A long guide to giving a short academic talk (2022) | Hacker News Reader

A long guide to giving a short academic talk (2022)

benjaminnoble.org

131 pointssebg20 comments

Regarding the q & a, a few tips.

Somewhat obvious, but try to anticipate what questions might be asked, and come up with good responses in advance. In addition, you can potentially create backup slides to support key points.

Make sure you understand the question(s). Feel free to ask the questioner to clarify. It's also fine to try and paraphrase their question and ask if you understand them correctly. In fact I recommend doing this even when you're pretty sure you understand, since it (i) buys you time to think of a response (ii) helps others in the audience who might not have heard (iii) is especially useful if the presentation is recorded and the questioners don't have a mic.

Try to keep your responses brief and to the point. This gives time to more people to ask questions. Rambling on and on when a short answer would suffice often makes people think you don't know what you are taking about. In a technical presentation it's often much better to simply say "that's not something we've considered, interesting", or "that's something I'd need to think a bit more about" than to try and bluff. But obviously you don't want to do this for every question.

If someone is being really awkward and/or you really can't seem to answer this question, better to suggest taking it offline to avoid hogging the whole q & a period.

Try not to get too defensive.

When practicing your presentation, ask people to be as picky as possible and make you think on your feet. Hopefully the actual questions you get will be easier.

IMO handling the q & a well is almost as important than the presentation itself. It's a real shame when I see some interesting work with a good presentation and the reviewer then undoes it all by not being able to respond well to basic questions.

To this list, I'd add one more thing: pay attention to talks that you see, writing notes not just on the material, but on things you liked and disliked about the presentation style, the slide format, etc. Avoid the things you dislike, and employ the things you like.

This applies to all manner of things.

You may notice that speakers who talk to the screen are unengaging. So avoid doing that.

You may notice that strong speakers have a moment near the beginning when they grab the attention of the audience, and that this makes it easy to stay awake for the whole talk. So try to find a way to do that -- a way that fits with your own work, of course.

You will certainly notice that talks that go overtime make for a terrible experience for everybody. So don't do that. (You'll need strategies. The simplest is to have "landing points" in the talk, so that if you see that time is running short, you can skip some material to get to a landing point. Obviously, the conclusions slide ought to be a landing point.)

And so on. Learn from the impression that other speakers make on you.

>You will certainly notice that talks that go overtime make for a terrible experience for everybody. So don't do that. (You'll need strategies. The simplest is to have "landing points" in the talk, so that if you see that time is running short, you can skip some material to get to a landing point. Obviously, the conclusions slide ought to be a landing point.)

I won't say I've never gone a few minutes over when there's a break coming up and the audience is engaging. But, yeah in general, it's especially rude when the next speaker is waiting to get setup, and that time they're left with 1 minute to plug in will be the time something goes wrong.

You should have a decent sense for pacing of the talk anyway but, as you say, you should also know which material lends itself naturally to "in the interests of time, I'm not going to spend a lot of time on this topic today but we can talk afterwards if you're interested sort of thing."

Conference talks are also different lengths and they've tended towards shorter over time. So, even if you can adjust on the fly to some degree, consider not using the same deck for a 30 minute talk and a 50 minute one. (I'm actually a fan of talks that are on the shorter side in general.)

I have mixed feelings about Q&A in the session itself in general for a lot of reasons. But you need to keep things from getting too diverted or getting down the "more a comment than a question" path. (Which is part of the mixed feelings--it's hard to cut off without seeming a bit rude.)

I've used this approach for slides and websites. Any time I see a great design, I'll take a screenshot and save it in a folder called "PowerPoint/Web Ideas". I'll even include some shots from typography style music videos. Then whenever I have some free time and am bored, I'll try to recreate it in PowerPoint.

Most of the designs are TOO MUCH for teaching slides, but its fun practice since making slides is sort of my job. But the practice gives me a better understanding of what I can do with presentations, building templates, messing with color theory, etc. There's a science/art to making a good lecture presentation and I'm enjoying the artistic expression, while still being educational task.

also part of paying attention to other talks...I always make a reference to one or two talks that just happened or one that I know will be coming up.

It's one more point of context for the audience, and people always love it when others talk about them.

Unfortunately in grad school I noticed a lot of cargoculting bad presentations. People would copy people that were good scientists but bad presenters because “that’s what good scientists are doing”.

Related: there’s a worry that if you make the presentation too easy to understand, people won’t think you’re smart. Inscrutable presentations must mean that the presnter is a genius operating on a higher level /s

"Once you write the slides and script, it’s time to practice… a lot."

All great speakers are look great because they practice and iterate on the material and end up with polished presentations that they've given dozens or hundreds of times.

I don't present nearly as much as I used to--and I'm mostly not great at doing serious practice runs (as opposed to mentally going through the material)--but I've often observed that my second or third time giving a presentation is usually a bit better than the first.

practicing talks is great at highlighting two things:

- helping you come up with natural, smooth segues between slides/ideas

- making it clear if there’s any missing background/detail you need to add or reorder slides. One of the worst things to do as a presenter is to realize this mid-talk and then start jumping through slides out of order

Or one can quickly learn how to achieve clear presentations by indexing the common errata:

"Chicken chicken chicken" (Yoram Bauman)

https://www.youtube.com/watch?v=yL_-1d9OSdk

This should be required viewing for every academic author. =3

Good point that the talk is as much a sales effort as anything else...

Is your audience receptive to what you have to say? are they curious and inclined to believe you? Tell them your conclusions right up front, so if they don't hear anything else you say, they learned what they need to know. It will also help focus their attention as you proceed.

If instead they are apt to push back on what you plan to say (because it would eliminate their jobs/research for example), then instead you start by building a case with a series of steps they will agree with till, having led them down that primrose path, you reveal to your new allies the inescapable conclusion. They'll still push back, but they will at least have learned what you have to say.

> every talk is a job talk

Absolutely. A subset of the people in the audience are tenured or tenure-track academics. These folks get grants for postdocs. Moreover, they may be on a hiring committee. Even better: if you leave a favorable impression and a colleague of theirs has a job opening, they might recommend you.

Treat (academic) presentations as job talks. Good impressions open doors, not just in the immediate future but also further out than that.

Here is my own list of recommendation (written down to give interns): https://nestordemeure.github.io/writing/guides/presentation/

It's not what you say, it's how you say it.

Obviously it's both. If you have something worthwhile to say, but say it poorly, people will find it difficult to care. Similarly, if you have nothing important to say, but present it in an engaging manner, people will be interested at first, but will cotton on to the lack of intellectual content pretty quickly.

To make a lasting impact you need both.

Most conferences I've attended are so boring that I'd take an engaging speaker saying nothing over a poor speaker presenting something of consequence.

There's really nothing I like about attending conferences and probably 1 in 100 speakers are both engaging and presenting something of value. The vast majority of them seem to be there to promote their latest book. Yet my employer seems to think they are important.

For me, conferences are about:

- The hallway track

- Having some less-distracted time around some general topic of interest

- Being made aware of things I might not have otherwise been made aware of

But, by and large, you're not IMO going to "learn about $X" by going to a breakout. It might put $X on your radar as something worth spending a day looking at.

By and large, I like (selectively) going to conferences but if someone has the mindset that they're a day of classroom-type sessions, they're probably less valuable. That's not a criticism. People just have different perspectives.

> if you want others to read your work, you cannot simply publish it and assume others will find (and cite) it. You need to sell it.

The LLM's will read it. Currently they will not cite it, but there is some work on linking knowledge representations with source material. If this could remove the need to sell it, I think academia would be much better off. The best researchers are not necessarily the best presenters.

And the most successful academics aren't necessarily the best researchers.

Reply on news.ycombinator.com