1
00:00:00,000 --> 00:00:00,810
DAVID MALAN: Hello.

2
00:00:00,810 --> 00:00:03,315
My name is David Malan, and
I teach CS50 at Harvard,

3
00:00:03,315 --> 00:00:05,940
which is our introductory course
in computer science for majors

4
00:00:05,940 --> 00:00:08,850
and nonmajors taught
primarily in C and Python.

5
00:00:08,850 --> 00:00:11,880
Back in spring of 2020, Harvard,
like a lot of universities,

6
00:00:11,880 --> 00:00:15,390
sent everyone home midsemester
with classes, thereafter, online,

7
00:00:15,390 --> 00:00:19,200
and it soon became clear that fall
of 2020 would be online as well.

8
00:00:19,200 --> 00:00:23,220
Now, CS50 itself has long been available
online via Harvard extension school

9
00:00:23,220 --> 00:00:25,170
and, more recently, via edX as well.

10
00:00:25,170 --> 00:00:27,780
So we initially considered
simply reusing the videos

11
00:00:27,780 --> 00:00:30,480
that we had filmed live
in fall of 2019 since I

12
00:00:30,480 --> 00:00:33,750
didn't think I could do nearly
as good a job from home via Zoom.

13
00:00:33,750 --> 00:00:36,090
But we realized that there
might be a silver lining

14
00:00:36,090 --> 00:00:37,770
of that particular moment in time.

15
00:00:37,770 --> 00:00:41,430
With everyone off campus, there was
an unprecedented availability of space

16
00:00:41,430 --> 00:00:44,280
on campus, in particular
the university's Drama

17
00:00:44,280 --> 00:00:46,950
Center, home to the American
Repertory Theater, was

18
00:00:46,950 --> 00:00:49,560
vacant, its team of
artisans idled as well

19
00:00:49,560 --> 00:00:51,698
with no shows scheduled any time soon.

20
00:00:51,698 --> 00:00:53,490
And so we reached out
to see if we might be

21
00:00:53,490 --> 00:00:55,770
able to teach the
course live on its stage

22
00:00:55,770 --> 00:00:59,280
with students present only via
Zoom and with just a few staff

23
00:00:59,280 --> 00:01:01,020
present in person.

24
00:01:01,020 --> 00:01:03,690
Amazingly, after quite a few
emails and Zoom meetings,

25
00:01:03,690 --> 00:01:07,290
the answer came back as a yes,
albeit subject to rigorous protocols.

26
00:01:07,290 --> 00:01:10,710
Only one person while teaching on
camera could be unmasked at a time

27
00:01:10,710 --> 00:01:13,830
with everyone else wearing
face masks and face shields

28
00:01:13,830 --> 00:01:16,830
separated from each
other by 15 or more feet.

29
00:01:16,830 --> 00:01:19,400
Now, I should mention
that, in healthier times,

30
00:01:19,400 --> 00:01:22,170
CS50 is already taught in a
theater, a beautiful one at that,

31
00:01:22,170 --> 00:01:26,640
but what that theater doesn't have is a
wide-open schedule, an entire prop shop

32
00:01:26,640 --> 00:01:29,580
in back, and experts in
lighting and set design.

33
00:01:29,580 --> 00:01:32,400
All things considered,
then, fall of 2020

34
00:01:32,400 --> 00:01:35,430
offered an amazing opportunity
for our computer science class

35
00:01:35,430 --> 00:01:37,530
to collaborate with a
whole team of artisans

36
00:01:37,530 --> 00:01:41,580
to bring computer science to life
with all the more theatricality

37
00:01:41,580 --> 00:01:43,530
than is ordinarily possible.

38
00:01:43,530 --> 00:01:47,430
Our pedagogical goal, ultimately, was
to create all the more memorable moments

39
00:01:47,430 --> 00:01:51,810
for students particularly in those
earliest days of COVID-19 with everyone

40
00:01:51,810 --> 00:01:56,790
online in front of a computer all day
to help students visualize, understand,

41
00:01:56,790 --> 00:02:01,530
and remember concepts more than a
lecturer's words alone might allow.

