1 00:00:00,420 --> 00:00:05,910 [Music] 2 00:00:10,639 --> 00:00:17,279 Good day and welcome to my talk, Why 3 00:00:14,080 --> 00:00:20,240 Teach the Y. Um, we're going to talk 4 00:00:17,279 --> 00:00:23,600 about how context empowers learning, and 5 00:00:20,240 --> 00:00:27,119 about how carving out time to dedicate 6 00:00:23,600 --> 00:00:29,039 to context, improves outcomes, and can 7 00:00:27,119 --> 00:00:31,039 reduce the time that you have to spend 8 00:00:29,039 --> 00:00:33,280 on revision. 9 00:00:31,039 --> 00:00:36,239 Um, let's start with who I am and why 10 00:00:33,280 --> 00:00:39,440 I'm giving this talk. Uh, I'm Jack. You 11 00:00:36,239 --> 00:00:42,640 probably got that already. Um, I'm the 12 00:00:39,440 --> 00:00:45,040 CTO and co-founder of Kumo Study. We 13 00:00:42,640 --> 00:00:47,680 make uh study management tools for 14 00:00:45,040 --> 00:00:51,280 learners with ADHD and tools for 15 00:00:47,680 --> 00:00:53,600 institutions to support them. Um, and 16 00:00:51,280 --> 00:00:56,000 I'm a computer science educator. I have 17 00:00:53,600 --> 00:00:59,039 a decade of experience teaching at camps 18 00:00:56,000 --> 00:01:01,359 and workshops and incursions as well as 19 00:00:59,039 --> 00:01:04,559 I spent two years teaching at a high 20 00:01:01,359 --> 00:01:06,720 school in Sydney. 21 00:01:04,559 --> 00:01:08,080 I want to kick things off with a little 22 00:01:06,720 --> 00:01:10,640 scenario that I think all of the 23 00:01:08,080 --> 00:01:12,799 educators in the room here have probably 24 00:01:10,640 --> 00:01:14,640 experienced. Don't worry too much about 25 00:01:12,799 --> 00:01:17,759 the specifics. It's more just the 26 00:01:14,640 --> 00:01:20,240 general shape of the scenario. So, you 27 00:01:17,759 --> 00:01:21,920 spend some time teaching, say, databases 28 00:01:20,240 --> 00:01:23,520 or some other topic to your students and 29 00:01:21,920 --> 00:01:26,400 it seems like they understand it and 30 00:01:23,520 --> 00:01:28,479 it's all going pretty well. And then you 31 00:01:26,400 --> 00:01:31,439 move on to another topic, say cyber 32 00:01:28,479 --> 00:01:32,799 security or electronics. Again, they're 33 00:01:31,439 --> 00:01:34,799 understanding it and things are going 34 00:01:32,799 --> 00:01:36,400 well. 35 00:01:34,799 --> 00:01:38,799 And then a student asks how they're 36 00:01:36,400 --> 00:01:41,040 related. 37 00:01:38,799 --> 00:01:42,960 Or for another example, suppose you're 38 00:01:41,040 --> 00:01:45,439 teaching a topic that you know 39 00:01:42,960 --> 00:01:48,320 necessitates a bit less time, maybe 40 00:01:45,439 --> 00:01:50,240 image file formats or different storage 41 00:01:48,320 --> 00:01:52,479 technologies and you cover what's 42 00:01:50,240 --> 00:01:54,399 discussed in the curriculum 43 00:01:52,479 --> 00:01:56,479 and then you move on to the next topic 44 00:01:54,399 --> 00:01:58,159 and then come time for the assignment or 45 00:01:56,479 --> 00:02:00,159 the exam, it seems like the students 46 00:01:58,159 --> 00:02:03,640 have more or less forgotten everything 47 00:02:00,159 --> 00:02:03,640 about that topic. 48 00:02:04,320 --> 00:02:09,360 These scenarios highlight the only 49 00:02:06,960 --> 00:02:13,640 problem with the curriculum. 50 00:02:09,360 --> 00:02:13,640 shared by computer. 51 00:02:24,720 --> 00:02:30,800 Of course, you try and put them in a 52 00:02:27,840 --> 00:02:33,680 sensible, cohesive order, but there is 53 00:02:30,800 --> 00:02:35,360 only so much that can be done. 54 00:02:33,680 --> 00:02:37,440 This leads to students feeling like the 55 00:02:35,360 --> 00:02:40,800 course is a mishmash of different 56 00:02:37,440 --> 00:02:42,800 elements unrelated to each other. And 57 00:02:40,800 --> 00:02:44,640 this in turn leads to students having a 58 00:02:42,800 --> 00:02:46,959 hard time building a strong mental model 59 00:02:44,640 --> 00:02:49,040 of the course. And so more time has to 60 00:02:46,959 --> 00:02:52,239 be spent on revision further cutting 61 00:02:49,040 --> 00:02:56,239 into time to cover the curriculum. So 62 00:02:52,239 --> 00:02:58,160 what can you do about it? Well, context 63 00:02:56,239 --> 00:03:00,480 is critical. 64 00:02:58,160 --> 00:03:02,720 I have found that when I spent more time 65 00:03:00,480 --> 00:03:05,760 on context, I was able to mitigate those 66 00:03:02,720 --> 00:03:09,280 problems to quite a considerable degree. 