Open Thread: How NOT to give a scientific talk

My favorite is to put up a slide with 7 million things on it - which is bad enough in the first place. And then to say "I know you cannot really see any of this....". Then why the heck put it up there in the first place!

It really annoys me when people try to cram every last bit of data they ever generated into a talk. People should really think about what is necessary to include to get your point across, without slamming the audience with every last western blot ever done in the lab...the message ends up getting lost. It also drives me nuts when people try to cram 40 slides into a 20 minute talk, etc.

I suggest the following:
1. Lots of unnecessary and inappropriate animation and sound effects. This is especially fun if you use two different versions of powerpoint where the properties of certain animations change over time and don't do the same thing they used to.
2. Yellow as a font color, or if the projector is older, whatever color it has trouble displaying.
3. One slide full of text which you read verbatim.
4. Leave the laser pointer on as much as possible. Wave it around, point it at nothing, attempt to draw with it to make up for an inadequate picture, etc.
5. On a slide with several graphics, address them in a random order.
6. Switch between light and dark backgrounds between slides if you're going to be in a dark room. A lab mate did this one time and everyone's eyes hurt for the entire presentation.
7. Comic Sans.
8. On images, provide no scale bars. On graphs, use no units. If this is a presentation on a journal article, use the worst resolution image you can find for important data so that it is completely illegible.
9. Don't run spell check.
10. If you're really dedicated, you can try to ensure that your presentation does not display well on the computer you use. By this I mean, compose it on a PC and display on a Mac or vice versa. There's always some random file that won't display properly (especially with images placed in Mac presentations using the paste command when trying to display on a PC).

That's all I've got for now.

ha, sounds like fun!

so many things you can do :)

- tons of animations on slides
- give talk while looking only at slides/computer
- play with your hair/pen/whatever

Lots of material on the same question are at:

http://scienceblogs.com/worldsfair/2008/07/a_list_of_things_to_avoid_at…
http://scienceblogs.com/worldsfair/2008/07/the_speaking_publicly_list_f…

Give no context.
I see this a lot in math talks, which often start with `consider this problem.' WHY should the audience consider this problem? Why is this problem interesting, or even Why does the presenter think the problem is interesting?

Ha!! I love that you will do this! It reminds me of when we used to practice AGU talks together.

Everyone has had good (bad) advice so far. (Although I like comic sans sometimes...) But I have one more - DRESS SLOPPY. Wear painter pants or bright, off the wall patterns or maybe a tiny tee. Tease your hair out or walk into class with one of Minnows toys stuck to you somewhere. Confuse the hell out of them! Make your students think there is a lunatic at the front of the room.

Should be fun!

When I was a grad student, the American Geophysical Union would distribute the classic "Guidelines for Giving a Truly Terrible Talk" along with the abstract acceptance letter (for oral presentations; poster presenters got the companion "How to Prepare a Perfectly Putrid Poster"). The link I've posted reproduces the article in English; the first sentence (in Norwegian) credits the original source. These guidelines assume 35-mm slides, but most of their techniques are equally achievable with PowerPoint.

PowerPoint just makes it easier to give a bad talk. As other posters have said, gratuitous animation (e.g., flying text or nontrivial transitions) is a major turnoff.

Mumble so that no one can hear you.
Only look at your shoes or the screen.
Randomly wave the laser pointer in people's faces.

"Artistic" PowerPoint slides are the worst. I recently sat through a conference presentation in which the text was dark blue on a purple background. No one past the second row could make out a single word. (People in rows 1 and 2 later reported that we didn't miss much, because the slides had no informational content.)

You know, I actually did a post (I hope that tag comes out) on bad presentations I saw at a recent conference! My personal favorites:

- do the entire presentation staring up at your own slides, with your back fully to the audience.
- make...um...sure that you...um...use a really...um...irritating...um...crutch-word, and make...um...sure you...um...use it really...um...often.
- don't dress casual, dress SLUTTY! Something entirely inappropriate with lots of cleavage.
- sayeverythingreallyfastandinamonotone

One of my favorites is to say something like "One thing I need to mention here" before every key point on the slide, this phrase should be reserved for true key points, not as a preface for every graph/figure/bullet point/random thought/parameter not included on the slide.