42
00:02:01,530 --> 00:02:03,870
That said, nearly all
of the demonstrations

43
00:02:03,870 --> 00:02:06,000
that we did that semester
on stage were really

44
00:02:06,000 --> 00:02:08,460
just higher-end equivalents
of ones that we have long

45
00:02:08,460 --> 00:02:11,730
done more modestly in person in
healthier times in classrooms

46
00:02:11,730 --> 00:02:12,840
large and small.

47
00:02:12,840 --> 00:02:15,090
Indeed, most everything
you're about to see

48
00:02:15,090 --> 00:02:18,570
can also be done with some
chalk paper and tape alone,

49
00:02:18,570 --> 00:02:21,000
but we thought we'd share
with this talk just some

50
00:02:21,000 --> 00:02:23,070
of what we did that
particular term and why

51
00:02:23,070 --> 00:02:26,490
we did it with even more in the paper.

52
00:02:26,490 --> 00:02:32,100
All of these slides can be
found at cs50.ly/sigcse23.

53
00:02:32,100 --> 00:02:34,620
And allow me to emphasize
that we only claim to have

54
00:02:34,620 --> 00:02:36,300
enacted these particular moments.

55
00:02:36,300 --> 00:02:39,720
Each was undoubtedly originated
or inspired by my own teachers

56
00:02:39,720 --> 00:02:42,900
professors and colleagues over time.

57
00:02:42,900 --> 00:02:46,230
Here, then, is the theater in question.

58
00:02:46,230 --> 00:02:49,170
Here are those same seats
filled virtually via Zoom

59
00:02:49,170 --> 00:02:55,120
with as many as 300 students at once
visible across six very large screens.

60
00:02:55,120 --> 00:03:00,780
And here, closer up, is what I would
see when interacting with students.

61
00:03:00,780 --> 00:03:03,690
What students saw, meanwhile,
looked a little something like this,

62
00:03:03,690 --> 00:03:06,990
and it was on this stage that we
brought one of our first demonstrations

63
00:03:06,990 --> 00:03:07,920
to life.

64
00:03:07,920 --> 00:03:09,810
Funny enough, the
demonstration that students

65
00:03:09,810 --> 00:03:13,050
anecdotally mentioned as the
most memorable over the years

66
00:03:13,050 --> 00:03:14,970
doesn't require a stage
at all just a phone

67
00:03:14,970 --> 00:03:18,120
book, which, admittedly, is perhaps
just as hard to come by these days.

68
00:03:18,120 --> 00:03:21,330
But it's a physical analog of the
context that students of course

69
00:03:21,330 --> 00:03:24,480
now have in their phones, so the
idea of searching for someone

70
00:03:24,480 --> 00:03:26,010
is no less familiar.

71
00:03:26,010 --> 00:03:29,430
Indeed, we use this prop to
introduce students to algorithms.

72
00:03:29,430 --> 00:03:31,500
We start by asking how
we might find someone,

73
00:03:31,500 --> 00:03:34,530
say, David in this here phonebook
if it's sorted alphabetically

74
00:03:34,530 --> 00:03:35,460
by first name.

75
00:03:35,460 --> 00:03:39,240
And I propose to start looking from
the beginning, one page at a time.

76
00:03:39,240 --> 00:03:41,550
After a few such page
turns, I ask students

77
00:03:41,550 --> 00:03:43,410
whether the algorithm is correct.

78
00:03:43,410 --> 00:03:45,990
Invariably, a few students
say yes, but a few also

79
00:03:45,990 --> 00:03:49,050
say no, conflating
correctness with efficiency.

80
00:03:49,050 --> 00:03:52,530
So I then define both and then
propose that we try to do better.

81
00:03:52,530 --> 00:03:55,896
I then start to flip two
pages at a time, 2, 4, 6,

82
00:03:55,896 --> 00:04:00,960
8, noting that it surely sounds
and is faster, but is it correct?

83
00:04:00,960 --> 00:04:04,590
A student usually recognizes that I
might accidentally skip over the person

84
00:04:04,590 --> 00:04:08,280
that I'm looking for, so I acknowledge
that I might need to double back a bit

85
00:04:08,280 --> 00:04:09,420
if I've gone too far.

86
00:04:09,420 --> 00:04:13,830
It's still twice as fast, give or take
a page, but can we do even better?

87
00:04:13,830 --> 00:04:15,870
I then propose to start in the middle.

