tag:blogger.com,1999:blog-9186457242428335144.post1053610130166831499..comments2024-03-28T00:36:13.790-07:00Comments on Volatile and Decentralized: Conference talk pet peevesMatt Welshhttp://www.blogger.com/profile/04255792550910131960noreply@blogger.comBlogger17125tag:blogger.com,1999:blog-9186457242428335144.post-49901037187836629522010-11-14T09:32:04.067-08:002010-11-14T09:32:04.067-08:00Actually in fundamental sciences, there are legion...Actually in fundamental sciences, there are legions of bad speakers whose work is very influential. Maths would be a primary source of examples, just go to your math department seminars some day and you will see. Theoretical computer science is a bit better but not always.<br />The major issue with your post is that you didn't mention talk structure, content or idea organization at all, just how the presentation is delivered. I do not challenge the fact that this can be a recipe for your personal success in many industries and even academic fields, but it does not mean you have made an interesting contribution.<br />And it's too bad if this becomes your only way to judge an academic talk or the reason why you'll "spent hundreds of dollars" to travel to the conference. I would have said meeting your colleagues and discussing ideas as the primary reason to do that.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-41089425532182820252010-11-06T14:16:11.510-07:002010-11-06T14:16:11.510-07:00Anon re: comparison to Stephen Hawking. You can...Anon re: comparison to Stephen Hawking. You can't be serious. How many people in this world are as evidently brilliant as Stephen Hawking? Note that he made his career not just by being a great physicist; he has also published popular books (e.g., "A Brief History of Time") which falls very much in the same category as being able to deliver a good presentation. I challenge you to name one successful academic computer scientist (or any other scientist, for that matter) who is incapable of giving a good talk, or presenting their work in some other manner.<br /><br />Anon re: "Not everyone will email you the PPT." This does not make it OK to take videos of the talks at conferences. You're not entitled to take home a copy of the presentation; you paid the conference registration fee, you get to listen to the talks, and you get the proceedings. Amateur conference videography without the speaker's consent is pretty obnoxious.Matt Welshhttps://www.blogger.com/profile/07077674014671176946noreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-40896316059258418522010-11-06T08:52:36.598-07:002010-11-06T08:52:36.598-07:00For the previous anonymous:
... and they precisel...For the previous anonymous:<br /><br />... and they precisely were...?<br />If you are sure they were wrong, please point them out clearly.<br />In the end, you are an anonymous, you risk nothing!<br />Otherwise, sorry, but you are just wasting your time...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-51778059232234303022010-11-05T11:23:52.939-07:002010-11-05T11:23:52.939-07:00I think my #1 pet peeve is "papers which are ...I think my #1 pet peeve is "papers which are wrong." There were a couple of those at this conference (there've been a few previous years, too), where you wonder what the PC was thinking...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-53943483436568400192010-11-05T09:16:51.571-07:002010-11-05T09:16:51.571-07:00Actually not everyone would email you the ppt. You...Actually not everyone would email you the ppt. You would be surprised. <br /><br />If you look at MSR people very few post ppts on their websites and a lot of them are not so responsive when you ask for the slides.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-54159927740778008332010-11-05T08:15:33.592-07:002010-11-05T08:15:33.592-07:00If you are unable to give a good talk on your work...<i>If you are unable to give a good talk on your work, you are unlikely to be successful in an academic career. Giving a bad talk at a conference can kill any hopes a grad student might have of getting good faculty job interviews - same goes if you give a bad talk during the interview. If you think this only matters for academics, think again. </i><br /><br />Hmm, I think of it this way. Say there is a brilliant mind trapped within a less useful body, say Stephen Hawking. And the body is an impediment to making a good presentation. Of course for whichever organization planning to hire Prof Hawking (this is an example of course), because he is so good <i>at being a physicist</i>, they're likely to overlook the fact that he is probably not going to be as impressive a public speaker as say John Stewart. But, really, who cares about his public speaking?