share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
SPEAKER 1: So it's a real pleasure to welcome Professor Michal Lipson back to Cornell. Those of you who need an introduction, Michal was a professor at Cornell. And one of the most distinguished professors at Cornell. She recently moved to Columbia.
SPEAKER 2: Do speak louder.
SPEAKER 1: And if I had to choose one person to give this writing seminar, I could not have chosen a better person than Professor Lipson. So to give you guys an idea, just between 2011 and right now, her written work, her publications, have over 15,000 citations. That's about more than any person I know, if I have to just think off the top of my head. And she has over 200 general publications. And about two years ago, she was rated as among the top 1% of the most highly cited physicists in the world.
So that's all these things, are testimony to the quality of scholarship that Michal has produced over the years. And that quality of scholarship just doesn't come by chance. It is something that one has to consciously make an effort, to be able to write clearly, to be able to write in a way that inspires, excites, people's imagination. As well as give them the information that they want to need.
So Michal has all the secrets that you would like to hear. And so would I. And it's a real pleasure that she is the one who is giving this talk right now. And I personally cannot think of a better person to give this talk. Personally, I've always found her scholarship to be extremely well-written and exciting. And it's a real treat to have her here to give this seminar. So please welcome her.
[APPLAUSE]
MICHAL LIPSON: Thank you very, very much. You put a lot of pressure on me, [INAUDIBLE]. I'm going to do my best here. And I'm really honored for all of you to come and listen to this talk.
So I'm going to start-- and by the way, those of you that are in the back, if you want, it's not a very formal talk. You guys can, if you want, you can sit on the side here. Just come here behind me and sit on the-- because it's going to be difficult for you guys to stand the whole talk. So feel free to come if you want.
So the goal of this workshop-- this mini workshop. 40 minutes workshop. Is for me to give you some very specific, pragmatic recipes on how to grab the attention of the reader. Your goal, when you write, is not to talk to the 10 people that do exactly what you do. They will read your paper no matter what. Don't worry about it. Your goal is to reach the much broader community. And it's not about a sexy title. It's not about the grabbing abstract. It's about you being able to write a paper. Someone grabs it, looks at it, one minute, they know what you did. One minute. And I'll give you the recipe on how to try to do that.
OK? Now let me tell you. Why did I come up with this one minute? So, in my field, in optics, we have CLEO, which is the most important conference for us. It's the conference that is the hardest to get in. Students work year round for this three page abstract. Every word there, they write.
And I sit on these committees. Typically, in the committee that I am in, there about 200 papers. There are 10 judges. 200 papers that need to be judged by 10 people. All of those 200 papers are read by each one of those judges. Now, how long do you think we dedicate for this? Realistically. Two, three hours max. That means one or two minutes max for you to tell your story in the whole paper.
There's also a very interesting story in the optics field. I'm not going to mention names. But there are very, very well known physicists, that a few years ago, both of them, by chance, wrote about the same topic in the Science magazine. And it was in the same issue of the Science magazine. And after that, it was a theory-- after that, the whole community started following up these works. And they basically were saying, in both of these papers, the same thing. Only one of them was cited and credited for that. Today, if you ask someone, who did this and that, it's going to be only one of them. Now, if you open that issue, and you read those two papers, you will understand in a second why. One of those papers was easy, crisp, clear, to read. The other was much harder. And they said the same thing.
So you might think, OK. What would have happened if the one that wrote better, and crisp, clearer, wouldn't have written that paper. Maybe that field wouldn't have happened at all. The guy wrote exactly the same thing. And maybe nothing would have happened. So, your goal as a scientist is to make sure that you grab their attention, and tell your whole story, in one minute.
And these are the people you have to deal with. OK. I don't know if you can see there. It says, most honorable Professor Klein, though I'm not worthy of your attention, it is with utmost respect and humble humility-- she said delete. This is it. That's the way. You have one minute to grab their attention.
OK. So all of this is going to be online. And I use these a lot. So you're going to have those references. So I go by golden rules. And my students have about 10 golden rules. I narrowed it down for you, the most important. Four golden rules we're going to cover here. And they're very pragmatic recipes.
The first one is to cut, reduce, any unnecessary word. OK. And I'm going to give you a few examples. And you might think oh, you know that. No you don't. Cut the word many. Cut the word-- anything subjective. Cut it.