88
00:04:15,870 --> 00:04:20,100
Looking down, I notice that I'm in
the M section, so what do I do now?

89
00:04:20,100 --> 00:04:23,070
Well, David, which starts with
a D, is surely to the left,

90
00:04:23,070 --> 00:04:27,990
so I can now figuratively and literally
tear the problem in half, theatrically

91
00:04:27,990 --> 00:04:30,960
so, throwing half of the problem away.

92
00:04:30,960 --> 00:04:34,590
And I can repeat, repeat, repeat
until I'm left with just one page.

93
00:04:34,590 --> 00:04:36,930
And to be fair, no more
phone book, but we then

94
00:04:36,930 --> 00:04:39,840
talk about how we might analyze
the algorithms, the first two

95
00:04:39,840 --> 00:04:42,720
linear and the third
logarithmic in running time.

96
00:04:42,720 --> 00:04:45,300
And we then translate the demo
to pseudocode, at which point

97
00:04:45,300 --> 00:04:48,030
we introduce vocabulary,
like functions, conditionals,

98
00:04:48,030 --> 00:04:51,030
boolean expressions, loops, and more.

99
00:04:51,030 --> 00:04:53,700
A few weeks later, we revisit
algorithms but in the context

100
00:04:53,700 --> 00:04:54,990
of arrays specifically.

101
00:04:54,990 --> 00:04:58,290
For this moment in time, I've long
used sheets of paper taped to the board

102
00:04:58,290 --> 00:05:01,740
with numbers written in chalk behind
each to represent some arrays,

103
00:05:01,740 --> 00:05:04,110
that the sheets are covering
the numbers is deliberate.

104
00:05:04,110 --> 00:05:06,990
Neither we nor a computer can
see all of the numbers at once.

105
00:05:06,990 --> 00:05:09,870
We must index into it
to see one at a time.

106
00:05:09,870 --> 00:05:12,360
Sometimes we've used
small gym lockers instead

107
00:05:12,360 --> 00:05:15,810
to make even clear that the numbers
are effectively behind closed doors.

108
00:05:15,810 --> 00:05:18,810
But in fall of 2020, the
theater had a whole warehouse

109
00:05:18,810 --> 00:05:22,890
of props used in past shows, and so
we brought those doors to life size

110
00:05:22,890 --> 00:05:26,070
with an eclectic array of
seven actual doors on stage.

111
00:05:26,070 --> 00:05:29,640
Behind each was a number, brought to
life by the theater's prop shop, some

112
00:05:29,640 --> 00:05:31,800
with fur, some even with googly eyes.

113
00:05:31,800 --> 00:05:34,680
I then used these doors to
search for a particular number,

114
00:05:34,680 --> 00:05:36,960
once with the numbers
unordered and, again,

115
00:05:36,960 --> 00:05:40,290
with the numbers then ordered after
which we named my two approaches--

116
00:05:40,290 --> 00:05:42,630
linear and binary search respectively.

117
00:05:42,630 --> 00:05:45,210
We then transitioned to a
discussion of asymptotic running

118
00:05:45,210 --> 00:05:47,760
time, including big O and omega.

119
00:05:47,760 --> 00:05:50,520
Now, in healthier times, we
would have two students volunteer

120
00:05:50,520 --> 00:05:52,500
to do the searches
instead, though we've found

121
00:05:52,500 --> 00:05:55,290
that it helps our own
pedagogical narrative to guide

122
00:05:55,290 --> 00:05:57,600
the students toward one
algorithm than the other

123
00:05:57,600 --> 00:06:01,260
rather than simply invite them to
find some number without any guidance.

124
00:06:01,260 --> 00:06:03,630
Otherwise, the first student
invariably hesitates,

125
00:06:03,630 --> 00:06:07,080
thinking there's some trick to finding
the number when there isn't, and then

126
00:06:07,080 --> 00:06:10,560
they start opening doors randomly, much
like a game show, rather than linearly.

127
00:06:10,560 --> 00:06:13,110
And worse, they sometimes
find the number in question

128
00:06:13,110 --> 00:06:17,250
behind their very first choice of doors,
which, of course, is hilarious and met

129
00:06:17,250 --> 00:06:18,750
with applause from their classmates.