<br /><br />On the other hand, I guess having such skills will help when one isn't as good a researcher as others: we use public speaking to 'shore up' our resumes. But in those instances, we're no longer looking at the 'best of the best', other... political factors get in the way. One might argue and say, "Well, that's because all of us are the best-of-the-bestest". In which case my reply would be: are we falling deeper into the mode of cultivating politicians rather than good scientists? And the answer clearly MUST be: we must first and foremost consider the intellectual prowess of the researcher. If all else is equal, then and only then should icings on the cake, like good presentation skills, come into the picture.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-1759143939101182782010-11-05T00:43:42.885-07:002010-11-05T00:43:42.885-07:00Anon - "What matters is the work which took m...Anon - "What matters is the work which took months, not the presentation which takes minutes." I could not disagree more. BOTH the work AND the presentation matter, tremendously. If you are unable to give a good talk on your work, you are unlikely to be successful in an academic career. Giving a bad talk at a conference can kill any hopes a grad student might have of getting good faculty job interviews - same goes if you give a bad talk during the interview. If you think this only matters for academics, think again. The most embarrassing talk I saw at a conference was from a fairly well-respected industrial lab where the person had clearly not practiced the talk AT ALL - made even more embarrassing by the fact that this was the award paper at the conference. People remember this kind of thing. Maybe you think it shouldn't matter, but the bottom line is that effectively communicating your work is just as important as the work you do.Matt Welshhttps://www.blogger.com/profile/07077674014671176946noreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-3028528809199949862010-11-04T17:34:43.759-07:002010-11-04T17:34:43.759-07:00A grad student once complemented me on the beautif...<i>A grad student once complemented me on the beautiful font choice in my talk -- it was Helvetica.</i><br /><br />That's nerd talk used for hitting on other nerds. So, was she (or was it a he) cute? :)<br /><br />On a more serious note, I find that the things mentioned above are more or less condiments. I mean, if the presenter is basically saying he's trashed my work, I'm going to be fuming and clinging onto his every word, thinking about what I'll say when I come up with a better algorithm / system / etc. Wait, I'm still joking. I think.<br /><br />Anyway, if our opinions of a researcher are influenced by how he / she presents, that's a slippery slope which unfortunately I think networking today is slipping further down on. Like anonymous above said, even if the presenter reads his / her paper aloud word-by-word, that should still be ok. What matters is the work which took months, not the presentation which takes minutes. On a similar note, what matters is that the work has significant impact, not that the researcher has lots of papers.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-64683553618250803502010-11-04T04:16:51.627-07:002010-11-04T04:16:51.627-07:00Also, if I can suggest another point...
Stick to ...Also, if I can suggest another point...<br /><br />Stick to the time limit! Not only is it annoying when a talk runs long, but it makes your talk suck because you end up zooming through the second half. The session chairs here are actually pretty lax - at NSDI they really cracked the whip and kept people on schedule (which I appreciated!).Ian Rosehttps://www.blogger.com/profile/05069842326920629876noreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-3067242244075121912010-11-04T04:06:05.380-07:002010-11-04T04:06:05.380-07:00"An excellent talk is still excellent even if..."An excellent talk is still excellent even if it is backed by black-text-on-white-background slides with figures and tables pasted in."<br /><br />I don't think this comment contradicts Matt's "no sense of design" point. Heck, black-and-white IS a good design! Its not particularly bold or original, but (imo) it still looks nice.Ian Rosehttps://www.blogger.com/profile/05069842326920629876noreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-14509927803812312912010-11-04T02:52:31.852-07:002010-11-04T02:52:31.852-07:00I'll soon be presenting a paper at a conferenc...I'll soon be presenting a paper at a conference for the first time. Your blog will surely help. <br />Thanks in advance.Devi Prasadnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-4840866691781171492010-11-03T18:58:52.613-07:002010-11-03T18:58:52.613-07:00In my experience, the use of an outline slide (wit...In my experience, the use of an outline slide (with its well-known structure) is often helpful for people who are feeling nervous. It allows them taking a slow(er) start...