Help Shape the Future of Space Exploration

Join The Planetary Society Now  arrow.png

Join our eNewsletter for updates & action alerts

    Please leave this field empty
Blogs

Headshot of Emily Lakdawalla

An editorial on the LPSC 2013 oral sessions

Posted by Emily Lakdawalla

28-03-2013 12:26 CDT

Topics: about science writing

As you can probably tell by now, I love the Lunar and Planetary Science Conference. It's where I get to hear about brand-new science, and it's also where I get to meet up with current colleagues and old friends. The talks in oral and poster sessions are interesting, but just as important are the hallway conversations.  This year there was a lot of talk about the sequester, and the NASA budget, and about Curiosity, of course.

But there was another topic that kept coming up: the bad quality of a majority of the presentations. I'll admit that it was I who kept bringing this up, but I found plenty of concurrence among fellow attendees. At the end of most talks, I found myself unable to summarize what the speaker had been attempting to convey. Slide after slide of thick text and data and graphs flashed before my eyes, and I took lots of notes. But after most talks, while I could tell you what the speaker had studied, and what data they'd used to study it, I struggled to generalize about what they'd learned.

This was in sharp contrast to how I felt after talks at the Division for Planetary Sciences of the American Astronomical Society meeting in October. There, I was pleasantly surprised to "get" most of the talks, despite the fact that my grasp of astronomy is much weaker than my grasp of geology. One could conclude that planetary astronomers are better at conveying their science than planetary geologists, but I don't think that's the explanation. There are several other differences between the planetary astronomy (DPS) and planetary geology (LPSC) meetings, both functionally and culturally.

  • Talk length: DPS talks are shorter (10 minutes) than LPSC talks (15 minutes). So it's rather surprising that I found DPS talks to be better.
  • Abstract length: DPS abstracts are short, similar in length to abstracts for professional papers. LPSC abstracts are two pages long, like mini-papers, complete with figures and references. LPSC abstracts are so long, in fact, that the abstracts need abstracts for summary purposes in the conference program. (Because this seems slightly silly, there's now a tradition at LPSC, originated by Alan Treiman and popularized by Ralph Lorenz, of submitting abstracts of abstracts in haiku form.) LPSC abstracts are way better and more useful over the long term than DPS abstracts, which are often little more than advertisements. You will often find LPSC abstracts cited in the references to published papers.
  • Culture: For reasons I don't understand, oral sessions at recent DPS meetings seem to be run tightly and discussion is encouraged, while LPSC sessions are less strict and there's less discussion. At recent DPS oral sessions, speakers were held pretty strictly to their 5 or 6 minutes for talks, 3 minutes for questions, and 1 or 2 minutes for transitions between speakers. At LPSC sessions, speakers are supposed to talk for 10 minutes with 3 for questions and 2 for transitions, but most speakers run to 12 or 13 minutes and there is rarely time for more than one quick question between talks. Some scientists (most of them senior ones) have a habit of running totally through their 15 minutes and eating into someone else's talk time, which is, in my humble opinion, arrogant and insulting to their peers.

I think it's a combination of these factors that made me enjoy the oral sessions at last October's Division for Planetary Sciences meeting more than at this year's Lunar and Planetary Science Conference. Consider the talk length. With such short talks at DPS, speakers were forced to distill their presentation into just one or two main points that they wished to convey. At LPSC, speakers rambled on about their data gathering methods and reported lots and lots of data and went down lots of side channels that were tangential to the focus of their investigations. Many of them seemed to be trying to cram most of a 45-minute colloquium talk into their 15-minute oral session slot.

It's possible that LPSC talks run long because the abstracts are long. LPSC abstracts are just awesome. They are condensed versions of professional papers that give you a good solid understanding of the science being done without requiring you to read a 30-page paper. If you find an abstract on a topic from a few years ago that is really important and worth investigating further, odds are good that you'll find a recent professional paper on the same topic. However, two printed pages of text is too much material to get through in a 10-minute talk. People try to do it anyway.