130
00:06:18,750 --> 00:06:22,260
But it's not exactly ideal for
a discussion of upper bounds,

131
00:06:22,260 --> 00:06:23,310
though great for lower.

132
00:06:23,310 --> 00:06:25,470
So the specific
narrative for this moment

133
00:06:25,470 --> 00:06:29,550
rather depends on just how much
the instructor leaves to chance.

134
00:06:29,550 --> 00:06:31,530
The next week we focus
for some time on how

135
00:06:31,530 --> 00:06:34,260
to swap two values in
memory as might be necessary

136
00:06:34,260 --> 00:06:36,540
when sorting the last
week's array so that you

137
00:06:36,540 --> 00:06:38,610
can, in fact, use binary search.

138
00:06:38,610 --> 00:06:41,460
We take out two glasses of
water, each a different color

139
00:06:41,460 --> 00:06:45,090
thanks to some food coloring, and we
ask a volunteer-- or, in fall of 2020,

140
00:06:45,090 --> 00:06:49,260
a colleague in another room via video
conference-- to swap the two values.

141
00:06:49,260 --> 00:06:51,990
Invariably, the volunteer
hesitates, at which point

142
00:06:51,990 --> 00:06:56,670
I dramatically offer up a temporary
variable in the form of a third glass,

143
00:06:56,670 --> 00:07:00,120
although one year, students simply
swapped the positions of the two

144
00:07:00,120 --> 00:07:01,680
glasses and thought we were done.

145
00:07:01,680 --> 00:07:05,820
And then we enact swapping the two
liquids using the third variable.

146
00:07:05,820 --> 00:07:08,610
We then transitioned to issues
of scope and the implications

147
00:07:08,610 --> 00:07:10,530
of passing by value or reference.

148
00:07:10,530 --> 00:07:13,620
We consider code, like this
here, which, similarly, swaps

149
00:07:13,620 --> 00:07:16,440
two values A and B
perfectly fine but fails

150
00:07:16,440 --> 00:07:19,080
to swap the callers, copies thereof.

151
00:07:19,080 --> 00:07:21,840
So at that point, we introduce pointers.

152
00:07:21,840 --> 00:07:23,760
Now, I should mention
that, a few years ago, I

153
00:07:23,760 --> 00:07:26,850
thought it'd be clever to demonstrate
XOR swapping as well whereby

154
00:07:26,850 --> 00:07:30,150
you can swap two values
using bitwise XOR operations

155
00:07:30,150 --> 00:07:31,890
without a temporary variable.

156
00:07:31,890 --> 00:07:34,170
And so I used water
and oil for that demo,

157
00:07:34,170 --> 00:07:36,120
thinking cleverly that
the two wouldn't mix

158
00:07:36,120 --> 00:07:38,670
and so I could get away
with just two glasses.

159
00:07:38,670 --> 00:07:41,190
It turns out that's easier
said than done in practice

160
00:07:41,190 --> 00:07:45,150
and yielded a memorable moment
that just didn't work very well.

161
00:07:45,150 --> 00:07:47,880
We continue our discussion of
memory, though, that same week,

162
00:07:47,880 --> 00:07:51,030
representing pointers
metaphorically as mailboxes.

163
00:07:51,030 --> 00:07:54,180
And one mailbox is an
address, like 0x123,

164
00:07:54,180 --> 00:07:56,880
and at that address is 50, an int.

165
00:07:56,880 --> 00:07:59,055
At this point, I tend to
put on an oversized foam

166
00:07:59,055 --> 00:08:01,680
finger akin to what you'd see in
stadium to make clear that one

167
00:08:01,680 --> 00:08:04,230
value effectively points to the other.

168
00:08:04,230 --> 00:08:06,570
Admittedly better than
standalone mailboxes

169
00:08:06,570 --> 00:08:09,510
might be a whole wall of apartment
mailboxes akin to an array,

170
00:08:09,510 --> 00:08:12,930
but the standalone mailboxes are
much easier to transport and use

171
00:08:12,930 --> 00:08:15,810
in traditional classrooms as well.

172
00:08:15,810 --> 00:08:19,590
Toward the end of our semester, we
also introduce students to a bit of SQL

173
00:08:19,590 --> 00:08:21,720
and, eventually, concepts
like race conditions.