"Artistic" PowerPoint slides are the worst.

Be thankful to live in the age of PowerPoint. I sat through one very disprganized talk where the speaker used an overhead projector, and frequently walked between the projector and the screen. Then he dropped his transparencies.

Apply verbal stress to the least important words in each sentence, and destress the most important. This will leave your audience wondering what the fuck you are talking about, and by the end of the talk they will all hate you with a burning passion.

My favourites include constantly walking between the projector and screen, forgetting to use slide show mode in powerpoint, having a movie/animation that doesn't work and spending ages on boring introductory stuff that means you skip through all the interesting slides.

I absolutely hate it when people use Powerpoint slide transitions or have arrows flying/dissolving in/out.

I'm so glad you posted this right now, next week is my seminar!

Go over time. A lot. Or just frighten your auidence into thinking you are going to go way over (e.g. put in an intro slide that mentions four subsections to your talk- then, when you've gotten through 80% of your time [maybe even 80% of your material], but only 1/4 of those subsections, put it back up to make your audience think you are going to go on forever).

As an aside- I am suprised by the hatred of animations a couple have expressed.
I'm doing signal transduction, and I work hard to get pathway pictures that don't have a ton of extraneous information (I'm terrified of being asked "what is protein X?" and having to say I don't know). At the same time, when I am watching a talk, I hate when My Favorite Protein (that I *know* is part of the pathway they are talking about!) is not displayed. So I'm going to try to strike a balance, and one of the ways I do this is to put up the slide, and then say "we are focusing on these cytokines" and have little red circles fly in to draw attention to those cytokines. Is this gratuitous animation? Would little arrows that *don't* fly in work better?

I have a few.

1) Instances where the presenters put entire paragraphs on their slide then read it to me. The point of a slide is a mini outline peoples!

2) When the letters clash with the background. Check it out from the projector first please... the computer monitor lies.

3) Inappropriate noises from the speaker. There was one speaker who inhales between her teeth so it was always a hiss. Another (and I know we all do it here where I live) who kept sucking her teeth. A third had an obnoxious nervous chuckle and his slides being out of order made it worse... noises... eek.

Not rehearsing a presentation. This overlaps with others here, but few things are more annoying than having someone explain a slide they haven't seen since they created it or thought about what they were going to say about it.

I laughed out loud at the comments above me. Hilarious!
One more I can think of is constantly saying: So there are 2 or 3 things to consider here... Discuss one, go off on a tangent and never, ever go back to the second or third. Also, make sure these points are not on your slides (whichever format you use).
And, use different lay-outs for each and every slide. Don't let them slide into expectations so they can focus on your content.

ha! I like all the suggestions above ... I would also add that you end each statement with your tone going "up" ... like you're asking a question (even if it's not a question). A lot of younger and less confident speakers tend to do that.

One of the worst talks I saw, the guy used both a laser pointer AND the cursor to follow the bullet points he was reading verbatim ... the laser dot and cursor danced around each other like they were racing to the end of the sentence. I have no idea what he said.

I attended a "read the slide to the audience" talk just yesterday. Slides are there to highlight main points and allow you to show pictures an graphs. They aren't a transcript.

Similar to only talking to the slides, just ignore everything and read your note cards or script instead.

Capitalize some titiles/bullet points, but not others. Choose randomly. Do the same for punctuation. It isn't as obvious as misspellings but much more common!

Ask a student to serve as time keeper and give them little signs for 5 and 1 minute. When they flash the first sign, look dumbstruck. When they flash the second, skip from your statement of the problem (on which you spent all the time), obviously blowing through slides, and go to the conclusion, which obviously makes no sense without the middle of the talk.

Arrange for someone to call your cell during the presentation. Take the call.

- Use irrelevant clip art (or images from some other source) while talking about a vaguely related topic. (For instance, show a picture of a hula girl to introduce Hawaiian volcanoes. Actually, that was sexist as well as being distracting.)