So the way to do it, to really cut it efficiently, is to first think, and sometimes even think aloud, what you want to say. And then write it. And I should tell you that this will come again and again in this talks. Often, when I sit with my students over a paper-- and this is after they know all the golden rules and everything, the way we write. And I say to them, and I ask, OK. I know it very well and I'm reading the paragraph and I don't grab it in 10 seconds. That's not good.
So I ask he or she, what do you want to say here? And they tell me. And I understand in two seconds. I understand. And then what do I tell them? I tell them, wait. Let's write word for word, what did they just tell me?
So for some reason, when we actually say it, we say it much more succinct and much more clearly. And all of us. So I'm going to go through several examples. And please be active and think about how to make it better. Otherwise you won't get out from this talk what I wanted you to get out.
So this is an example. So this is a real example. So from someone very famous. So studies on the bending losses, for a variety of materials, will provide the fundamental information for optimizing the bending radius and the ratio between straight sections versus bent sections for future miniaturized gas sensor design.
We could do better. You want me to give you 10 seconds to think how to do it better? Try to think, how could you do that better. There's nothing wrong with this sentence. It's just definitely not crisp clear.
So this is an example of what you could say. Basically, you're not cutting any of the contents. The optimization of the radius and length of bent sections will enable future miniaturized gas sensors. And you're not cutting anything. You read this in a second. You get it.
OK. Let's go through several examples. This letter is to state that this field of research-- I got a lot of those-- should be regarded as non-fundable. So you tell me. Come on. This one is easy.
AUDIENCE: This research is non-fundable.
MICHAL LIPSON: Yeah. What else? OK. Field, or research. That's fine. Yeah. OK.
OK. Another one. The expected prevalence of mental retardation, based on the assumption of a normal distribution of intelligence in the population, is stated to be theoretically about 2.5%.
OK. You guys tell me. I mean, you can't do worse than that. So whatever. It's not very bad, but still.
OK. Next one you guys tell me, OK.
OK. The expected prevalence of mental retardation, if intelligence is normally distributed, is 2.5%. The way-- now, I should tell you, that when I give these exercises, it's a little hard to actually look at the sentence and think, OK, what am I going to cut, or not. Welcome to the world of faculty, right, that get a paper and say, ah, how do I make this paper more crisp, clearer. It's hard.
It's much better, before you write, to actually think clearly what you want to say, and then write. And if it doesn't work, fine. Try again. Say what you want to transmit and then write.
After you write, try to hunt down every word that is not needed. And the way to do it is to try it out without. Without that word. And see how it sounds.
Actually, interestingly, some words that is supposed to make it very-- supposed to make your point even stronger, without it, the sentence becomes even stronger. So for example, if you say, this is very important. If you say this is important, it might be even stronger. You might make your point stronger. I wouldn't say this is important ever. Nothing subjective goes in the paper. But this was just an example.
For example, very, really, quite, basically, generally. All of those, remove. I mean, the way to do it-- everything I'm saying here-- try without it. If you are in doubt, remove it.
There are-- usually, not necessary. Not necessary at all. For example, there are many researchers who are good teachers. How would you-- what would you do here?
AUDIENCE: [INTERPOSING VOICES]
MICHAL LIPSON: Yeah. Researchers, good teachers. I would actually remove the 'many.' Researchers are already many.
For practical application as a portable sensing device, it is required that the waveguide is coiled. We say that a lot, right. This typical sentence. You actually don't need that. For practical application as a portable sensing, the waveguide is coiled.
Now, this might not seem very important. All these removing words. But you'll see that when you do that, your paper becomes so much easier to read and to skim. Right, the whole goal is just to skim.
OK. Come on, unnecessary phrases. I see that all the time in my field, and in physics in general. For the most part, for the purpose, in case of-- usually, you can remove. In the final analysis, in the event that-- there is a lot of us that, especially when you write, you want to always protect yourself.
So for example, there is a lot of, as far as we know. No. That doesn't add anything. Of course it's as far as you know. If not, you wouldn't write it. So don't do that, OK?
It has been estimated that. It may be argued that. No. That doesn't help. If you are saying, you're saying. Take ownership. Don't worry about it. Don't try to put yourself in [INAUDIBLE]. Because all these sentences, what they do, you're trying to kind of, OK. You're addressing the aggressive referee. That won't help. And it just makes the paper more cumbersome.
In the case of measurements, the laser was rotated by 90 degrees clockwise. How would you say that?
AUDIENCE: For measurements, the laser was rotated by 90 degrees clockwise.
MICHAL LIPSON: Right. Or, I mean, it is a measurement, right, so.
AUDIENCE: The laser was rotated by 90 degrees clockwise.
MICHAL LIPSON: Right. So, the laser rotated by 90 degrees.
You can see here that I took all this-- this is from my field, right. But I wanted to give you concrete examples. So these are all from very, very well known scientists. A direct-- and several, I'm sure, from my own, also. A direct comparison study between conventional and photonic bandgap waveguides has not been reported to date for the purpose of optimizing gas sensors based on these devices.
So the person is saying, OK. Maybe people did a lot of work, but they did not report these comparisons for sensing. Maybe they reported other comparisons, but not for sensing. So how would you-- Chris, that's relevant to you. So, right? So how would you say that sentence?
You know what? Let's do that. Because if not, if you just sit passive, and I keep giving you examples, guys, you won't get anything out of it. Let's take one minute. Turn to your neighbor, and explain. One of you explain what this-- forget about the sentence. Explain what this sentence wants to say. And then give me an answer. OK. Let's take three answers at random. OK. One minute. Turn to your neighbor.
AUDIENCE: [DISCUSSION]
MICHAL LIPSON: OK. In the back. I need a courageous soul in the back. You can't do worse than that, so whatever.
Anyone like, half to the back. Yes.
AUDIENCE: A direct study between the two waveguides has not been reported for optimizing gas sensors.
MICHAL LIPSON: Yeah. That's actually already much better. Good. Anyone else? OK, the front. Yes, yes, yes.
AUDIENCE: Conventional and photonic bandgap waveguides have not been compared for optimizing gas sensors.
MICHAL LIPSON: Much better. OK. One more. OK.
So I actually don't have a concrete answer, because all of those are much better. Phrases that should be removed. A majority of, a number of, are of the same opinion, at the present moment, by means of, less frequently occurring. OK. Instead of that, what would you say instead of a majority of?
AUDIENCE: Most of.
MICHAL LIPSON: Right. Instead of a number of, many. Are of the same opinion, agree. OK. So we all know that. But it's important-- don't fall on this trap. Any clunky sentence you want to say, put it together.
OK. More wordy sentences. Instead of in spite of the fact that, you say although. In the event that, if. And so on. You're going to have these examples online. But these are just examples. There are just so many of those. So I would say, every sentence, after you write it. Every sentence, think about it. Is there a way for me to reduce, or remove-- remove. Take a red pen, be courageous.
I've made my point. You always want to ask yourself, is it really necessary? It might add a little bit of a different tone, but we are not in English class here. Our point is to be just crisp, clear. So even if it maybe changes the tone, forget about it. If it's not critical, remove it.
What happens if I take it out? OK. If it didn't really change anything, remove. Just a little bit more and I'll move to the second point. Brain injury incident-- that's a good one-- shows two peak periods in almost all reports. Rates are the highest in young people and the elderly.
So this sentence is a good one because it's hard to read. It's a hard sentence to read. I remember reading it twice and saying, what are they saying?
AUDIENCE: Highest in young people and elderly.
MICHAL LIPSON: Right. OK. So, brain injury incidence peaks in the young and in the elderly. Or peaks in the young and in the elderly. OK. And just that.
So you got the first point. And perhaps, this is one of the most important points. The second point everybody tells you-- and it is usually, whenever I say that, there is controversy-- remove the controversial. Just use active voice. And that's because there are situations where you want to use, where you could use passive voice. But in doubt, here, we just want to do the golden rules. We just want to not think and do a crisp, clear paper. So just use active voice. You can't go wrong with that.
The catch is that most of us feel more comfortable using passive voice. And when we do that, two things happen. The sentence usually becomes more cumbersome. And second, it is not clear who actually did that. Who actually did the experiment. It just sets a bad tone. And I'll give you an example.
Samples were fabricated. So who fabricated the samples?
Mistakes were made. That's a great one, right. It's OK. Nobody is responsible. Oh, yeah.
So this is an example of a passive voice. Major differences in response time of the two samples were found. So you want to use active voice. Who found these major differences in responses? You want to say, we found. This sentence is a little-- and I'm putting this, I don't want to confuse you, but this is important, so that you understand the big picture.