174
00:08:21,720 --> 00:08:24,630
We suppose that there are some
popular posts on a social media

175
00:08:24,630 --> 00:08:27,150
site that lots of users
are liking at once,

176
00:08:27,150 --> 00:08:29,850
and we suppose that the site
is using two SQL statements

177
00:08:29,850 --> 00:08:33,480
to keep track of those likes, one to
select the current number of likes

178
00:08:33,480 --> 00:08:35,820
and one to update that number by one.

179
00:08:35,820 --> 00:08:39,360
We ask students what might happen if
those two statements aren't executed

180
00:08:39,360 --> 00:08:43,169
atomically for each user's like,
together at once without interruption,

181
00:08:43,169 --> 00:08:45,090
by some other user's like.

182
00:08:45,090 --> 00:08:48,630
We answer that we might
miscount the likes, but why?

183
00:08:48,630 --> 00:08:51,720
Well, we walk over to an
actual refrigerator on stage,

184
00:08:51,720 --> 00:08:53,340
or a plastic miniature thereof.

185
00:08:53,340 --> 00:08:56,580
And we suppose that two roommates
have the same in their dorm

186
00:08:56,580 --> 00:08:59,010
room, and both of them really like milk.

187
00:08:59,010 --> 00:09:01,170
One roommate comes home
one day, opens the fridge,

188
00:09:01,170 --> 00:09:02,790
and they discover they're out of milk.

189
00:09:02,790 --> 00:09:04,170
So they head to the store.

190
00:09:04,170 --> 00:09:06,780
Meanwhile, the other roommate
comes home, opens the fridge,

191
00:09:06,780 --> 00:09:09,120
discovers the same, and
heads to another store

192
00:09:09,120 --> 00:09:11,400
without crossing paths with the other.

193
00:09:11,400 --> 00:09:14,820
Eventually, both of them return home
only to realize that both of them

194
00:09:14,820 --> 00:09:15,720
have bought milk.

195
00:09:15,720 --> 00:09:18,300
But they don't really like
milk that much, so some of it

196
00:09:18,300 --> 00:09:19,500
will now go to waste.

197
00:09:19,500 --> 00:09:22,350
So how do we avoid that problem?

198
00:09:22,350 --> 00:09:24,630
Now, at least one
student usually suggests

199
00:09:24,630 --> 00:09:27,720
that we leave a note on the fridge,
which would, indeed, suffice,

200
00:09:27,720 --> 00:09:29,910
but we take the opportunity
more dramatically

201
00:09:29,910 --> 00:09:32,220
to lock the fridge all together.

202
00:09:32,220 --> 00:09:35,250
We explain that the solution
for likes in a database

203
00:09:35,250 --> 00:09:38,460
is also a lock or
lower-level transaction,

204
00:09:38,460 --> 00:09:40,830
and we emphasize that
both problems arose

205
00:09:40,830 --> 00:09:43,260
because we inspected the
state of some variable

206
00:09:43,260 --> 00:09:48,180
while it was being updated and made
a decision based on that state.

207
00:09:48,180 --> 00:09:51,240
Those then are just a few
of the memorable moments

208
00:09:51,240 --> 00:09:52,740
that we present in the paper.

209
00:09:52,740 --> 00:09:54,360
We discuss others as well.

210
00:09:54,360 --> 00:09:58,740
We use a black box to explain
functions with inputs and outputs.

211
00:09:58,740 --> 00:10:01,290
We rearrange plastic
numbers to explain sorting,

212
00:10:01,290 --> 00:10:05,040
using one shelf for selection sort
and bubble sort but multiple shelves

213
00:10:05,040 --> 00:10:07,140
and, thus, more space for merge sort.

214
00:10:07,140 --> 00:10:10,380
We create a pyramid of cardboard
bricks to explain recursion

215
00:10:10,380 --> 00:10:13,920
whereby a pyramid of height n is
but a pyramid of height n minus 1

216
00:10:13,920 --> 00:10:16,290
plus an additional layer of bricks.

217
00:10:16,290 --> 00:10:19,230
We draw on a grid of
tiles to explain memory,

218
00:10:19,230 --> 00:10:21,600
each representing some number of bytes.

219
00:10:21,600 --> 00:10:25,650
We allocate wooden blocks to
build a linked list of nodes,

220
00:10:25,650 --> 00:10:28,200
though I should mention that
of all the demonstrations

221
00:10:28,200 --> 00:10:30,750
we did that term, this
one I actually regret.