- Show graphs that have been scanned directly from a paper, with too much data to read at once, and symbols that can't be distinguished. And then don't talk about the graph at all when it's on the screen.

- "As you can see..." Especially when it's impossible to see anything that supports your point.

(Also, you've seen the Gettysburg Address powerpoint presentation, haven't you?)

i have to second scicurious in that using um.... repetitively, or 'like' is horrendous to listen to.

The worst three I've seen:

* Someone whose entire presentation consisted of tiny graphs of all the data he'd ever collected, with minimal spoken or written explanation.

* Someone whose entire presentation consisted of a laundry list of all the molecules he and his students had purified or synthesized in the lab.

* Someone who gave two presentations that each consisted of over 50 slides, tied together by no discernible thread, while the presenter yammered away a mile a minute.

All were ~45- minute presentations. The last two were job talks (the middle one by someone who already has a tenured professorship somewhere!) While the first two were merely boring, the third was truly disturbing; I seriously think the person in question was off meds against doctor's orders that week.

Awesome suggestions so far!
My 3 pet peeves for talks:
1. Too many words on a slide. The worse example of this I ever saw had over 75 words on a slide (yes I counted, no I wasn't listening to the speaker.)

2. Reading the slide to an audience! This one really drives me crazy, it's so tedious and always makes me want to leave the room.

3. Using the laser pointer like it's a light-saber and the words on the screen are attaching you. It takes a great amount of self control to not overdo it with the laser pointer but the pointer dance will just about leave me seasick but the end of a talk.

Use completely uninformative slide headings. The worst is when half the slides are headed "Experiment 1", a lot more are named "Experiment 2", and the only digressions are things like "The Question" and "The Prediction", instead of something useful.

Alternatively, go too far in descriptive headings - "One-Way Ancova supports the role of self-importance in academia's bombastic tendencies" - that take up tons of space and require a tiny font.

These suggestions are hysterical...my only bit of advice not previously mentioned is:
Wardrobe Malfunction
Especially if you can make everyone in the audience truly embarrased & uncomfortable.

Can't wait to hear how the presentation goes...sounds like it will take a lot of preparation & stone-faced acting to weave all these elements together. Should be a riot!

Use a different font/color scheme on each slide. Make some crappy and some have a good style. That way, at the end of the talk you can demonstrate that, no matter how much they like Broadway font, it's impossible to read. The same can be done for the color schemes.

ah! SUCH a good idea! if (when?) i'm a professor i am totally going to steal this idea!

Wow! You all rock! I *love* the suggestions and thanks for the links. Now I just have to figure out how to work as many of these into a 15 minute talk as I can....of course, I may say it's a 15 minute talk, but of course it's going to have 50 slides and about 30 minutes of material. :)

You might want to do two shorter presentations: one showing what should be done and the second showing what shouldn't be done. As much fun as it might be to break all the rules about what makes a good presentation, your students will probably appreciate the contrast between the two presentations rather than just being pounded with everything done wrong. They need to on some level have confidence that you're teaching them because you know how to present information effectively.

Good luck!

Oooh! Ooh! My favorite is when the presenter shows a slide, but then talks about something totally different and doesn't explain at least one of the panels.

A few to "ditto:"
1. Use the interrogative tone all the time? It makes the presenter sound so confident?
2. Use a pre-set PowerPoint background (some have called these "artistic" which has built-in positions for the title, etc...and then have text/graphics which clearly violate those pre-determined spaces. You could combine this with some awkward color combos and text.
3. Poor attire, back to the audience.

I think these may be at least slightly novel:
1. When you get close to going overtime, either blow through or skip entirely a large number of slides, and then expect your audience to remember what was on them when you get to the conclusion.
2. Keep turning the room lights on and off depending on what's your slide...and always say, "oh, this looks so much better on the screen! Maybe if I hit the lights you'll get a better view," when you do this.

I actually do a similar assignment for one of my classes, and then I give them 5-10 truly awful PowerPoint slides and have a contest to see who can find the greatest number of reasonable criticisms. The winner gets a few extra credit points.