67 00:03:05,760 --> 00:03:12,480 I actually spent very little time on 68 00:03:09,280 --> 00:03:14,720 revision. Only a couple of handful, only 69 00:03:12,480 --> 00:03:17,280 a handful of weeks across all of 9 and 70 00:03:14,720 --> 00:03:19,519 10. And even then, with very little 71 00:03:17,280 --> 00:03:21,200 prompting, the students themselves 72 00:03:19,519 --> 00:03:22,800 actually provided the solutions for the 73 00:03:21,200 --> 00:03:25,120 initial examples I worked through on the 74 00:03:22,800 --> 00:03:26,640 board before I handed out revision 75 00:03:25,120 --> 00:03:28,800 sheets or cheat sheets or went through 76 00:03:26,640 --> 00:03:30,959 it again with them. But most 77 00:03:28,800 --> 00:03:33,920 importantly, I found that students grew 78 00:03:30,959 --> 00:03:35,760 more engaged and excited by our classes. 79 00:03:33,920 --> 00:03:37,599 They were very motivated to join in on 80 00:03:35,760 --> 00:03:39,519 class discussions. They always wanted to 81 00:03:37,599 --> 00:03:43,440 know more and they became excellent at 82 00:03:39,519 --> 00:03:45,040 connecting disperate topics together. 83 00:03:43,440 --> 00:03:47,680 So, let's clear up what I mean by 84 00:03:45,040 --> 00:03:49,519 context here. Since it's a bit of an 85 00:03:47,680 --> 00:03:52,080 ambiguous word, it could really mean 86 00:03:49,519 --> 00:03:57,200 almost anything you want it to mean. 87 00:03:52,080 --> 00:04:00,560 What I mean at its core is why is this 88 00:03:57,200 --> 00:04:02,720 the way that it is? The answers to this 89 00:04:00,560 --> 00:04:05,040 question vary wildly depending on the 90 00:04:02,720 --> 00:04:07,360 topic at hand. There's historical and 91 00:04:05,040 --> 00:04:09,920 social reasons. There's limitations of 92 00:04:07,360 --> 00:04:11,439 technology of when the topic was 93 00:04:09,920 --> 00:04:14,640 developed. And sometimes it's just the 94 00:04:11,439 --> 00:04:17,359 plain physics of the universe. 95 00:04:14,640 --> 00:04:21,199 I always ask myself and I encourage my 96 00:04:17,359 --> 00:04:23,600 students to ask why did 97 00:04:21,199 --> 00:04:26,800 uh why did the people making this 98 00:04:23,600 --> 00:04:30,560 technology system standard or whatever 99 00:04:26,800 --> 00:04:34,560 do it this way? People by and large 100 00:04:30,560 --> 00:04:37,759 don't do things for no reason. 101 00:04:34,560 --> 00:04:39,680 They may have a bad reason, but even a 102 00:04:37,759 --> 00:04:42,000 bad reason 103 00:04:39,680 --> 00:04:45,720 helps you understand. 104 00:04:42,000 --> 00:04:45,720 For example, you've 105 00:04:52,960 --> 00:04:58,240 on. It was the age of wisdom. It was the 106 00:04:56,000 --> 00:05:01,120 age of foolishness. 107 00:04:58,240 --> 00:05:02,479 He continues further. It was the epoch 108 00:05:01,120 --> 00:05:04,160 of belief. It was the epoch of 109 00:05:02,479 --> 00:05:08,400 incredility. 110 00:05:04,160 --> 00:05:11,440 Light and darkness, hope, despair. On 111 00:05:08,400 --> 00:05:13,199 and on he goes. 112 00:05:11,440 --> 00:05:15,440 He's supposed to be this great writer 113 00:05:13,199 --> 00:05:18,639 and certainly best of times, worst of 114 00:05:15,440 --> 00:05:21,120 times is a banger of a line. But why 115 00:05:18,639 --> 00:05:23,199 does he just keep going, driving his 116 00:05:21,120 --> 00:05:25,440 point into the ground? We get it. Things 117 00:05:23,199 --> 00:05:28,000 were both one way and the other way at 118 00:05:25,440 --> 00:05:29,520 the same time. London is two cities at 119 00:05:28,000 --> 00:05:33,199 once. 120 00:05:29,520 --> 00:05:36,759 Why are you like this? Well, Dickens was 121 00:05:33,199 --> 00:05:36,759 paid by the word. 122 00:05:38,080 --> 00:05:42,880 And suddenly all of that makes sense, 123 00:05:40,400 --> 00:05:44,639 doesn't it? And odds are much slimmer 124 00:05:42,880 --> 00:05:47,360 that you'll forget either that he was 125 00:05:44,639 --> 00:05:50,160 paid by the word or that there is much 126 00:05:47,360 --> 00:05:54,840 more after that famous line. 127 00:05:50,160 --> 00:05:54,840 So I do this in class 128 00:05:59,039 --> 00:06:05,120 with Python including using multiple 129 00:06:01,520 --> 00:06:08,000 conditions in an if statement and we 130 00:06:05,120 --> 00:06:11,039 moved on to logic gates and boolean 131 00:06:08,000 --> 00:06:13,840 algebra. So multiple conditions in an if 132 00:06:11,039 --> 00:06:17,120 statement and they were getting it but 133 00:06:13,840 --> 00:06:19,520 not as quickly or as well. 134 00:06:17,120 --> 00:06:21,600 And I realized that the problem was that 135 00:06:19,520 --> 00:06:24,400 all the exercises, even when wrapped up 136 00:06:21,600 --> 00:06:26,960 in a scenario or a story, just kind of 137 00:06:24,400 --> 00:06:30,120 look like this. And it's just shapes on 138 00:06:26,960 --> 00:06:30,120 a page. 139 00:06:39,520 --> 00:06:43,080 I knew that it 140 00:06:48,639 --> 00:06:51,639 used 141 00:06:53,520 --> 00:06:58,960 that circuitry build 142 00:06:56,639 --> 00:07:01,440 of course. So I decided it was time to 143 00:06:58,960 --> 00:07:05,960 stop doing abstract problems and start 144 00:07:01,440 --> 00:07:05,960 building a computer in from scratch. 145 00:07:06,080 --> 00:07:10,479 As a reminder, these kids are in grade 146 00:07:08,639 --> 00:07:11,759 nine. 147 00:07:10,479 --> 00:07:12,880 So, we start out with the simplest thing 148 00:07:11,759 --> 00:07:15,440 that I could think of that could be 149 00:07:12,880 --> 00:07:17,599 called a computer, a one bit adder. It 150 00:07:15,440 --> 00:07:19,680 takes in two one bit numbers and it adds 151 00:07:17,599 --> 00:07:21,280 them together, making a new two-bit 152 00:07:19,680 --> 00:07:24,720 number. 153 00:07:21,280 --> 00:07:27,280 So, I explained what we were doing and 154 00:07:24,720 --> 00:07:30,639 together we drew up a truth table for 155 00:07:27,280 --> 00:07:33,039 the one bit adder and I told them I told 156 00:07:30,639 --> 00:07:35,599 them to design a logic circuit that 157 00:07:33,039 --> 00:07:37,039 replicated this truth table. And they 158 00:07:35,599 --> 00:07:39,680 all made things that looked roughly like 159 00:07:37,039 --> 00:07:41,520 this. 160 00:07:39,680 --> 00:07:44,319 Once they did that, I congratulated them 161 00:07:41,520 --> 00:07:47,360 on having built a computer from scratch. 162 00:07:44,319 --> 00:07:49,440 It seems silly, but they genuinely had. 163 00:07:47,360 --> 00:07:53,599 This wasn't an abstract puzzle. This 164 00:07:49,440 --> 00:07:55,759 wasn't just plugging numbers into a 165 00:07:53,599 --> 00:07:57,360 design. 166 00:07:55,759 --> 00:07:59,599 So, 167 00:07:57,360 --> 00:08:01,520 I escalated. 168 00:07:59,599 --> 00:08:04,800 I said, "Okay, we're going to make a 169 00:08:01,520 --> 00:08:06,960 bit." Now, we can't quite just copy what 170 00:08:04,800 --> 00:08:09,840 we did because we've got that carry bit 171 00:08:06,960 --> 00:08:11,680 coming in from the one bit adder. 172 00:08:09,840 --> 00:08:13,840 So, we had to make a new truth table 173 00:08:11,680 --> 00:08:16,560 that accounted for that and then design 174 00:08:13,840 --> 00:08:18,720 a new circuit. And so, we made a one bit 175 00:08:16,560 --> 00:08:21,520 full adder. 176 00:08:18,720 --> 00:08:24,400 Once we had that, then we just had to 177 00:08:21,520 --> 00:08:26,720 glue two of them together like so. 178 00:08:24,400 --> 00:08:30,080 And then you can glue two of those 179 00:08:26,720 --> 00:08:32,399 together and two of those together. And 180 00:08:30,080 --> 00:08:34,719 now you've got a one bite adder. So now 181 00:08:32,399 --> 00:08:36,560 we're really cooking with gas. We've 182 00:08:34,719 --> 00:08:39,440 designed a computer that can take any 183 00:08:36,560 --> 00:08:41,200 two numbers between 0 and 255 and add 184 00:08:39,440 --> 00:08:45,000 them together. And that's pretty cool 185 00:08:41,200 --> 00:08:45,000 for a bunch of nicates 186 00:08:52,000 --> 00:08:56,399 and if they wanted they could go and buy 187 00:08:53,920 --> 00:08:58,720 some buttons and wires and more gates 188 00:08:56,399 --> 00:09:01,120 and they could build the design that I 189 00:08:58,720 --> 00:09:04,240 had that they had just made and it would 190 00:09:01,120 --> 00:09:05,760 work. I think actually doing that in 191 00:09:04,240 --> 00:09:07,360 class would have been a great thing to 192 00:09:05,760 --> 00:09:11,200 do but unfortunately I didn't quite get 193 00:09:07,360 --> 00:09:13,279 around to it. next time hopefully 194 00:09:11,200 --> 00:09:15,680 by the end 195 00:09:13,279 --> 00:09:17,519 uh and and then we continued on. We kept 196 00:09:15,680 --> 00:09:20,080 expanding this computer adding more 197 00:09:17,519 --> 00:09:22,480 features and functionality to it. By the 198 00:09:20,080 --> 00:09:24,800 end it could add and subtract bytes 199 00:09:22,480 --> 00:09:27,279 using two's complement. 200 00:09:24,800 --> 00:09:30,720 Uh you could put the result into one of 201 00:09:27,279 --> 00:09:33,120 either two bytes in memory. um and it 202 00:09:30,720 --> 00:09:36,080 could be given an instruction consisting 203 00:09:33,120 --> 00:09:38,000 of two bytes of data whether to add or 204 00:09:36,080 --> 00:09:40,560 subtract them and which memory address 205 00:09:38,000 --> 00:09:42,959 to store the result in. So this is 206 00:09:40,560 --> 00:09:46,320 genuinely a computer with no hint of 207 00:09:42,959 --> 00:09:48,560 exaggeration. I showed them how you can 208 00:09:46,320 --> 00:09:50,800 take the binary for that input there and 209 00:09:48,560 --> 00:09:53,200 that's called an op code and this is the 210 00:09:50,800 --> 00:09:55,519 basis of machine language. That's what 211 00:09:53,200 --> 00:09:57,920 assembly languages directly translate 212 00:09:55,519 --> 00:10:01,360 into. It's what Python indirectly 213 00:09:57,920 --> 00:10:03,920 becomes. All told, we spent most of term 214 00:10:01,360 --> 00:10:06,720 two in grade 9 working on these computer 215 00:10:03,920 --> 00:10:09,519 designs. And then after their assignment 216 00:10:06,720 --> 00:10:12,640 was done, we did not touch logic gates 217 00:10:09,519 --> 00:10:15,440 again until term three of year 10. So 218 00:10:12,640 --> 00:10:17,839 over a year later, 219 00:10:15,440 --> 00:10:20,800 guess what? They remembered it pretty 220 00:10:17,839 --> 00:10:24,399 well. They weren't entirely certain of 221 00:10:20,800 --> 00:10:27,120 which symbols were which gates 222 00:10:24,399 --> 00:10:29,040 but they were largely correct on that 223 00:10:27,120 --> 00:10:30,880 and they knew how to make a truth table 224 00:10:29,040 --> 00:10:32,399 and make a circuit and they knew how to 225 00:10:30,880 --> 00:10:34,959 take a circuit and make the truth table 226 00:10:32,399 --> 00:10:36,800 from that. They knew how to actually use 227 00:10:34,959 --> 00:10:40,519 the logic gates in a way that I only 228 00:10:36,800 --> 00:10:40,519 ever learned in university. 229 00:10:40,800 --> 00:10:46,079 This also works with smaller topics. So, 230 00:10:43,680 --> 00:10:49,200 I was teaching the Cambridge IGCSE 231 00:10:46,079 --> 00:10:51,200 curriculum and according to it, there's 232 00:10:49,200 --> 00:10:53,040 only one thing you need to know about 233 00:10:51,200 --> 00:10:54,160 sound files. So, you're getting the 234 00:10:53,040 --> 00:10:56,959 inside scoop on what that whole 235 00:10:54,160 --> 00:10:59,200 curriculum covers there. Sound files 236 00:10:56,959 --> 00:11:01,680 have a sample rate and a higher sample 237 00:10:59,200 --> 00:11:04,800 rate. 238 00:11:01,680 --> 00:11:07,120 Congratulations, you've graduated. 239 00:11:04,800 --> 00:11:10,240 So that's just one sentence and you can 240 00:11:07,120 --> 00:11:13,200 probably memorize it pretty easily. But 241 00:11:10,240 --> 00:11:15,360 memorization I feel isn't a satisfying 242 00:11:13,200 --> 00:11:17,519 learning experience. It's not going to 243 00:11:15,360 --> 00:11:21,839 enuse the students and it's prone to 244 00:11:17,519 --> 00:11:23,680 being forgotten. So I 245 00:11:21,839 --> 00:11:25,839 sample rate and why does it make a 246 00:11:23,680 --> 00:11:27,760 higher quality sound? 247 00:11:25,839 --> 00:11:30,720 Before I could answer that, I had to ask 248 00:11:27,760 --> 00:11:33,839 what even is a sound? I forgot to click 249 00:11:30,720 --> 00:11:35,680 through my slides. 250 00:11:33,839 --> 00:11:38,399 Sound is the vibration of particles in 251 00:11:35,680 --> 00:11:40,240 air. And we can represent that vibration 252 00:11:38,399 --> 00:11:43,200 with waves. 253 00:11:40,240 --> 00:11:47,880 A single a simple sound of just a single 254 00:11:43,200 --> 00:11:47,880 note is a nice sine wave. 255 00:11:54,959 --> 00:12:00,720 speakers in controlled by an 256 00:11:57,600 --> 00:12:04,320 electromagnet to push the air in the 257 00:12:00,720 --> 00:12:07,279 same pattern. And so this complex wave 258 00:12:04,320 --> 00:12:09,680 or even the simple wave in fact shows 259 00:12:07,279 --> 00:12:12,240 the position of the speaker at any 260 00:12:09,680 --> 00:12:14,000 moment in time. 261 00:12:12,240 --> 00:12:16,240 And that and the sample rate is how 262 00:12:14,000 --> 00:12:18,720 often we check where the speaker should 263 00:12:16,240 --> 00:12:20,800 be. 264 00:12:18,720 --> 00:12:23,279 If we go back to the simple example, 265 00:12:20,800 --> 00:12:24,560 we've got just a single note here at 500 266 00:12:23,279 --> 00:12:28,079 htz. 267 00:12:24,560 --> 00:12:30,800 This graph just shows an 8 slice of it. 268 00:12:28,079 --> 00:12:33,200 At one extreme of sampling, we have the 269 00:12:30,800 --> 00:12:35,920 perfect sample. 270 00:12:33,200 --> 00:12:39,040 This is if we could somehow sample every 271 00:12:35,920 --> 00:12:41,600 infinite decimal slice of this graph and 272 00:12:39,040 --> 00:12:42,959 then we can reproduce the sound. 273 00:12:41,600 --> 00:12:45,200 But then if we're going to store this 274 00:12:42,959 --> 00:12:47,839 digitally, then that requires infinite 275 00:12:45,200 --> 00:12:49,920 digital storage space. Um, and I don't 276 00:12:47,839 --> 00:12:52,880 have that. 277 00:12:49,920 --> 00:12:56,079 So let's jump to the other extreme and 278 00:12:52,880 --> 00:12:57,440 go for a very low sample rate. Um, you 279 00:12:56,079 --> 00:12:59,760 can see our original signal in blue 280 00:12:57,440 --> 00:13:03,120 here. And if we sample the note at the 281 00:12:59,760 --> 00:13:06,399 same frequency, 500 times a second, uh, 282 00:13:03,120 --> 00:13:08,079 we get that red line here. Um, so it's 283 00:13:06,399 --> 00:13:10,000 always at zero and the speaker doesn't 284 00:13:08,079 --> 00:13:13,839 move. 285 00:13:10,000 --> 00:13:15,600 So that's not going to be enough. 286 00:13:13,839 --> 00:13:18,880 If we go just a little bit more 287 00:13:15,600 --> 00:13:20,720 frequently, say 502 times a second, we 288 00:13:18,880 --> 00:13:23,440 can see that we've got a sample that 289 00:13:20,720 --> 00:13:25,760 slowly moves the speaker. 290 00:13:23,440 --> 00:13:28,000 Um, this is clearer to see what it's 291 00:13:25,760 --> 00:13:30,240 doing wrong. If we expand this view out 292 00:13:28,000 --> 00:13:32,160 to a whole second of the soundwave, 293 00:13:30,240 --> 00:13:34,800 we've got our original soundwave at 500 294 00:13:32,160 --> 00:13:38,320 Hz still in blue, and we've got what 295 00:13:34,800 --> 00:13:39,839 comes out of our sample here. And that's 296 00:13:38,320 --> 00:13:44,839 definitely not the same note as the blue 297 00:13:39,839 --> 00:13:44,839 wave. Uh that's only about 2 herz 298 00:13:47,920 --> 00:13:52,120 but actually produce a sound 299 00:13:57,760 --> 00:14:02,079 triangular wave pattern. So it will 300 00:13:59,519 --> 00:14:05,760 still sound weird and robotic but at 301 00:14:02,079 --> 00:14:07,440 least it is the right pitch. Now, 302 00:14:05,760 --> 00:14:09,600 it will also struggle with a more 303 00:14:07,440 --> 00:14:11,920 complex sound wave. You can see that 304 00:14:09,600 --> 00:14:13,279 much of this wave gets cut off and so 305 00:14:11,920 --> 00:14:14,800 the sound is going to be heavily 306 00:14:13,279 --> 00:14:16,480 distorted. 307 00:14:14,800 --> 00:14:19,120 So, we need the sample rate to be at 308 00:14:16,480 --> 00:14:22,800 least four times the frequency of the 309 00:14:19,120 --> 00:14:25,120 highest sound that we want to record. 310 00:14:22,800 --> 00:14:27,760 Human hearing range is generally between 311 00:14:25,120 --> 00:14:30,240 15 to 20 kHz, but the top end of that 312 00:14:27,760 --> 00:14:33,279 isn't commonly used in audio files or 313 00:14:30,240 --> 00:14:36,560 music. Um, the highest note on a piano 314 00:14:33,279 --> 00:14:39,920 is just under 8 kHz. So, we'd need a 315 00:14:36,560 --> 00:14:41,920 sample rate of at least 32 kHz. We can 316 00:14:39,920 --> 00:14:45,120 go higher for smoother, more natural 317 00:14:41,920 --> 00:14:47,680 sounds, as well as to accommodate uh the 318 00:14:45,120 --> 00:14:51,040 top ends of the hearing range. So, audio 319 00:14:47,680 --> 00:14:54,160 is commonly recorded at 44.1 kHz or 48 320 00:14:51,040 --> 00:14:56,800 kHz. This graph shows here an 8 kHz 321 00:14:54,160 --> 00:14:59,040 sound recorded at 44.1. And you can see 322 00:14:56,800 --> 00:15:02,800 that it's not perfect. You do get little 323 00:14:59,040 --> 00:15:05,839 clipped edges on a lot of the waves, but 324 00:15:02,800 --> 00:15:08,959 overall it's a pretty close match. 325 00:15:05,839 --> 00:15:11,839 Now, in class, I actually went through 326 00:15:08,959 --> 00:15:13,839 all of these graphs with my students. We 327 00:15:11,839 --> 00:15:17,000 saw what the sample rate was. We saw how 328 00:15:13,839 --> 00:15:17,000 it impeds 329 00:15:17,920 --> 00:15:22,399 behind it. We covered this over a single 330 00:15:20,639 --> 00:15:25,760 lesson. It didn't take much longer than 331 00:15:22,399 --> 00:15:27,440 I just spent with you. Um, 332 00:15:25,760 --> 00:15:29,680 and we covered this in the first few 333 00:15:27,440 --> 00:15:31,120 weeks of grade 10 and then we didn't 334 00:15:29,680 --> 00:15:32,959 discuss it until for the rest of the 335 00:15:31,120 --> 00:15:35,519 year. 336 00:15:32,959 --> 00:15:38,000 All of them got this question right on 337 00:15:35,519 --> 00:15:39,839 their final exam though. I didn't I 338 00:15:38,000 --> 00:15:42,160 didn't discuss it again since then. We 339 00:15:39,839 --> 00:15:46,519 didn't revise it. Nobody asked me about 340 00:15:42,160 --> 00:15:46,519 it. They all got it right. 341 00:15:50,240 --> 00:15:54,240 But I wanted to emphasize the kind of 342 00:15:52,240 --> 00:15:56,720 approach that I take in these 343 00:15:54,240 --> 00:15:59,759 discussions really exploring the space 344 00:15:56,720 --> 00:16:02,399 and uncovering that context. But how do 345 00:15:59,759 --> 00:16:05,519 I achieve this? Am I just some kind of 346 00:16:02,399 --> 00:16:07,759 super genius who knows everything? No, I 347 00:16:05,519 --> 00:16:10,320 use Wikipedia loads. 348 00:16:07,759 --> 00:16:13,040 Um, I don't want to be too flippant 349 00:16:10,320 --> 00:16:15,759 here, but I cannot overstate how good 350 00:16:13,040 --> 00:16:17,440 Wikipedia is. 351 00:16:15,759 --> 00:16:20,480 Let's take a look at the Wikipedia page 352 00:16:17,440 --> 00:16:22,880 for audio file format. It's got a bunch 353 00:16:20,480 --> 00:16:26,000 of stuff here that seems well out of my 354 00:16:22,880 --> 00:16:29,360 depth, but if we scroll down or if we 355 00:16:26,000 --> 00:16:31,759 search the page, hey, it says sample 356 00:16:29,360 --> 00:16:33,199 rate. And I knew from the line in the 357 00:16:31,759 --> 00:16:35,600 textbook that that was the thing that we 358 00:16:33,199 --> 00:16:37,680 cared about. So, let's click that and go 359 00:16:35,600 --> 00:16:39,199 to that page. 360 00:16:37,680 --> 00:16:41,279 and we're going to check out this 361 00:16:39,199 --> 00:16:43,440 article and it goes into more detail 362 00:16:41,279 --> 00:16:45,839 than I did, but it covers a bunch of the 363 00:16:43,440 --> 00:16:47,360 same stuff. You can see there that graph 364 00:16:45,839 --> 00:16:49,920 not too dissimilar from the kind of 365 00:16:47,360 --> 00:16:52,320 graphs that I was drawing. 366 00:16:49,920 --> 00:16:54,959 So, with just two pages on Wikipedia, we 367 00:16:52,320 --> 00:16:58,399 can get a decent idea of what a sample 368 00:16:54,959 --> 00:17:00,480 rate is and then convey that explanation 369 00:16:58,399 --> 00:17:02,480 to our students. 370 00:17:00,480 --> 00:17:05,679 This is true for pretty much every 371 00:17:02,480 --> 00:17:09,760 technical topic. Um 372 00:17:05,679 --> 00:17:13,120 I I think largely because Wikipedia has 373 00:17:09,760 --> 00:17:17,039 come you know has come up in the age of 374 00:17:13,120 --> 00:17:19,120 computers fairly obviously. Um it has a 375 00:17:17,039 --> 00:17:20,559 lot of stuff about computers and about 376 00:17:19,120 --> 00:17:21,760 computer science and everything on it. 377 00:17:20,559 --> 00:17:23,839 So if you are teaching computer science 378 00:17:21,760 --> 00:17:25,760 at all it is an excellent source to 379 00:17:23,839 --> 00:17:27,679 learn about the context and the history 380 00:17:25,760 --> 00:17:29,919 and the why things were done the way 381 00:17:27,679 --> 00:17:33,559 they were. 382 00:17:29,919 --> 00:17:33,559 The other two big 383 00:17:37,679 --> 00:17:43,360 enough to make it real. 384 00:17:40,480 --> 00:17:48,360 A few little tidbits to tie it to the 385 00:17:43,360 --> 00:17:48,360 context is entirely sufficient. 386 00:17:49,360 --> 00:17:54,640 All that matters is that there is enough 387 00:17:52,799 --> 00:17:56,960 there for them to latch on to and go, 388 00:17:54,640 --> 00:17:59,600 "This is something that impacts the real 389 00:17:56,960 --> 00:18:02,720 world. This is something that a real 390 00:17:59,600 --> 00:18:04,400 human being put thought into. Maybe not 391 00:18:02,720 --> 00:18:07,280 a lot, but they put some thought into it 392 00:18:04,400 --> 00:18:10,799 and they made it and there were reasons 393 00:18:07,280 --> 00:18:12,960 behind what they did. 394 00:18:10,799 --> 00:18:15,440 The other big one, involve your 395 00:18:12,960 --> 00:18:18,640 students. You can give them research 396 00:18:15,440 --> 00:18:21,760 tasks or have an open discussion. I 397 00:18:18,640 --> 00:18:25,039 found asking why do you think they made 398 00:18:21,760 --> 00:18:29,039 it like that is a great way to get a 399 00:18:25,039 --> 00:18:30,320 discussion going. Um, this Oh no, I 400 00:18:29,039 --> 00:18:31,280 pressed a button because I dropped the 401 00:18:30,320 --> 00:18:32,480 thing. 402 00:18:31,280 --> 00:18:34,160 Why do you think they made 403 00:18:32,480 --> 00:18:35,840 Why do you think they made it like that? 404 00:18:34,160 --> 00:18:37,360 That is a good question. Why is there a 405 00:18:35,840 --> 00:18:41,600 button right under where your thumb sits 406 00:18:37,360 --> 00:18:44,320 that exits the whole presentation? 407 00:18:41,600 --> 00:18:47,679 Um, 408 00:18:44,320 --> 00:18:50,559 yes, maybe not a good reason. Um, 409 00:18:47,679 --> 00:18:54,000 so a an excellent example I had of this 410 00:18:50,559 --> 00:18:57,200 was um well I I found that this worked 411 00:18:54,000 --> 00:19:01,120 particularly well for topics that are 412 00:18:57,200 --> 00:19:03,679 maybe a little controversial. Not you've 413 00:19:01,120 --> 00:19:06,080 got to it pick the right ones but a 414 00:19:03,679 --> 00:19:07,760 little controversial or current or 415 00:19:06,080 --> 00:19:09,840 there's a lot of ambiguity floating 416 00:19:07,760 --> 00:19:13,600 around. We had a great discussion in 417 00:19:09,840 --> 00:19:15,600 class about NFTTS. I explained what they 418 00:19:13,600 --> 00:19:19,280 were, how they worked. I explained 419 00:19:15,600 --> 00:19:21,200 blockchain and how that worked and 420 00:19:19,280 --> 00:19:24,160 there was and I asked them, you know, 421 00:19:21,200 --> 00:19:26,640 why do why do you think NFTts work like 422 00:19:24,160 --> 00:19:29,919 this? And there was a pause and one of 423 00:19:26,640 --> 00:19:34,559 them said, it just seems like gambling. 424 00:19:29,919 --> 00:19:38,559 And I said, okay, go on. and she said, 425 00:19:34,559 --> 00:19:41,679 "Well, it just seems like the only way 426 00:19:38,559 --> 00:19:43,360 that NFTts make money is if somebody 427 00:19:41,679 --> 00:19:45,120 buys the one that you bought for more 428 00:19:43,360 --> 00:19:47,679 than you bought it. So, you're just 429 00:19:45,120 --> 00:19:49,120 hoping that somebody will. So, it's just 430 00:19:47,679 --> 00:19:51,600 gambling, right?" I'm like, "You're 431 00:19:49,120 --> 00:19:56,160 entirely correct. The whole thing is a 432 00:19:51,600 --> 00:19:57,919 scam. Maybe don't buy NFTts." Um, she 433 00:19:56,160 --> 00:20:00,400 later told me that her dad had been 434 00:19:57,919 --> 00:20:05,960 planning to buy NFTts and she told him 435 00:20:00,400 --> 00:20:05,960 and then he did not. So success. 436 00:20:07,760 --> 00:20:11,919 There are some caveats I do want to 437 00:20:09,919 --> 00:20:15,919 address here. 438 00:20:11,919 --> 00:20:19,919 Got applause from the other room too. 439 00:20:15,919 --> 00:20:21,440 Um firstly I had a rare environment uh 440 00:20:19,919 --> 00:20:23,760 that I understand that I was privileged 441 00:20:21,440 --> 00:20:26,320 to be in. I had small classes at most a 442 00:20:23,760 --> 00:20:28,880 dozen at a private school. And while 443 00:20:26,320 --> 00:20:30,320 none of these girls had prior digital 444 00:20:28,880 --> 00:20:32,480 technologies experience when they 445 00:20:30,320 --> 00:20:34,480 entered my class for the first time, 446 00:20:32,480 --> 00:20:37,120 they were girls who were expected to do 447 00:20:34,480 --> 00:20:39,919 well academically by their parents and 448 00:20:37,120 --> 00:20:41,919 by the school. I was also teaching the 449 00:20:39,919 --> 00:20:44,000 Cambridge curriculum and it is designed 450 00:20:41,919 --> 00:20:45,520 around the students getting more time 451 00:20:44,000 --> 00:20:48,080 per 452 00:20:45,520 --> 00:20:50,720 year. The school that I was at spread 453 00:20:48,080 --> 00:20:53,039 this time out over two years, but it 454 00:20:50,720 --> 00:20:55,360 probably only needed a year and a half. 455 00:20:53,039 --> 00:20:57,360 This afforded me some time to experiment 456 00:20:55,360 --> 00:20:59,360 with this approach that I knew that I 457 00:20:57,360 --> 00:21:01,679 could make up at the end if I needed to 458 00:20:59,360 --> 00:21:04,080 do just as much revision as I had 459 00:21:01,679 --> 00:21:05,840 initially thought. Luckily, I have done 460 00:21:04,080 --> 00:21:07,919 some of this experimentation for you and 461 00:21:05,840 --> 00:21:09,360 so you can try this out with a little 462 00:21:07,919 --> 00:21:11,280 more confidence that it will work 463 00:21:09,360 --> 00:21:14,159 hopefully. 