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-32111083069301753452010-11-03T18:58:28.687-07:002010-11-03T18:58:28.687-07:00To quote a (slightly cynical) former colleague: th...To quote a (slightly cynical) former colleague: the purpose of the talk is to get people to *cite* the paper.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-3829493757576979022010-11-03T18:25:17.062-07:002010-11-03T18:25:17.062-07:00And, the purpose of the talk is to get people to r...And, the purpose of the talk is to get people to read the paper.Thomasnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-20384234537414508892010-11-03T15:42:03.195-07:002010-11-03T15:42:03.195-07:00It could be worse... In some fields, the speaker s...It could be worse... In some fields, the speaker stands at the front of the room and reads the paper verbatim.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-34223712317619191922010-11-03T09:51:03.283-07:002010-11-03T09:51:03.283-07:00I think I understand the underlying point you are ...I think I understand the underlying point you are trying to make with three of these points, but I disagree with the conclusions and advice, or at least with where the emphasis was placed.<br /><br /><i>Reading the slides</i><br /><br />If the speaker does not verbally state the text in the slide during their speech, I have trouble "syncing" between the two sources of information, and it's as if I am trying to listen to two different speeches at once. I think that the purpose of text on slides is to remind listeners what is the most important couple of points of the slide in case they miss it in the explanation. Obviously reading <em>nothing</em> but the text on the slides is bad, but if one follows your advice on keeping the amount of text small, then reading nothing but the text would lead to a four-minute talk. I think it is perfectly acceptable to put four minutes worth of total text into the slides, read precisely that text as you get to it, and spend the rest of the time filling in the details, with the text up there to remind the audience what the main points are in case they doze off for 30 seconds. I don't think this is lazy or at all conflicts with the advice to rehearse the talk.<br /><br /><i>No sense of design</i><br />Certainly an innate sense of design can help improve a talk. But if you don't have a sense of design, you can still be a good scientist, do good work, give good talks, and have <em>decent</em> slides by keeping them simple. An excellent talk is still excellent even if it is backed by black-text-on-white-background slides with figures and tables pasted in. The fact that you can tell the difference between one Powerpoint feature, the clip art, and the other Powerpoint features that let you "design beautiful and visually compelling talks" means you have an innate sense of design that not everyone has. I think the presence of the Microsoft stick figures can be attributed to people with no sense of design, like me, caving in to pressure to add snazzy design to our spartan slides, but not having the graphic design talent to do this well.<br /><br /><i>No sense of humor</i><br />I've heard this one about teaching as well. Just like the design stuff, if you are a natural stand-up comedian, that's a great skill and you should use it in your career just like any other skill. But some people, like me, are no good with planned jokes. I can be funny in a group if it is spontaneous, but I have learned that whenever I plant a joke into a talk or a lecture, it bombs. The lesson I learned is, don't use humor if it doesn't work. I imagine the real point you wanted to make is, "Don't give a boring talk". But humor is not the only antidote for boredom. The best antidote for boredom is to give a talk that everyone in the room can follow. I only learned very slowly over the years how to do that, and I don't have any simple rules to follow to achieve that. But for me personally, I think that I get it done without any planned jokes.Dave Dotyhttp://www.dna.caltech.edu/~ddoty/noreply@blogger.comtag:blogger.com,1999:blog-9186457242428335144.post-38691902734964255772010-11-03T09:43:43.872-07:002010-11-03T09:43:43.872-07:00I was actually surprised to hear a senior research...I was actually surprised to hear a senior researcher advising to repeat everything you say on the slides, "because anyway people are not listening and only pay attention every 5' ". But this was here in France, and I think non-fluency of english makes that the speaker wants to ensure that the part of the audience that does not understand him gets the message.Matthias Galléhttp://www.irisa.fr/symbiose/matthias_gallenoreply@blogger.com