"Be thankful to live in the age of PowerPoint" ?
Doesn't anyone remember that PowerPoint is evil? :-)
http://www.wired.com/wired/archive/11.09/ppt2.html

How about this?
http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001yB&topic…

If you want to make a truly bad presentation, duplicate some of the techniques displayed at the second link. (Both items from Edward Tufte)

If you are not drunk, you should at least spill bourbon on your sweatshirt just before entering the room. If you can exhale bourbon bon-bon breath on a few students that may help the illusion.

The good ones are taken, but also: after you forward to the next slide, say "Oh, I forgot to tell you something on the last slide", then click forwards on accident, then back one or two times too many on accident, and then finally forward to the one you wanted. Then, when you click twice to get where you should be, go one too far forward. Repeat.

MiddleO'Nowhere: 3. One slide full of text which you read verbatim.

You mean like these?

Want to give an interesting talk = DO NOT use PowerPoint (or any other slide-ware). Tufte says it best:

http://www.edwardtufte.com/tufte/books_pp

It's interesting that so many of the comments just assume that slide-ware will be used. When I think bad to the few really good presentations I have seen, slide-ware was nowhere to be found (not to be confused with projecting an occasional "illustration" - which is a very different thing.)

I assume your audience came to see *you* - not your "slides."

P.S.

I've seen Tufte give a presentation. He practices what he preaches, and held an audience spell-bound for several hours!

Addendum: A link to great tips - this can be hard work (see tip #11), but it pays off.

Tufte's Tips for Successful Presentation

http://www.edwardtufte.com/tufte/advocate_flatland2

1. Too many words on a slide. The worse example of this I ever saw had over 75 words on a slide (yes I counted, no I wasn't listening to the speaker.)

Expand your horizons. The second slide over there has well over a hundred.

All of my favorites are taken, but here's another one: accidentally bop out of PowerPoint slide mode into design mode, fiddle around trying to get back into slide mode, and then ask if anyone knows how to help. This works best if you've got more than one helpful student, because then you'll have a crowd of three hanging around your PC punching various buttons and arguing about how to do it.

Yes, I have actually seen this in action, more than once, and with more than one presenter.

Hey, noone mentioned one of my faves - when a slide comes up, look at it in amazement and discuss aloud with yourself how it came to be there, wonder what you meant to say to it, then tell a totally irrelevant anecdote that it reminded you of.

Also, definitely mumble.

These are all really great. ScienceWoman, PLEASE record your talk for YouTube!

When they flash the first sign, look dumbstruck. When they flash the second, skip from your statement of the problem (on which you spent all the time), obviously blowing through slides, and go to the conclusion, which obviously makes no sense without the middle of the talk.

Ouch! I've done that, but in my defense I had only 5 minutes to start with and had to rewrite my carefully crafted comments at the last moment when the event organizers decided to nix the panelists' visual aids. I was less than halfway through my comments when the moderator flashed the "1 minute" sign, so I had to cram looking dumbstruck and rushing through to the conclusion into a paltry 60 seconds.

I would add the following "how not's" for presenters:

For lay audiences drawn to your talk by the extremely colorful, charismatic nature of its subject, use the worst illustrations possible, preferably your own crude sketches and/or black and white photos.

Give your audience members a combined math and vision test by presenting all your data as barely legible numbers in densely-packed tables (charts and diagrams are for sissies!).

And these for meeting organizers (not specific to science talks):

If practical, place the podium next to or even in front of the room's exit doors so that anyone trying to sneak out will be as conspicuous and disruptive as possible.

Use the smallest, shortest screen available so that no one past the first two rows can see the whole screen.

Crinkle the water bottle or rattle the change in your pocket every 2 secs.

Take a swig of water after each slide while pacing in the front looking very important and pensive and rubbing your head.

Read the text slide to me... then go back to a figure slide before. Then jump ahead to another text slide... then go back to the same figure slide before. Then jump ahead to next text slide... rinse repeat.

Put a jumping banana in the upper corner on a loop. And make sure that jumping banana screws up the font on all your slides.