When we use active voice, we want to transmit the message that we did it. We did this, we did that. Here, the main point-- so what's the main point of the sentence? The main point of the sentence is the response time. It's not the fact that we did it. So this could remain in passive voice. And that would be OK.
But in general, if you are in doubt, you don't know, forget about it. Just say we found. OK. So I honestly always use just active.
So this is an example when it is appropriate to use, but I would just actually skip that, because I don't want to confuse you.
So we went through two golden rules. Cut everything. Use active voice, take ownership. Use parallel constructions. So that's actually a tool that makes the paper much easier to understand. I would say, if you don't use this, the paper becomes very cumbersome, almost always. And I'll give you examples. And with this, you really have to think. What's parallel here? So, for example.
Aluminum is most quickly removed using a nitric and hydrochloric acid etching solution while sulfuric acid and hydrogen peroxide solution remove photoresist. OK. I just read it, and I'm saying, what did I just say? It's not easy to understand.
So, and that's because it is unparallel. So let's see. We want to say aluminum is most quickly removed using nitric. OK. How do I start the second part of the sentence?
AUDIENCE: [INAUDIBLE]
MICHAL LIPSON: Exactly. I start with photoresist, the second. So I say aluminum removed with this. Photoresist is removed with that. We make this mistake a lot.
And you should understand that, even standard, in general writing, in English writing, this might not be so critical. But here, remember, our goal is to make the reader skim and understand. And that makes it much easier.
If you have a list of ideas, they should be written in a parallel form. For example, if you want to be a good doctor, you must study hard, critically think about the medical literature, and you should be a good listener. Ahh. How do we make that better?
Study hard. Critically think about, OK. The medical. And?
AUDIENCE: Be a good listener.
MICHAL LIPSON: Be a good listener. Yeah.
Not parallel. Another example. Sorry. Let me just-- OK. And you must study hard, listen well-- this is even better, OK-- and think critically. Or, just like we said-- good student, good listener, critical thinker.
One more. This research follows four distinct phases: establishing measurement instruments, pattern measurements, developing interventions, and the dissemination of successful interventions to other settings and institutions. It's actually-- you'll have a blast when you finish this talk. Go open papers of very well-known people. And you'll see all of this. So how do we make that better?
So the problem here, the reason why it's not parallel-- it's not what we are saying. It's the verb. So we're saying establishing in the first. Second, we say pattern measurements. Right? And three, we go back to developing. And four, we are completely off. We're saying, the dissemination of successful. OK. So choose what you want to say. Choose the way you want to say it. Either what is being done, or how it's being done. And be consistent.
Establishing. Measuring. Developing. Dissemination.
Now. We went through three golden rules. And I'm going to go to my favorite. And this is the one. If you remember anything about all of this, this is what makes the paper crisp, clear.
My own students know that if they don't do this, I take one second to look at the paper. Go back and do this. And I should say, this is all relatively subjective. This is the way I do it. And these are my golden rules, OK. But if you do this, your paper will be crisp, clear.
Every paragraph should start with the main point of the whole paragraph. So much that the rest of the paragraph is just details. So this is a little different. Because your main point, often, is at the end of the first paragraph of the introduction. Or, it could be, often, at the second or third sentence.
For example, photonics is important. First sentence. Second sentence, nothing has been done on modulators. Third, we demonstrate modulators. Oh my God. You said the most important thing at the third sentence in the paragraph. Someone skimming it would not see that. And by the way, I would definitely cut photonics is important.
So this is so-- if you do it really good, this is so obvious, that if you take just the first sentence of each paragraph, and put it together, it will read like an abstract. So what I did, before this talk, yesterday evening, I went to-- I looked for people that write very well. And are extremely highly cited. And they obey, I would say, 90% of this. So if you take the first sentence of each paragraph, maybe one paragraph is not needed. Maybe one-- the beginning is not needed. The first sentence. Otherwise, you can really understand what they did.
And that's again, because you skim through the paper. What you see- you see the figures, and you see the first sentence of the paragraph. That's what your brain sees. There's a very strong temptation of putting your main point at the end. Don't do that.
If you have a lot of details, for example, in the fab. Right, you fabricate, you make. You chemically mix this and that. And you have to put it in the paper. But it's not very interesting. So the way to do it-- think about, what was the general idea of the fab? Is there anything different than anyone else did? Probably. Otherwise you wouldn't give all the details, right? Say that in the first sentence. And then hide all those details in the middle of the paragraph. So only those that are really close to your field will read it.