222
00:10:30,750 --> 00:10:33,690
When the prop shop asked, how
large I'd like the blocks to be?

223
00:10:33,690 --> 00:10:35,760
I proposed three feet.

224
00:10:35,760 --> 00:10:40,170
Well, it turns out that three foot
nodes made out of wood are much, much,

225
00:10:40,170 --> 00:10:43,200
much heavier than I
thought, a testament perhaps

226
00:10:43,200 --> 00:10:45,960
to how difficult it is to manage memory.

227
00:10:45,960 --> 00:10:48,690
We hash oversized playing
cards into actual buckets

228
00:10:48,690 --> 00:10:50,790
to explain hashing and hash tables.

229
00:10:50,790 --> 00:10:54,510
We enact callback functions by phone
whereby I ask a colleague a question,

230
00:10:54,510 --> 00:10:56,700
and he calls me back with the answer.

231
00:10:56,700 --> 00:10:58,860
And we use light bulbs
to explain binary,

232
00:10:58,860 --> 00:11:02,010
each week encoding an Easter egg
in the theater's own stage lights

233
00:11:02,010 --> 00:11:03,780
using unicode.

234
00:11:03,780 --> 00:11:07,170
Ultimately, our use of so many
physical props in fall of 2020

235
00:11:07,170 --> 00:11:10,410
was well received by
students with 65% of students

236
00:11:10,410 --> 00:11:14,847
reporting that they loved, use thereof,
and 27% reporting that they liked.

237
00:11:14,847 --> 00:11:16,680
We asked students more
specifically, though,

238
00:11:16,680 --> 00:11:19,080
just how helpful they
found each demonstration,

239
00:11:19,080 --> 00:11:22,050
and atop the list, funny enough,
was arguably the simplest of them,

240
00:11:22,050 --> 00:11:24,420
tearing a phonebook to
explain binary search.

241
00:11:24,420 --> 00:11:28,050
At the bottom of the list, meanwhile,
was the black box to explain functions.

242
00:11:28,050 --> 00:11:30,930
Here along the vertical
axis is each demonstration,

243
00:11:30,930 --> 00:11:33,900
and each horizontal bar depicts
just how helpful students

244
00:11:33,900 --> 00:11:35,340
found each demonstration.

245
00:11:35,340 --> 00:11:37,830
From left to right,
blue is very helpful,

246
00:11:37,830 --> 00:11:41,100
red is somewhat helpful, and
then neutral, somewhat unhelpful,

247
00:11:41,100 --> 00:11:43,230
very unhelpful, and unsure.

248
00:11:43,230 --> 00:11:47,460
So a long blue bar means that many
students found a demo very helpful,

249
00:11:47,460 --> 00:11:51,300
and a shorter blue bar means that fewer
students found a demo very helpful.

250
00:11:51,300 --> 00:11:54,210
We explore in the paper why we
think that might be for some demos.

251
00:11:54,210 --> 00:11:56,520
For instance, we suspect
we're not spending enough time

252
00:11:56,520 --> 00:11:59,310
using the black box to explain
functions and that we're actually

253
00:11:59,310 --> 00:12:03,210
spending more time than is necessary
using light bulbs to explain binary,

254
00:12:03,210 --> 00:12:06,990
though future work includes identifying
more precisely why some didn't work

255
00:12:06,990 --> 00:12:09,360
as well as others so that we can retool.

256
00:12:09,360 --> 00:12:11,325
From surveys of students,
though, we already

257
00:12:11,325 --> 00:12:13,200
know that the course's
use of physical props,

258
00:12:13,200 --> 00:12:17,400
help me understand at a different
level and help me remember the material

259
00:12:17,400 --> 00:12:19,920
better, which was precisely a name.

260
00:12:19,920 --> 00:12:23,190
They made me comprehend something
instead of just recalling it.

261
00:12:23,190 --> 00:12:26,040
For instance, I will never forget
about locking transactions.

262
00:12:26,040 --> 00:12:27,630
Think double milk jugs.

263
00:12:27,630 --> 00:12:31,800
Or link lists, think boxes
connected by arrows pointers.