Walk up and down the aisles while talking, and make sure you touch people on the tops of their heads or their shoulders as you pass by. But seriously, try not to give the old dudes a heart attack...

Keep moving the podium around.

Let a really whack screensaver kick on (like say, more jumping bananas in full screen)

Thanks for all the laughs! Will you be able to get through your talk without laughing?

Here's another: tell a really inappropriate joke. And then laugh at it :)

- Open with a painfully unfunny or borderline offensive joke.

- Screw up the punchline to the aforementioned joke.

- Illustrate one of your points with a lame, outdated pop-culture reference, for example, Bob Uecker.

- Use "air quotes."

I love all the suggestions (we all act as if we haven't had the too many word problem in ours lives, ha!)

As a professor, I think this is a great idea (one that I might have to co-opt in my seminar class). I, however, second the idea of giving the students an example of how to give good talk as well. Perhaps the don't talk first, followed by a discussion of all the parts that went wrong, followed by a good talk, again with a discussion of why it was good talk. I think the contrast will be striking to them and a valuable learning experience (man, the amount of time I have seen students spend on choosing the right color scheme, vs the actual content has been astounding).

These are great!

In one of the worst talks I ever saw, the speaker had a microphone. He held it up, I swear, so that it was touching his lips, then proceeded to talk at about 100 miles per hour, very loudly and incoherently, while turning his back to the audience. It was the most painful talk I have ever heard. (And to top it off, he was also guilty of the 100-words-per-slide affliction.)

Although on the bright side, his talk preceded mine, so I looked like a superstar by comparison! :)

In one of the worst talks I saw during my grad school days, the speaker appeared to have put his slides in the carousel in semi-random order. Not only did he have to search back and forth through the slides to find the one he wanted to talk about, some of them were backwards or upside down. In the last five minutes he just skipped through about 20 to get to the conclusions. PowerPoint at least fixes the problem with slides in the wrong orientation.

Ugh.

The thing is, you can't teach people to do a good thing by showing them how not to do something. If you want someone to do something well, you need to show them how to do it right.

You should probably put together two presentations--- first do everything you can wrong, then turn around and re-present the data as well as you can in another 15 minute talk.

Cheers.

Lots of great suggestions for a horrible talk. I'll just add a couple (which do overlap with some comments above):
- make sure your talk illustrates your slides, not the other way around.
- try to average less than 1.5 minutes per slide (the less the better, to inflict maximum confusion)
- Plan on keeping the audiance hostage for at least 25% more time than you are allocated (because your project is the most important and most interesting one in the world to anybody who would be in attendance, they won't mind at all)
- Assume everyone there is as familiar as you are with the field, the problem, and the jargon. (especially the jargon)

Make sure to do this after the drop date or give them some hint in advance that the lecture is non-standard (will they have had other lectures from you before this one). If I get a sufficiently bad lecture I'll just walk out and drop the class.

1) Read from a paper. Use no slides or visual aides of any kind. Never look up from the aforementioned paper to see if *any* of the audience is with you. (No joke - I sat through such a talk.)

2) Give a talk that is so technical/specialized that only one other person in the room understands it. (It's okay - the rest of us needed a nap anyway.) Then direct your entire talk at this person, never making eye contact with anyone else. (Next time just offer to buy the guy lunch!)

3) Be LATE for your talk and then insist on interupting the next person who is already talking so that you can give your talk NOW because you have someplace else to be. (Moderators should lay the law down here, but for some reason let this person get away with this.)

Tangential issues...

1) Moderators - allow your favorites/students all the extra time they need, but hold everyone else to the alloted time.

2) Department heads - introduce new faculty who are giving a talk by saying how you are not paying them nearly enough. (Happened to me. No joke.)

Two points, one of which overlaps with another made above:

*make sure any special characters (like variables) turn in to skull-and-crossbones or something due to lack of compatibility between Mac and PC fonts

*talk about five slides worth of material while on the first of the five slides, then flip through the other four you've talked about, muttering that you already covered that as you flip slides; that way students have no chance to take notes

Rush through your equations like you're afraid of them, preferably without defining any of the variables or explaining how they related to the scientific question being explored.

