, then tell them, then tell them what you told them.
My key advice is to gear the talk to the audience and context. A TED talk will have a completely different presentation style than an academic conference where people might be actually interested in the more nitty gritty details.
> Tell them what you are going to tell them, then tell them, then tell them what you told them.
I see this a lot in talks and even in canned presentations and I don't recommend it.
I think it comes from people padding their talks to fit within a certain time frame. Or perhaps, some speakers have realized that enough audience members utterly tune-out during the talk and need to be reminded of what the talk was about just as they're waking-up and getting ready for the next meeting. Somehow it got drilled into people that this is a necessary structure, but it is not!
Much better, I think, to remove material from the talk to the point where the speaker is making one, maybe two interesting assertions that are supported succinctly. Every talk should leave time for discussion.
This specific aphorism is given as advice by some well known MIT public speaking guy. I think people take it too literally and make an outline slide that's like "motivation, problem statement, method, experiments." So useful...
The best way to make your talk memorable is to focus on substance over style. Work hard on giving something new for the audience to chew on. Without interesting content the audience will tune out no matter how well you structure the talk.
Toastmasters International was mentioned in the article. I just finished my Toastmasters “competent communicator” a few months ago. Here are some of my tips for public speaking in general:
1. Breathe! This is easier said than done when you are caught in an anxiety driven feedback loop where your breadth gets more shallow making you even more anxious. Halt this process by taking deep breadths.
2. Limit caffeine. Caffeine exacerbates feelings of anxiety especially in public speaking scenarios.
3. Beware of ruminating and catastrophic thoughts. You don't need to fight or eliminate them. In fact, doing so may make anxious feelings worse. But in a kind and compassionate manner towards yourself, just acknowledge they are there and the role that anxiety may play in those negative thoughts. Moreover, if you really do screwup the presentation, with some time no one will remember. We can recall great speeches but rarely terrible ones.
4. Attend a local public speaking club such as toastmasters (plug here for Speakeasy II here in Boulder, Colorado). Regularly speaking in front of audiences will take you through the desensitization process and over time giving a speech in front of a group of people will seem not so scary.
Incidentally, yesterday I came across a great scientific talk that I think checked all the boxes here, and then some. 100k views+, but deserves an even larger audience IMO, could clear up some of the confusion about climate (the facts parts):
Dan Britt - Orbits and Ice Ages: The History of Climate
Obviously, I think this is a great talk since I find the topic interesting too. There is probably not much you can do if your audience is uninterested in the first place.
If people are interested in some good scientific and philosophical lectures, I found the ones by Royal Institution to be interesting. Haven't watched all of them, but the ones on memory and neuroscience were particularly good.
Uri Alon gave excellent advice on how to give a good talk. I feel Uri Alon gives much more specific details on how to structure the talk, body language, and most importantly how to handle questions.
For very important presentations, recording yourself can be useful too. It can be excruciating to watch it again, but it forms part of a powerful iterative loop.
Unrelated to the posted article, but native English speakers might have a small disadvantage in getting understood, because they don't always know which word choices are less known among those who learned English as a second language.
And native British English speakers can have a larger disadvantage in getting understood, among people who know English well but are not used to British accents. I was recently in a workshop where I had no trouble understanding the talks given by Chinese, Finnish, German, Indian, Russian, Swedish and whatever speakers. But a native British person, I really had to concentrate because they spoke fast and with a heavy accent. American accents I always find easy, though.
> American accents I always find very easy, though.
That might possibly be due to the hegemony of the American movie/television industry? In my case, I know that a majority of the English-spoken entertainment I consume is American-made, so much that British English now gives me a distinct, "exotic" feeling.
There is this observation that British and American singers sound much alike when they sing. They lose most of their accent, and they just sing in "neutral English". And that neutral English is much closer to American than British. So I'm inclined to think that the British just have more of an accent than Americans. The British accent contains more intonations, which one cannot use when the intonations are dictated by the music.
"Why British Singers Lose Their Accent When Singing"
I can't understand why people would think a non-geographic-specific American accent would be 'neutral' compared to any other. What does 'neutral' even mean? The letter sounds as in the written version of a word?
American accents have as many weird dropping of sounds, adding of extra sounds, and other little curiosities compared to the written version as any other accent does.
For example a non-geographic-specific American accent pronounces 'water' as 'warder', 'butter' as 'budder', and 'duty' as 'doody' - it's definitely taking things away and adding things and not just being neutral.
Maybe I don't know enough about linguistics and terminology, but aren't Americans famous for their dramatic intonation, especially at the end of sentences which aren't questions?
> They lose most of their accent, and they just sing in "neutral English".
There's no such thing. It may be the case that most commercial music is sung in approximately a General American (or maybe sometimes the otherwise largely deceased affected Mid-Atlantic) accent, but that's probably more about commercial market appeal than anything else.
> The British accent contains more intonations, which one cannot use when the intonations are dictated by the music.
It's quite possible to sing music in any of a number of British accents.
Consider Green Day, a band made up of Californians who sing in a working class British accent. The British bands of the 60s sounded semi-American when singing because they were trying to, and the British Invasion bands were trying to sound more American because it was where the market was. The British punk and Britpop bands didn't try to sound American or even mid-Atlantic.
I guess he meant "commercial English", i.e., encompassing large markets. For instance, I love Iron Maiden, but they sound very North-American. Then, in interviews, I find really hard to understand what they're saying.
Yes, it's true that some English bands sing in mid-Atlantic accents, but that's not because it's somehow impossible to sing in an English accent.
I think song lyrics are often easier to understand than ordinary speech just because they're slower, independent of accent. For example, I find it easier to understand Spanish in songs than I do when someone is speaking at a normal conversational pace.
I can think of 3 exceptions. All have the same kind of academic/aristocratic sounding accent: Simon Peyton Jones, Lindybeige, and Rowan Atkinson (in Jonny English, and possibly in Blackadder too). For a time, I thought that's how British sounded like…
Note that you native British English speakers may have trouble with american accents. Though with the influence of Hollywood in movies they probably can pick up some no problems. However don't expect them to pick up a heavy southern drawl though.
There's an interesting point buried here (emphasis mine):
> Presenters often fail because they try to deliver too much complex information. Language and content, normally, has to be designed with the non-specialist SCIENTIST in mind. “You have to think about the least knowledgeable person in your audience THAT YOU CARE ABOUT REACHING” says Rubenson.
If you are presenting a breakthrough in compilers in a compilers workshop, maybe you don't want to start your talk explaining what a compiler is. Step 0 of every presentation should be "know your audience", and if your audience is expecting complex information you will have to get into details at some point.
That doesn't mean that elitism is a good thing. I think both beginners and professionals should walk our of your talk understanding something new. And a good scientific presentation should be also a good presentation. But with limited time (major conferences implemented 20 minutes talks this year), you need to set your baseline knowledge somewhere.