So what I'm going to do now-- bear with me-- I am going to show you a paper published this year, actually, by this fantastic scientist-- and if you can't read it from the back, I will read loud, so no worries.
Published in [INAUDIBLE] Technology by Marin Soljacic who is just a fantastic scientist. And they do exactly that. So let's see this example. The reason why I chose that person, Marin Soljacic, is that they come from the same mentorship that I had. So it's not a coincidence.
This all comes from the tree from Professor John Joannopoulous at MIT. So [INAUDIBLE] if you're looking for papers that are very well written, in that sense, meaning first paragraph, and so on-- [INAUDIBLE], John Joannopoulous, Marin Soljacic, they all write in this way. And we will put this online, this paper, as well.
One example, if you want to take-- this is like a mini example, mini paper-- showing this point. OK. I wrote a fictitious paper, where the paragraphs are very, very short. But just for you to get a feeling. So let's forget about everything. Let's just take the first sentence of each paragraph. We designed and fabricated a ring resonator using CMOS compatible fabrication method. OK. We tested the structure using ultra fast pulses. The structure demonstrated high switching performance. Applications include telecom-- sorry, there's another, too much application-- requiring small sized components. That's it. I gave my abstract. And everything else are details.
The facts that are the most important need to go up front, is also for individual sentences. I did not include that in the golden rule, but in the extended version of the golden rule, the 10 golden rules, it's there. So if you have a sentence and you don't know what order to write it, always take the most important thing that you want to say, up front.
OK. I'm going to-- let me ask you if you guys need that. Do you know what's the difference between introduction, right, and discussion? Because that's often a question I get. Do you guys know?
That actually does not have too much to do with crisp, clear. This is in general an issue that people have. I'm just going to say that very briefly. The introduction, you start with what's available. And you see what's missing is your work. In the discussion, you say-- so you start from big and you go small, right, in the introduction.
From, at the end, in a discussion, you say what you-- what the amazing thing that you did is going to revolutionize the world. So then you explain, you know, what's the implications, why what you did, somebody else would care-- and why, what, the deficiencies that you have, might or might not be fundamental. And that's actually pretty important. Make sure to put your work in the big picture, in the sense that, maybe your performance is not, you know, the best, but maybe it only has to do with a capacitance or something that is very minor. Put it there in the discussion.
We are already late. Maybe, let me just say, the most important things not to do in the discussions. We often just take the beginning, take the introduction, or the results, and repeat it in the discussion. The discussion should be just the big picture. And often, if I'm very busy, I would just read the discussion.
So I don't care about the number. I care about the big picture. So the results, you show the number. The discussion, you show the big picture.
Another mistake that we often do, we go into the details of explaining why this happened, or why that happened. All these details should be embedded in a paragraph. Don't worry about that. The discussion should be the big picture. Not the little details.
Definitely, you want to make your paper readable. You want to attract a very wide audience. But definitely don't overclaim. Don't exaggerate.
With the student population, you should know that in general, the problem is the opposite. We usually don't have the problem of over-exaggerating, overclaiming. Usually we have the opposite problem, of under claiming, or being a little too specific, or too careful. Make sure to be in the middle.
So I'm going to summarize. With the golden rules, you want to cut, reduce, hunt every word. And say it before you write it. The most effective way is to write, leave it for a few days, and then come back. You'll see tons of words that shouldn't be there, and sentences that shouldn't be there. I sometimes look at proposals I wrote, that I worked on every single sentence there. I thought it was an amazing proposal. And after two months, I read it, and I say, oh, my God. It's impossible to read. Why are there so many words there? And so on. So you can never, never edit too much.
Use active voice. Don't be afraid. Use parallel constructions. Parallel construction is almost everywhere. Because we're always comparing whatever we did. And we usually do more than one thing. So parallel construction makes it much easier to read.
That's the key. Start the paragraph with what you want to say in the paragraph. And then check yourself. Take the beginning of each paragraph. Put it together. And see if it's an abstract.
And then I talked a little bit about the introduction versus discussion.
OK. I'm going to stop here. Thank you.
[APPLAUSE]
Michal Lipson, professor of electrical engineering at Columbia University, gives a tutorial on scientific writing April 8, 2016. She emphasizes clarity, omitting unnecessary words and phrases, parallel construction and putting important ideas first. She gives examples and challenges the audience to improve them. Sponsored by the Cornell Center for Materials Research (CCMR).