I think that LPSC presenters should lean on their abstracts more. They waste precious minutes talking about calibrations and results from all their different model runs. They should stick all that stuff in the space provided by the two-page abstract, briefly describe their study framework, then use most of their talk time to deliver the most important results, discussion, and conclusions. And then stop when they get to that 10-minute mark, and welcome discussion.

The Lunar and Planetary Science Conference organizers have got to get tough on speakers running over their length. The whole reason for professional meetings is to have face-to-face discussion. I learned a lot from most questions and responses. Presenters shouldn't fear these questions; I'd rather hear people's objections before I submit a paper, so I could anticipate them and handle them in the first draft of a submitted paper rather than on revision (or worse, rejection). Isn't that why I'd be presenting my in-progress work at a meeting in the first place? Do I need to employ reverse psychology and accuse people of not having the guts to face their peers' questions?

There is one problem that's common to the two meetings: bad PowerPoint presentations. This, too, was a topic I discussed with a lot of people at LPSC. I shared some of my recommendations, and people shared some great ones with me. I plan to write a post about that soon, but before I do, I'd like to solicit suggestions on giving good conference talks. Name websites or just offer tips. I have a few resources already (for instance, three volumes of Edward Tufte's work), but I welcome more suggestions!

 
See other posts from March 2013

 

Or read more blog entries about: about science writing

Comments:

Paul Wren: 03/28/2013 01:28 CDT

LPSC 2013 was my first conference of its kind, and I'm gratified to know that speakers blowing through their time limits is not just accepted practice. I found it really annoying, but figured that was just how it is. Thanks for shining a light on this.

Saul Cohen: 03/28/2013 02:08 CDT

Oh man, I've got a million such pet peeves. Here's 10: 1) No font sizes under 20pt. If it can't be read from the back of the room, don't bother putting it on the slide. 2) ONE figure per slide. Two are allowed if and only if you are comparing the two figures. The figure(s) should take up at least 50% of the total slide area, and most of the words you say should be explaining to people what it is they're looking at. Likewise, most of the text on the slide should be explaining to people what it is they're looking at. You've probably been looking at this data every day for a year, but nobody else can figure it out in the minute they'll be looking at it unless you help them out. 3) The axes labels should be at least 16pt. If you can't read it from the back of the room, don't bother putting it on your slide. Note that this means you probably can't cut and paste the graphics out of your paper. Don't be lazy; remake them in a format appropriate for your talk! If you have not put axes labels (or error bars) on your plots, please re-enroll in the nearest undergraduate institution and do not return until you can tell us why units (and uncertainty) are important. 4) Never put a figure or equation on a slide unless you plan to tell people what it means. In the case of an equation, that also means you plan to tell people what all the variables mean. This takes a LOT of time, so minimize the number of equations per slide, probably no more than one. 5) Don't use bright green colors like RGB(0,1,0). On a projector, this will be invisible. That means no default "green" from most graphics tools (I'm looking at you, gnuplot). While you're looking for things that will turn out to be invisible, also make sure your graphics don't have invisibly thin lines, light grays or aliasing artifacts. 6) Spend about a minute per slide. Do not flipbook past 10 slides in a minute; nobody can understand what they're looking at that fast, they'll just get disoriented. Notice that this implies your talk will be only 10-15 slides long! If you just can't bear to leave out this one technical detail you're so sure somebody will ask you about, put it in a "bonus slides" section at the end of the talk and refer to it during the questions. Also to minimize disorientation, try not to flip back and forth through your slides. If something is relevant at two times, just make two copies of it. 7) You can't go wrong with the old advice: Tell people what you're going to tell them. Tell them that. Tell people what you told them. Implicit is that you should know in advance what it is you're trying to tell people. Your title and short abstract should be sufficient that people know what you're going to tell them. End your talk with a summary slide that tells people what you told them. Do not have some silly "Thanks" slide or "The End" slide or a picture of your project team. Leave the summary slide with the takeaway message up on the wall for as long as you can during the final questions. Of course, you may have to go away from it to re-show other slides relevant to a question. 8) Don't run over time. If you're not sure how long your talk is or you've had time problems before, practice in advance so you know the timings of every slide. However... it's not the speaker's responsibility to stay on time. It's the session chair's job to enforce time. Chairs who cannot perform their jobs should be fired and replaced by the conference organizers. (Public shaming optional.) This also goes for questioners who go over time; the question session is for questions, not grandstanding or self-promotion. It is the responsibility of the chair to protect the audience and speaker from such delinquents. 9) This is a little controversial, but I stand by it: your slides should tell the story of your talk all by themselves. Long after your talk is over, when your slides are living in an archived repository of slides from this conference, people will still read them. But they won't get much out of it unless the slides are understandable from the text and graphics alone. You can always embellish and add details verbally, but the key points should be there in eternal text for people of the future. 10) Put your name and title on every slide. Not everybody is going to have been in the room or paying attention while your title slide was up. If somebody realizes halfway through your talk that you're actually really interesting, you want to make it as easy as possible for them to learn more about you and your project.

Elizabeth Koprucki: 03/28/2013 03:45 CDT

Re: presentation visuals: Someone just recommended the book The Functional Art: An Introduction to Information Graphics and Visualization to me. http://www.thefunctionalart.com/ The excerpt I've seen looks excellent. The author's background is in journalism.

squawky: 03/28/2013 04:08 CDT

Planning to assign the chapter on oral communication from Communicating Rocks (Peter Copeland, ISBN 978-0321689672) for our research methods class... it gives some slide design suggestions with examples - including problems with figures. My biggest pet peeves are when the speaker acknowledges their table/graph isn't readable (especially when they're referring to only a couple of rows/columns in a large scanned table), presenters who just read dense text bullets off their slides, and those who follow their text with the laser pointer like the bouncing ball on a karaoke machine. Have gone over my time before - it can be hard to tell where you are in the talk when there's only a "two minute warning" before your time is up... perhaps some of the time issues could be avoided by giving speakers more feedback on how much time they have left. (Yes, practice helps - but I cannot recall ever giving a talk that went exactly as I practiced, and it's not always easy to see your watch while at the podium, nor to use the "presenter's view" with the elapsed time if you are not using your own laptop)

Saul Cohen: 03/28/2013 04:48 CDT

@squawky: When I remember to bring it, I like to use this nifty remote-clicker thing that has a built-in countdown timer. You save time by not having to wander over to the laptop to move to your next slide, and you can set it to vibrate in your hand to indicate time intervals. You can also sneak a peek at the timer display to get the exact time.

Michael Nolan: 03/29/2013 10:41 CDT

It seemed to me this year that a lot of people got to about 8 minutes and it seemed (to me) to be a good place to stop, but they had time left and so then went into way too much detail on something and ended up over (possibly like this sentence). Next time I organize a meeting, I'll try 5 per hour: DPS (6 per hour) is too short, with too much changeover overhead, and LPSC (4 per hour) seems to let people ramble on.

Leave a Comment:

You must be logged in to submit a comment. Log in now.
Facebook Twitter Email RSS AddThis

Blog Search

LightSail - Flight by Light

Support LightSail!

In 2016, The Planetary Society’s LightSail program will take the technology a step further.

I want to help!

Featured Images

Surveyor 7 panorama

Earth, as seen by Surveyor 7
Cratered caprock within Elorza crater, Mars
Point Lake, Curiosity sol 302
More Images

Featured Video

View Larger »

Space in Images

Pretty pictures and
awe-inspiring science.

See More

Join the New Millennium Committee

Let’s invent the future together!

Become a Member

Connect With Us

Facebook! Twitter! Google+ and more…
Continue the conversation with our online community!