📄 http:^^www.cs.wisc.edu^~markhill^conference-talk.html
字号:
Date: Tue, 05 Nov 1996 20:46:56 GMTServer: NCSA/1.5Content-type: text/htmlLast-modified: Mon, 20 May 1996 16:39:36 GMTContent-length: 6414<html><head><title>Oral Presentation Advice</title></head><body bgcolor="#ffffff" vlink="#0060f0" link="#FF3300"><font size=6><center><b>Oral Presentation Advice</b><p></font><font size=4>Mark D. Hill<p>Computer Sciences Department<br>University of Wisconsin-Madison<p>April 1992</font></center><p>Below are points to consider and an outline for a conference talk. Theoutline is a starting point, not a rigid template. Most goodspeakers average two minutes per slide (not counting title and outlineslides), and thus use about a dozen slides for a twenty minutepresentation.<ul><li> <!WA0><A HREF="#think">Things to Think About</A><li> <!WA1><A HREF="#outline">A Generic Outline</A><li> <!WA2><A HREF="#badtalk">How to Give a Bad Talk by David Patterson</A></ul><p><hr size=5><p><p><h3><A NAME="think">Things to Think About</A></h3><p><p><ol type=1><li><b>Oral Communication is different from written communication</b><p>Listeners have one chance to hear your talk and can't "re-read"when they are confused. Often they have or will hear many talkson the same day. There are two well-know ways to communicateyour point effectively. The first is to K.I.S.S. (keep it simplestupid). Second, repeat key insights: tell them what you'regoing to tell them (Forecast), tell them, and tell them what youtold them (Summary).<p><li><b>Think about your audience</b><p>Most audiences should be addressed in layers: some are expertsin your sub-area, some are experts in the general area, andothers know little or nothing. Who is most important to you?Can you still leave others with something? For example, pitchthe body to experts, but make the forecast and summaryaccessible to all.<p><li><b>Think about your rhetorical goals</b><p>I recommend two goals for conference talks: leave your audiencewith a clear picture of the gist of your contribution, and makethem want to read your paper. Your presentation should notreplace your paper, but rather whet the audience appetite forit. Thus, it is commonly useful to allude to information inthe paper that can't be covered adequately in the presentation.(The goals for an interview talk, for example, are radicallydifferent.)<p><li><b>Practice in public</b><p>It is hard distilling work down to 20 or 30 minutes.<p><li><b>Prepare</b><p>See David Patterson's <!WA3><a href="http://www.cs.wisc.edu/~markhill/bad-talk.html">How to Give a Bad Talk</a><p></ol><p><hr size=5><p><h3><A NAME="outline">A Generic Outline</A></h3><ul><li><b>Title/author/affiliation</b> (1 slide)<li><b>Forecast</b> (1 slide)<br>Give gist of problem attacked and insight found (What is theone idea you want people to leave with? This is the "abstract"of an oral presentation.)<li><b>Outline</b> (1 slide)<br>Give talk structure. Some speakers prefer to put this at thebottom of their title slide. (Audiences like predictability.)<li><b>Background</b><ul><li><b>Motivation and Problem Statement</b> (1-2 slides)<br>(Why should anyone care? Most researchers overestimatehow much the audience knows about the problem they areattacking.)<li><b>Related Work</b> (0-1 slides)<br>Cover superficially or omit; refer people to your paper.<li><b>Methods</b> (1 slide)<br>Cover quickly in short talks; refer people to your paper.</ul><li><b>Results</b> (4-6 slides)<br>Present key results and key insights. This is main body of thetalk. Its internal structure varies greatly as a function ofthe researcher's contribution. (Do not superficially cover allresults; cover key result well. Do not just present numbers;interpret them to give insights. Do not put up large tables ofnumbers.)<li><b>Summary</b> (1 slide)<li><b>Future Work</b> (0-1 slides)<br>Optionally give problems this research opens up.<li><b>Backup Slides</b> (0-3 slides)<br>Optionally have a few slides ready (not counted in your talktotal) to answer expected questions. (Likely question areas:ideas glossed over, shortcomings of methods or results, andfuture work.)</ul><H3>Acknowledgments</H3>Thanks to Jim Goodman, Jim Larus, and David Patterson for their usefulcomments.<p><hr size=5><p><font size=6><center><b><A NAME="badtalk">How to Give a Bad Talk</A></b><p></font><font size=4>David A. Patterson<p>Computer Science Division<br>University of California-Berkeley<p>Circa 1983</font></center><p><p>Ten commandments (with annotations gleaned from Patterson's talk byMark D. Hill):<ol type=I><li><b>Thou shalt not be neat</b><p>Why waste research time preparing slides? Ignore spelling,grammar and legibility. Who cares what 50 people think?<p><li><b>Thou shalt not waste space</b><p>Transparencies are expensive. If you can save five slides ineach of four talks per year, you save $7.00/year!<p><li><b>Thou shalt not covet brevity</b><p>Do you want to continue the stereotype that engineers can'twrite? Always use complete sentences, never just key words.If possible, use whole paragraphs and read every word.<p><li><b>Thou shalt cover thy naked slides</b><p>You need the suspense! Overlays are too flashy.<p><li><b>Thou shalt not write large</b><p>Be humble -- use a small font. Important people sit in front.Who cares about the riff-raff?<p><li><b>Thou shalt not use color</b><p>Flagrant use of color indicates uncareful research. It's alsounfair to emphasize some words over others.<p><li><b>Thou shalt not illustrate</b><p>Confucius says ``<i>A picture = 10K words</i>,'' but Dijkstra says``<i>Pictures are for weak minds.</i>'' Who are you going to believe?Wisdom from the ages or the person who first counted goto's?<p><li><b>Thou shalt not make eye contact</b><p>You should avert eyes to show respect. Blocking screen canalso add mystery.<p><li><b>Thou shalt not skip slides in a long talk</b><p>You prepared the slides; people came for your whole talk; sojust talk faster. Skip your summary and conclusions ifnecessary.<p><li><b>Thou shalt not practice</b><p>Why waste research time practicing a talk? It could takeseveral hours out of your two years of research. How can youappear spontaneous if you practice? If you do practice, arguewith any suggestions you get and make sure your talk is longerthan the time you have to present it.<p></ol><p>Commandment X is most important. <i>Even if you break the other nine,this one can save you.</i></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -