1
00:00:03,693 --> 00:00:06,373
I am going to speak about the act of writing,

2
00:00:06,502 --> 00:00:08,962
which is something every
researcher has to do a lot

3
00:00:09,621 --> 00:00:12,491
and the good news is
that some pretty simple things

4
00:00:12,579 --> 00:00:14,639
can make your
written output a lot better.

5
00:00:14,900 --> 00:00:16,930
So that's why I titled this:
Seven simple suggestions.

6
00:00:16,930 --> 00:00:19,380
So here we go.
Seven simple suggestions.

7
00:00:19,631 --> 00:00:22,271
Here's the first one:
Don't wait: WRITE.

8
00:00:22,271 --> 00:00:23,211
This is what mistake,

9
00:00:23,536 --> 00:00:24,606
what a lot of people make.

10
00:00:24,606 --> 00:00:26,756
Here is a typical plan for doing research

11
00:00:27,153 --> 00:00:27,783
Number one:

12
00:00:28,033 --> 00:00:29,083
Have a brilliant idea.

13
00:00:29,260 --> 00:00:29,860
Number two:

14
00:00:29,881 --> 00:00:32,571
Spend months doing good research to back it up

15
00:00:32,571 --> 00:00:33,121
Number three:

16
00:00:33,121 --> 00:00:33,851
Write the paper.

17
00:00:33,851 --> 00:00:36,361
In last two weeks before the deadline for the conference.

18
00:00:37,151 --> 00:00:38,401
This is a bad plan, right.

19
00:00:38,931 --> 00:00:41,021
Here is a good plan for doing research:

20
00:00:41,061 --> 00:00:42,281
First: have an idea.

21
00:00:42,592 --> 00:00:44,602
Second: start writing the paper.

22
00:00:44,882 --> 00:00:47,842
Third: use the paper as a forcing function

23
00:00:48,373 --> 00:00:51,493
to make you do the research that articulates the paper.

24
00:00:51,493 --> 00:00:51,893
Ok?

25
00:00:52,010 --> 00:00:53,370
See how diferent it is?

26
00:00:53,440 --> 00:00:53,940
Right?

27
00:00:53,973 --> 00:00:55,833
It means you start writing very early.

28
00:00:55,833 --> 00:00:57,163
Why is this good? Well.

29
00:01:00,471 --> 00:01:02,421
hacking around and writing a code and thinking

30
00:01:02,723 --> 00:01:04,223
and I start to write the paper,

31
00:01:04,223 --> 00:01:06,243
than I relise that some of my previous work

32
00:01:06,294 --> 00:01:07,724
was misdirected, right?

33
00:01:07,844 --> 00:01:09,644
It wasn't able to do useful goal,

34
00:01:09,655 --> 00:01:10,575
certainly not for this paper

35
00:01:10,595 --> 00:01:11,745
and I also discover

36
00:01:11,868 --> 00:01:14,698
some key parts of the paper that need more work,

37
00:01:14,698 --> 00:01:15,138
right.

38
00:01:15,138 --> 00:01:19,678
So you wanna use the paper as a forcing function to learn that early.

39
00:01:20,129 --> 00:01:23,639
It also gives you a good way to communicate with another people

40
00:01:23,639 --> 00:01:26,061
Research is all about communication. Right?

41
00:01:26,061 --> 00:01:29,621
If you...Maybe you are kind of person who likes work just on their own

42
00:01:29,640 --> 00:01:31,920
in a windowless room with no lights.

43
00:01:31,920 --> 00:01:33,770
But I like to work with other people

44
00:01:33,770 --> 00:01:35,700
I spent a lot of time at the white board

45
00:01:35,700 --> 00:01:39,700
If you have something written you have a new mechanism for communicating with them

46
00:01:40,428 --> 00:01:40,958
Right?

47
00:01:40,958 --> 00:01:43,978
Is very important is a mechanism for communicating

48
00:01:44,072 --> 00:01:45,572
That's all along way saying

49
00:01:45,782 --> 00:01:49,502
that really writing is not the way in which we just report research

50
00:01:50,467 --> 00:01:54,467
for me is a way which I do research

51
00:01:54,681 --> 00:01:57,141
and I think you should think about writing in that way.

52
00:01:57,331 --> 00:02:00,231
It's not just the output, it's a computation

53
00:02:00,321 --> 00:02:02,731
It's the stuff that makes research happen

54
00:02:02,789 --> 00:02:08,099
And it's because somehow we think more clearly when we write, than when we just think.

55
00:02:08,414 --> 00:02:10,134
At least I do.

56
00:02:10,270 --> 00:02:10,950
So.

57
00:02:10,950 --> 00:02:15,590
Ehm. One problem with this plan here is that it starts with this idea bit.

58
00:02:15,590 --> 00:02:16,000
Right?

59
00:02:16,000 --> 00:02:17,670
Where this idea come from?

60
00:02:17,740 --> 00:02:21,580
And so..ehm...it's tempting when you looking at another peoples' work

61
00:02:21,580 --> 00:02:27,790
to think "oh my" everybody else has very clever ideas and I am a mere worm

62
00:02:27,849 --> 00:02:31,849
and I have such a trivial boring ideas that nobody would be interested in them

63
00:02:31,901 --> 00:02:33,191
And do you know that feeling? Yeh?

64
00:02:33,191 --> 00:02:35,541
You sit there and you don't feel very creative

65
00:02:35,541 --> 00:02:39,541
Let me tell you that this is what every researcher feels most of the time

66
00:02:40,226 --> 00:02:42,726
There are days when you have miraculous breakthrough

67
00:02:42,726 --> 00:02:45,836
and you really know your God is in heaven and you know that all is right with the World.

68
00:02:45,836 --> 00:02:50,966
But most of days you thing I am worm this is the natural state of the researcher.

69
00:02:50,966 --> 00:02:51,466
Right?

70
00:02:51,466 --> 00:02:56,616
So what you have to release is that even you know [Tarjan], and so forth, thinks that he's a worm most of the time

71
00:02:56,616 --> 00:02:58,816
and what what the researcher do? Good researchers

72
00:02:58,816 --> 00:03:01,586
is they simply start writing anyway, right?

73
00:03:01,586 --> 00:03:07,056
They write a paper about any idea no matter how trivial or insignificant it may be.

74
00:03:08,036 --> 00:03:14,316
My experiences that the clever the research student then more they prone to this failure mode

75
00:03:14,316 --> 00:03:17,246
either they don't understand something which case they are depressed

76
00:03:17,246 --> 00:03:22,626
well stay do undestand something which they do believe, in which case they think it is trivial and nobody would want to know about it, all right?

77
00:03:22,882 --> 00:03:25,422
So in both cases depressed.

78
00:03:25,422 --> 00:03:27,592
This is not a good situation, right?

79
00:03:27,653 --> 00:03:30,793
So just write an idea no matter how insignificant it seems to be

80
00:03:30,793 --> 00:03:36,243
because my experience consistently is that when you write the paper

81
00:03:36,243 --> 00:03:42,013
your idea develops and ramifies computer science is like a snowflake or flower

82
00:03:42,100 --> 00:03:46,100
you start with a little seed and its ramifies ahead of you in an interesting things

83
00:03:46,507 --> 00:03:50,507
something looked boring turns out to be actually rather interesting

84
00:03:50,656 --> 00:03:55,956
not always sometimes you start writing the paper and it turns out indeed to be weedy and insignificant

85
00:03:55,956 --> 00:04:00,759
and maybe when you published it and it will be done quickly and you put it on your homepage, all right?

86
00:04:01,542 --> 00:04:05,422
So, write early!

87
00:04:05,422 --> 00:04:10,452
And, and and I really do believe that the second thing is much the most a common case.

88
00:04:10,472 --> 00:04:14,112
It almost invariably turns out to be more interesting than you thought.

89
00:04:14,112 --> 00:04:19,362
All right, number two if you gonna have this idea and wrote about it you need to be clear what it is.

90
00:04:19,362 --> 00:04:27,222
The business of writing a paper is to convey from your brain into the minds of your readers your idea.

91
00:04:27,222 --> 00:04:30,352
So think of your papers like, like a virus, right?

92
00:04:30,352 --> 00:04:34,352
You trying to, you try to infect your readers' minds with your idea.

93
00:04:34,982 --> 00:04:38,172
And then you will infect them and then they talk to another people and they will infect them

94
00:04:38,172 --> 00:04:42,642
So it's like a kind of contagious play which is going to sweep the world

95
00:04:42,642 --> 00:04:46,642
And everybody is been thinking about your idea because it is so well infectious.

96
00:04:47,327 --> 00:04:47,697
Right?

97
00:04:47,810 --> 00:04:52,880
So...ehhmm....my, my, my knowledge here with, with Mozart

98
00:04:52,880 --> 00:04:58,190
It's just houndred of years after the Mozart died we are listening the people read his papers.

99
00:04:58,297 --> 00:05:00,237
Or more precisely play his music.

100
00:05:00,237 --> 00:05:02,827
Where you go to concert hall to hear his papers because

101
00:05:02,827 --> 00:05:04,667
their ideas were so infectious

102
00:05:04,667 --> 00:05:06,337
Don't you think that's amazing?

103
00:05:06,337 --> 00:05:10,467
Won't it be remarkable that in four hundred years time people will still reading you papers.

104
00:05:10,889 --> 00:05:11,929
That's not all likely

105
00:05:11,929 --> 00:05:15,219
But I think that's the..,that's a kind of idea like to get

106
00:05:15,219 --> 00:05:18,429
your your..the papers is not mechanism for getting promotion

107
00:05:18,429 --> 00:05:22,429
It's mechanism for convey ideas from your head to somebody elses head

108
00:05:22,553 --> 00:05:26,553
If you don't convey them, right, if don't bother to tell anybody about your ideas,

109
00:05:26,553 --> 00:05:28,633
then your minds don't have them

110
00:05:29,311 --> 00:05:29,791
Right?

111
00:05:29,868 --> 00:05:33,868
Even if Einstein had sat in the window this box not tell anybody about relativity.

112
00:05:33,933 --> 00:05:36,009
Then we wouldn't know relativity, right?

113
00:05:36,009 --> 00:05:41,559
That means you need to know what your idea is. Right? So when you write a paper in the end at least by the time you finished.

114
00:05:41,559 --> 00:05:45,559
You must know what is the idea that your paper conveys.

115
00:05:45,690 --> 00:05:49,220
It's surprising how hard that could be to determine from people's papers

116
00:05:49,220 --> 00:05:52,680
As a viewer if you read somebody's paper when you finish you can say:

117
00:05:52,680 --> 00:05:54,510
"What idea did that paper convey?"

118
00:05:55,549 --> 00:06:03,763
If you find that hard to articulate as a reader, then you know, in your view, you can say I really couldn't figure out what this what the idea this paper was about.

119
00:06:03,763 --> 00:06:06,613
And so than you apply to your own papers, right?

120
00:06:06,613 --> 00:06:10,180
It's actually sometimes quite hard to know exactly what your idea is to begin with.

121
00:06:10,180 --> 00:06:12,110
But you must now by the time you finish.

122
00:06:12,110 --> 00:06:16,140
Eeehrrr...if you find that yourself thinking, ohh..I have actually three ideas in this paper.

123
00:06:16,140 --> 00:06:18,830
Than what you do? Just write three papers.

124
00:06:18,830 --> 00:06:19,350
Right?

125
00:06:19,350 --> 00:06:26,570
It's wrong to try to merge them all in. In which case each one becomes cryptic and incomprehensible. Which you try to squidge into ten pages.

126
00:06:27,088 --> 00:06:29,508
Just write three papers, that's cool, right?

127
00:06:29,661 --> 00:06:33,511
That's not salami slicing, that is taking ideas and expressing them.

128
00:06:33,511 --> 00:06:38,461
A good idea in your paper is to say explicitly when you get to main idea

129
00:06:38,461 --> 00:06:40,801
It's suprising how seldom is this happens.

130
00:06:40,801 --> 00:06:47,941
So I often try to write phrases in my paper that say the beginning of section three somewhere the main idea of this paper is this..

131
00:06:48,068 --> 00:06:51,559
Because I have to explain some context and set up and background.

132
00:06:51,559 --> 00:06:57,339
And I wanted to be absolutly clear when I move from describing the context saying: here is the payload.

133
00:06:57,904 --> 00:06:58,314
Right?

134
00:06:58,598 --> 00:07:02,598
The viruses is about to arriving in your brain. Prepare!

135
00:07:02,714 --> 00:07:04,304
Does it make sence? All right?

136
00:07:04,377 --> 00:07:08,377
If you don't say that, if you leave you reader to be a detective.

137
00:07:08,606 --> 00:07:13,996
So, subsequently they have to reverse engineer what they think you meant was the key idea. That's not good, right?

138
00:07:13,996 --> 00:07:17,996
Why not said it explicitly, be completely up front about this.


139
00:07:18,806 --> 00:07:20,326
Third thing: Tell a story!

140
00:07:20,326 --> 00:07:23,572
So if you know write a paper this applies I am focusing mainly on papers.

141
00:07:23,572 --> 00:07:27,512
But I do also mean dissertations and so forth. Everything applies everything you like really.

142
00:07:27,512 --> 00:07:29,282
Is tell some kind of a story

143
00:07:29,282 --> 00:07:34,692
I always try to imagine when I'm writing a paper, that I am standing a front of the white board and explain it to a colleague.

144
00:07:35,167 --> 00:07:40,257
It's amazing how differently people present things at a white board and they will in a paper.

145
00:07:40,257 --> 00:07:40,647
Right?

146
00:07:40,674 --> 00:07:47,464
Without white board they start away ..??..with examples and explaining quite diffrently than if they..if they will not be out there..you you you get the idea..

147
00:07:47,464 --> 00:07:52,524
They often explain in much more accessible and engaging way at the white board than in the paper.

148
00:07:52,524 --> 00:07:58,524
So if you want to by accessible and engaging ......... do the same in the papers you do at the white board. Nearly.

149
00:07:58,524 --> 00:08:00,094
You need a little have a bit more substance.

150
00:08:00,094 --> 00:08:03,591
So here is a narrative flow that I usually try to follow with my papres. Right?

151
00:08:03,591 --> 00:08:08,171
You want to say here what a problem is, you want to motive your readers to say way it is an interesting problem

152
00:08:08,171 --> 00:08:14,441
You want to say at least briefly why it is unsolved problem, so be aware of solving, it is unsolved.

153
00:08:14,454 --> 00:08:19,075
And then you want to present your idea, that's the payload, right now

154
00:08:19,075 --> 00:08:19,495
Right?

155
00:08:19,495 --> 00:08:21,345
And then you give quite a bit of detail

156
00:08:21,345 --> 00:08:24,455
about exactly how what your idea is and how it works

157
00:08:24,455 --> 00:08:26,877
Ehm... Then you want to say something about

158
00:08:26,877 --> 00:08:28,653
how your idea compares with other people.

159
00:08:28,653 --> 00:08:30,659
See, you are kind of trying to lead people in

160
00:08:30,659 --> 00:08:32,419
You know, no everybody will read

161
00:08:32,419 --> 00:08:34,258
right the way through your paper.

162
00:08:34,258 --> 00:08:35,278
Your ideal is

163
00:08:35,278 --> 00:08:37,625
That whereever anybody stops reading

164
00:08:37,625 --> 00:08:40,021
they take away something valuable with them.

165
00:08:40,421 --> 00:08:43,031
And moreover every bit they read

166
00:08:43,031 --> 00:08:44,971
makes them want to read more, right?

167
00:08:44,971 --> 00:08:47,341
This is back to make it accesible, right?

168
00:08:47,341 --> 00:08:48,612
So, you know they say:

169
00:08:48,612 --> 00:08:50,132
Ah, here is an interesting problem

170
00:08:50,132 --> 00:08:51,122
I wish I could solve that one!

171
00:08:51,122 --> 00:08:52,602
I wonder if he can solve it!

172
00:08:52,602 --> 00:08:54,012
So, you know - they got a hook!

173
00:08:54,012 --> 00:08:55,712
They incline to read some more.

174
00:08:55,712 --> 00:08:59,152
Ehm, here is my - so, the typical outline

175
00:08:59,152 --> 00:09:00,272
for most papers that I write

176
00:09:00,272 --> 00:09:02,392
and indeed it works for dissertations as well.

177
00:09:02,542 --> 00:09:03,543
A short abstract.

178
00:09:04,514 --> 00:09:05,492
An introduction.

179
00:09:05,492 --> 00:09:06,882
Then something about the...

180
00:09:06,882 --> 00:09:08,762
stating what the problem is.

181
00:09:08,762 --> 00:09:10,142
Then a quite a bit of

182
00:09:10,142 --> 00:09:12,402
a short piece explaining what the idea is

183
00:09:12,402 --> 00:09:13,372
kind of intuitively

184
00:09:13,372 --> 00:09:16,236
and then a longer piece explaining the details

185
00:09:16,236 --> 00:09:17,156
behind the idea

186
00:09:17,156 --> 00:09:18,746
that sort of fills out the evidence.

187
00:09:18,746 --> 00:09:20,056
Then something about relaive work.

188
00:09:20,056 --> 00:09:21,016
And I am going to say a bit more about

189
00:09:21,016 --> 00:09:22,485
each of these sections

190
00:09:22,485 --> 00:09:23,405
in the following piece

191
00:09:23,405 --> 00:09:24,725
but this is my picture of

192
00:09:24,725 --> 00:09:26,945
how the structure of a paper might go.

193
00:09:26,945 --> 00:09:28,385
And look at the numbers of readers, right?

194
00:09:28,385 --> 00:09:30,505
More people read your abstract and title

195
00:09:30,505 --> 00:09:31,795
that will read all the rest of the paper

196
00:09:31,795 --> 00:09:33,795
so each time you want to give them a hook

197
00:09:33,795 --> 00:09:34,369
to continue.

198
00:09:34,369 --> 00:09:34,939
Here is what I think

199
00:09:34,939 --> 00:09:35,949
your introduction should do.

200
00:09:35,949 --> 00:09:36,726
One page!

201
00:09:36,726 --> 00:09:38,656
Describe the problem briefly

202
00:09:38,656 --> 00:09:41,099
and articulate what your contributions are.

203
00:09:41,415 --> 00:09:43,255
So, here is some by way of example

204
00:09:43,255 --> 00:09:44,853
Ehm, describing the problem

205
00:09:44,853 --> 00:09:46,933
I would suggest you introduce your problem

206
00:09:46,933 --> 00:09:47,873
with an example.

207
00:09:48,183 --> 00:09:49,993
So, my good way of doing to say:

208
00:09:49,993 --> 00:09:52,593
is there any typewriter fonts on the first page?

209
00:09:52,593 --> 00:09:53,994
I am a programming languages guy

210
00:09:53,994 --> 00:09:55,894
and programming languages people tend to

211
00:09:55,894 --> 00:09:58,347
put example programs in a typewriter font.

212
00:09:58,347 --> 00:10:00,447
So, I know if there is a typewriter font on the paper

213
00:10:00,447 --> 00:10:01,279
it's probably an example.

214
00:10:01,279 --> 00:10:02,289
That's good, right?

215
00:10:02,289 --> 00:10:03,959
But in your fields it may differ

216
00:10:03,959 --> 00:10:05,687
but anyway, start with an example

217
00:10:05,687 --> 00:10:06,597
that'll ilustrate your problem.

218
00:10:06,597 --> 00:10:07,947
So, here is an example of

219
00:10:07,947 --> 00:10:09,347
a paper that I wrote some time ago

220
00:10:09,347 --> 00:10:11,247
and the very first thing I did was to give

221
00:10:11,247 --> 00:10:13,257
a little program and explain a problem with it

222
00:10:13,257 --> 00:10:14,797
and explain that somehow

223
00:10:14,797 --> 00:10:16,417
my paper is gonna fix this problem.

224
00:10:16,417 --> 00:10:17,321
OK?

225
00:10:17,321 --> 00:10:18,791
So, that's an initial hook.

226
00:10:18,791 --> 00:10:20,721
Not a general description of a problem

227
00:00:57,631 --> 00:01:00,351
I often find, if I spend invested months of effort