264
00:12:31,800 --> 00:12:35,040
It helped to visualize
all of the code and ideas

265
00:12:35,040 --> 00:12:38,910
that would otherwise just be letters,
symbols, numbers, and indents.

266
00:12:38,910 --> 00:12:41,550
It really helped me
stay engaged, especially

267
00:12:41,550 --> 00:12:44,040
in the virtual nature of this semester.

268
00:12:44,040 --> 00:12:46,710
The flip side, though, is
that all the theatricality,

269
00:12:46,710 --> 00:12:49,230
even if modestly implemented
with just paper and tape,

270
00:12:49,230 --> 00:12:51,810
does come at a cost, namely time.

271
00:12:51,810 --> 00:12:55,260
As one student put it, the
demos were often unnecessary,

272
00:12:55,260 --> 00:12:56,790
helpful but unnecessary.

273
00:12:56,790 --> 00:12:59,640
For me, an animated diagram on the
slides would have been sufficient

274
00:12:59,640 --> 00:13:03,280
and often felt that these demonstrations
took up more time than necessary.

275
00:13:03,280 --> 00:13:07,120
And another student distinguished
between engagement and actual learning.

276
00:13:07,120 --> 00:13:09,670
I like them because they were
engaging, but I don't think

277
00:13:09,670 --> 00:13:11,590
they helped a lot with my learning.

278
00:13:11,590 --> 00:13:13,960
That said, another acknowledged
that I don't really

279
00:13:13,960 --> 00:13:17,500
get value from stuff like that,
but I know it helps others.

280
00:13:17,500 --> 00:13:20,213
So that's a balance we ourselves
have long struggled with.

281
00:13:20,213 --> 00:13:22,630
Indeed, if there's been one
long-standing critique of CS50

282
00:13:22,630 --> 00:13:25,780
it's just that it's too theatrical
for some students tastes.

283
00:13:25,780 --> 00:13:29,020
In fact, one of his papers on
reviewers couldn't have said it better.

284
00:13:29,020 --> 00:13:31,150
In a teaching environment,
there is a balance

285
00:13:31,150 --> 00:13:33,850
that needs to be struck between
the inquisitive newcomer, who

286
00:13:33,850 --> 00:13:36,670
we want to inspire, and the
more advanced early student

287
00:13:36,670 --> 00:13:37,930
that we want to challenge.

288
00:13:37,930 --> 00:13:42,220
How much theater in the classroom
for an individual class is advisable?

289
00:13:42,220 --> 00:13:45,670
Finding that balance, then, is among our
own goals ahead or perhaps confirming

290
00:13:45,670 --> 00:13:48,130
that we have found it, even
if the global optimum isn't

291
00:13:48,130 --> 00:13:50,170
locally optimal for some learners.

292
00:13:50,170 --> 00:13:52,660
And we don't yet, but it
would be nice to track

293
00:13:52,660 --> 00:13:54,730
the impact of these
memorable moments on students

294
00:13:54,730 --> 00:13:57,880
beyond the end of our own one course
or, better yet, beyond graduation

295
00:13:57,880 --> 00:14:00,490
as via surveys or even
assessments to ascertain

296
00:14:00,490 --> 00:14:03,790
if they help students visualize,
understand, and remember concepts only

297
00:14:03,790 --> 00:14:07,330
in the short term or in
the long term as well.

298
00:14:07,330 --> 00:14:12,550
Videos of all these demonstrations and
more can be found at youtube.com/cs50.

299
00:14:12,550 --> 00:14:14,680
Please do reach out via
email with any questions,

300
00:14:14,680 --> 00:14:17,710
and thank you to all of my colleagues
at Harvard, to all of my teachers

301
00:14:17,710 --> 00:14:21,970
and professors past, and to our friends
at the American Repertory Theater.

302
00:14:21,970 --> 00:14:25,180
Theatricality does not
require an actual theater,

303
00:14:25,180 --> 00:14:27,400
but in that particular
moment in time, we

304
00:14:27,400 --> 00:14:29,770
do think our course
was the better for it.

305
00:14:29,770 --> 00:14:34,990
And it seems only appropriate to end
this talk now on a theatrical note.

306
00:14:34,990 --> 00:14:38,340
[MUSIC PLAYING]

307
00:14:38,340 --> 00:15:33,000