464 00:21:11,280 --> 00:21:16,240 Um, I also studied computer science at 465 00:21:14,159 --> 00:21:17,919 university. I am a software developer by 466 00:21:16,240 --> 00:21:19,440 trade and have spent many years as a 467 00:21:17,919 --> 00:21:22,720 professional developer before I taught 468 00:21:19,440 --> 00:21:25,120 at school. And so I have had plenty of 469 00:21:22,720 --> 00:21:28,159 time to familiarize myself with a lot of 470 00:21:25,120 --> 00:21:30,320 this context before I got into the 471 00:21:28,159 --> 00:21:34,000 classroom. So I had to do less of that 472 00:21:30,320 --> 00:21:35,840 Wikipedia research now and I did that at 473 00:21:34,000 --> 00:21:39,679 university. 474 00:21:35,840 --> 00:21:41,440 Lastly, I have wicked ADHD. Um, and I 475 00:21:39,679 --> 00:21:44,799 absolutely live for going down a rabbit 476 00:21:41,440 --> 00:21:46,880 hole and learning how a thing works. 477 00:21:44,799 --> 00:21:49,360 um digging into the history or the 478 00:21:46,880 --> 00:21:52,320 politics or the physics behind something 479 00:21:49,360 --> 00:21:56,000 is when and where I thrive. I am told 480 00:21:52,320 --> 00:21:58,400 this is not for everyone. 481 00:21:56,000 --> 00:22:00,960 Um 482 00:21:58,400 --> 00:22:03,120 I want to focus that while focusing on 483 00:22:00,960 --> 00:22:05,039 the I want to emphasize that while 484 00:22:03,120 --> 00:22:07,679 focusing on the context worked for me, I 485 00:22:05,039 --> 00:22:09,600 understand it might not work for you. 486 00:22:07,679 --> 00:22:12,080 Still, I would encourage you to consider 487 00:22:09,600 --> 00:22:14,400 it and give it a try. I certainly saw 488 00:22:12,080 --> 00:22:16,320 results with it and not just because I'm 489 00:22:14,400 --> 00:22:18,480 a big softy and gave everyone great 490 00:22:16,320 --> 00:22:20,720 marks even if they didn't deserve them. 491 00:22:18,480 --> 00:22:22,400 Um, these students did the Cambridge 492 00:22:20,720 --> 00:22:24,880 exam which is run and assessed 493 00:22:22,400 --> 00:22:26,960 independently by people from Cambridge. 494 00:22:24,880 --> 00:22:29,840 Um, so I have some external validation 495 00:22:26,960 --> 00:22:31,360 here. I only had a single cohort that I 496 00:22:29,840 --> 00:22:33,200 taught in the staff for the whole two 497 00:22:31,360 --> 00:22:35,760 years of the course. So unfortunately my 498 00:22:33,200 --> 00:22:38,480 sample size is low but of the seven 499 00:22:35,760 --> 00:22:40,559 seven students who did that exam five of 500 00:22:38,480 --> 00:22:42,799 them scored an A star which is 501 00:22:40,559 --> 00:22:45,840 Cambridge's highest grade and one scored 502 00:22:42,799 --> 00:22:47,440 an A. These are girls who had never had 503 00:22:45,840 --> 00:22:50,400 a computer science lesson before our 504 00:22:47,440 --> 00:22:53,400 classes. They'd never 505 00:22:50,400 --> 00:22:53,400 elected 506 00:22:54,880 --> 00:23:02,480 on a curve. So to get an A you need to 507 00:22:57,679 --> 00:23:07,039 get 88 or above out of 150. These girls 508 00:23:02,480 --> 00:23:08,960 scored an average of 112.8. 509 00:23:07,039 --> 00:23:10,960 My supervisor said that the school had 510 00:23:08,960 --> 00:23:13,200 never achieved results like this before 511 00:23:10,960 --> 00:23:15,200 in computer science. 512 00:23:13,200 --> 00:23:17,679 It is not a silver bullet, however. And 513 00:23:15,200 --> 00:23:19,679 if a student is disengaged, no amount of 514 00:23:17,679 --> 00:23:21,280 context can fix that. You might have 515 00:23:19,679 --> 00:23:23,360 noticed I said I had seven students and 516 00:23:21,280 --> 00:23:25,520 then I gave you six grades. The last 517 00:23:23,360 --> 00:23:28,480 one, unfortunately, did not do as well 518 00:23:25,520 --> 00:23:30,080 and got an E. There is only so much that 519 00:23:28,480 --> 00:23:32,960 you can engage somebody who does not 520 00:23:30,080 --> 00:23:35,280 wish to be engaged. 521 00:23:32,960 --> 00:23:36,960 So all in all, I'd encourage you to try 522 00:23:35,280 --> 00:23:41,559 it. I would love to hear how you go with 523 00:23:36,960 --> 00:23:41,559 it. Thanks for coming to my talk.