Don't forget to use really small font (10 point maximum, 8 is better) so that even those in the front row can't read the slides. Put tons of equations on them, then explain them in more detail than necessary, or don't explain them at all. Gear the talk for the wrong audience. I think either too technical or too basic would work.

When you are talking look really frightened, speak quietly and partially hide behind some furniture like desk, computer stand, chairs or anything else that is at the front of the lecture room.

Use a gratuitous 'fuck' or two. Hell if it is good enough for some to use in every blog comment they post it oughta work great in an oral presentation where everyone is falling asleep anyways.

Oi, I took a class similar to the one you're teaching and had to sit through many painful presentations by my peers. These are my pet peeves:

-Rock back and forth so everyone gets seasick.
-Put up graphs and tables but no figure legends or descriptions of any sort.
-Talk in a monotone and/or either really quietly or really loudly.
-"Recite" your presentation. I've seen several presentations (even at conferences) where students obviously practiced, but practiced so much they memorized the entire presentation word-for-word. Not only did it sound like a recitation (and was extremely boring), but if they forgot one of their memorized phrases, they were completely thrown and sometimes couldn't get back on track.
-Use lots of technical terms and don't explain what they mean.

I realize not everyone likes powerpoint presentations, but speaking as someone who is hard of hearing, I benefit a great deal from having visual aids and summaries of what is being discussed. People have a wide variety of learning styles, and not everyone can follow presentations that are only accessible aurally. (Even with ASL intepreters, I still have trouble following along. The presentation is still being presented "orally" for me, and I don't learn best through that route.)

Oh dear, these wouldn't be half as funny if we hadn't all seen them before.

I'm not sure if it's been emphasised enough that not only should you use lots of animations, they should all be different - see if you can use every trick powerpoint can provide. This includes bringing in key words one letter at a time and making the whole word bounce when the last letter arrives.

My absolute favourite talk was by a student who apparently needed to stroke himself for reassurance all the way through. It started by wiping sweaty palms on his jeans, then rubbing his belly, then hand under shirt rubbing belly and chest, then his back, and the piece de resistance - hand down the back of the jeans, rubbing his arse. No idea what he was talking about.

Erm, not sure if a lecturer should try that one though...

Please begin your talk with "The title of my talk is... "
Include plenty of scientific cliches such as "I'm going to shift gears here..."
-----
Becca asked about "gratuitous" animation. I don't think it's inherently counterproductive to have a simple arrow or circle that *usefully* draws the viewer's attention. But since it does annoy some people, I use the pointer instead.

A good result of this method is that I have to slow down enough to clearly indicate the desired item and then check in with the audience to be reassured that they got it.

On the other hand, movies that show an actual process (in photograph or cartoon form) are great.

Awesome post and comments. I'm sending this link to my labmates.

Agreed, TCS. Some animation is useful. I actually have a model that I animate to show neurobiological changes following chronic drug exposure. If I didn't animate it, it would be hard to explain, and the visual really helps people.

However, I will say, don't animate your text unless you need people to focus on it in particular. I recently attended a talk where the guy finished it off with a "thank you" slide of his lab. The "thank you" waved around like a flag the whole time and was absolutely horrid. Don't do it!

Ok, as someone whose major research and teaching field is technical communication - and specifically how students learn to communicate as professionals and what teaching practices support/hinder student learning, I'm going to dive in here.

All the comments on good and bad talks are interesting and helpful, and the idea of starting off with a bad talk is great. BUT for your students' sakes as well as your own, I'd also step back and think about what you're actually trying to teach them. If "how to give a talk" is simply a set of guidelines (use more PowerPoint; use less PowerPoint; use more animation; use less animation), you're not really teaching them anything transferrable that they can apply in multiple settings under different conditions. Even things on this list that people claim are terrible to do in a talk can in fact be useful under the right set of conditions (ok, except maybe the one about coming drunk - though it worked for Dean Martin an awful lot).

What you really want them to understand is that a talk - scientific, business, or otherwise - is about an exchange of information between one person who has this information and other people who want it in order to accomplish some larger goal (this, of course, leads to the problem with most classroom instruction in oral communication - most of the audience doesn't want or need the information the speaker is giving . . . there are ways to address that but I do have actual work to do at some point this evening).

So, what you want your students to learn to do is to think about who the audience is, why they've shown up in the room, and what they're looking for. At the same time, as speakers they need to think about what they want to accomplish with the talk (get useful feedback? make connections? get grant funding? find collaborators? bore people to tears?) Those questions can have very different answers depending on the circumstances. A good communicator will design a talk that helps everyone accomplish their goals as efficiently and effectively as possible.

So when you do your bad talk and then critique it, push the meta issues, not the rules: Why is this bad? What does the audience need from the talk that they're not getting? How does this element of the talk work against the speaker's and/or the audience's goals? When might a strategy like X or Y be good? When would it be bad? What does it achieve or hinder?

The truth is that there is no real list of things that "should not be done in an oral presentation." If I had world enough and time enough, I could go back through almost every "thing not to do" and come up with a reason it could be effective under a particular set of conditions. Teach them to design communication to meet the needs of the individuals involved, not to follow someone's set of rules that apply to this medium in this context at this point in time.

Ok, off the soap box and back to work.

I am going to share the two most excruciating examples. First, I set through an hour long seminar where the presenter used 5 slides. Yes FIVE. It was painful how he droned on and on and on, still pointing to the same slide after 20 minutes. The kicker is, he went over time. The second example would fall under the category of know your audience. A dermatologist presented a bunch of molecular biologists with slide after slide of HPV warts. We wanted graphs and pvalues. Instead we got butt-holes, vaginas, toes, fingers, etc. overflowing with warts. These were close-ups too. We were all sitting there horrified. I decided this was the talk he normally gave to doctors and med students. It was so gross. I am still suffering from this.

I'm surprised noone has come up with this: if you are forced to present in a language that is not your first language, translate your slides using Babelfish and don't get anyone to look them over. I kid you not. It was a real shame for the guy I saw do this actually: I was at an English language conference where there were a lot of Japanese students nervously presenting their research, apparently with their supervisors in the audience. They were so scared. One guy did the above, and I am embarrassed to say I got the giggles really, really badly. There were some genuinely funny mis-translations. Luckily it was a huge auditorium and I was near the back and was able to stifle myself so he couldn't hear me.

Use a bunch of highly technical terms and acronyms. Its always nice to go to a talk that is given to the 3-4 people in the audience that work specifically in that area.

Great advice! I'm going to send my grad student to read this post.

Here aretwo more things that I think detract from a presentation:

1) Ending a talk with a slide that says "Thanks for listening" or "Questions?" and leaving that slide on for the entire question and answer session.

Why not end with a conclusions slide: this gives your message more air time with the audience and can also help prompt questions from listeners who may have forgotten some points they wanted to ask about during the exciting discussion section.

2) Empty overview slide: This is similar to the comment about empty headers by gradstudent above. It always puzzles me when people have a talk overview slide with the points:

1. Introduction
2. Background
3. Method
4. Results
5. Discussion

Huh? This gives no information relating to the content of the talk and just wastes time. Totally predictable information like this is unnecessary. Either the overview should be specific enough to be unique for that particular talk, or it is unnecessary.

Are you focusing on the oral presentation, the content of the presentation, or the use of a supplementary A/V presentation? Seems like most people are focusing on the technology, and that's really disappointing. A good presenter doesn't require the crutch of MS Powerpoint. And I think a telltale sign of the times is the number of academics commenting on that very point.

My personal pet peeve, mentioned only briefly, is the speaker who stares at the same person for the entire presentation, only rarely and briefly breaking the stare to glance quickly at someone else or a point on the back wall before reverting back to the original staree. And I especially hate it when I am the one being stared at.

A second peeve is the person who declines the microphone and proceeds to mutter the entire presentation under his breath.

A third is the speaker who takes questions but doesn't repeat them to the entire audience and so loses the attention and interest of everyone except the person who asked the question.