>> David: Welcome, my name is David Malan and this is CS50. And, this is not your typical high school computer science course. This course is a little something special. And it's not because I'm teaching it. it's because this is one of, we like to think, those rare courses that actually squeezes your brain so much and your schedule so much that by semester's end you genuinely feel smarter. So, that was the way I felt when I myself took this course back in 1996. I was a government major at the time here living in Mather House. And, I. 

[ Cheering ] As are the other, the five others there. So, I was a government major at the time. And I was a government major because I'd always liked history. I liked constitutional law in high school. And so, I set off on this path freshmen year doing what I do best, doing what I knew I liked. And I'd always been a bit of a geek, I'd always babbled with computers. But, I myself, perhaps ironically in retrospect, I'd always viewed the guys who were taking computer science in my high school, my friends as the real geeks. And yet, finally did, sophomore year I take this little course called CS50 which actually had scared me away my own freshman year. And I was a user assistant at the time. So, I was certainly already qualified, at least, to be among those, those geeks who I made fun of in high school. But, I was scared. There is, it had this reputation at the time and perhaps still in part that it was a, it was a scary course. It was a lot of work. It was hard and yet, it's one of those things where, my God, the bark was far worse than its bite. I found myself sophomore year, for the first time, actually finding homework fun. And I'm not saying this just to pitch this course to you. this was a genuine realization for me. Now, I think, at one point, I did cross the line to true geekdum where I actually used to look forward to Friday night because I could curl up in front of my little MacBook and work on CS50's problem sets. So, at that point, I think I had clearly cross the line. And, our goal in this course is not to turn all of you into that person, but, just to say that there's a little something special about this course, a little something special about computer science in general especially these days. And, the more acclimated we've all got into technologies, the more toys we carry around in our pockets that are themselves computers. I mean, my iPhone is actually a higher performing computer than my laptop was in college. You can do a lot of neat things with these things. And now, most of us in this room, probably just use other people's software, other people's tools, other people's solutions to problems because we go to iTunes, we download some app from the app store and walla we can do something really neat. But, by the end of this semester, you can be the person doing that problem solving. You can be the person solving some problem, building that app, that piece of software, that tool that, either, one just amuses me, amuses you. for instance, controlling, say, your iTunes library from with your apartments clearly an app that already exists or you can find shuttle schedules more easily for instance. So in fact, one of the very first things I did after taking CS50 was again, it was probably at this point Friday nights and Saturday nights, there was this opportunity at the time where Harvard was running shuttle buses. And, to this day, they still have printed schedules. But, there was really no means online of finding out when the next shuttle was. Now, I didn't care, because even though I lived in Mather I wasn't among those who generally took the shuttle. But, I had a lot of friends in Pfoho for instance. And one of them asked me, I think in at one point. 

[ Cheering ] Okay Pfoho's better than Mather apparently. So, one of my friends kind of got me thinking maybe I can write a little program to kind of help my friends check the shuttle schedules. And thus was born the awkwardly named shuttle boy, this piece of lexicon that has stayed around for some number of years at least among some of the upperclassmen. At the time, this was this little command line program similar in spirit to Dos. It was running in a UNIX environment. So, it wasn't very sexy at all. You just typed a couple commands and it told you when the shuttles were. Well, these days it's a little more sophisticated. And, it's actually represented of precisely the types of projects that students in this course end up making by terms end. So, this is version four, some 15 years later, of Shuttleboy. Seems Google Maps is being a bit slow there. But, for those of you who have to get out of here after class and go, say, from Memorial Hall and need to go up to, say, the Quad. Well, you can either leave in two minutes if by, this really isn't working for you just yet, or you have as many other options as you might like. And, coming soon, we can pitch this from CS50, it will be the same capability by SMS. You can text Shuttleboy via phone. You call Shuttleboy. CS50's phone number this year, well, amazingly this was available, 617-BUG-CS50. So, that will debut soon. So don't call it just yet because I've gotten reprimanded for, right now it's being routed to our system head teaching fellow's cell phone. So, you'll just get him if you have questions about the shuttle schedule now. But, that will soon be implemented. So, other things, actually, that we ourselves spent this summer on really in preparation for this semester was a whole bunch of other problems that we wanted to solve. For instance, there's a whole lot of stuff that goes on in this campus on a given day. And in fact, put another way, there's a whole lot of stuff going on at this campus that I certainly never take advantage of. And, even though I'm now overwhelmed with how many things that are going on, on campus, I still seem never to find the time. But, walla, even there, CS50's lecture. That's where you is, so this was a, this was a program we wrote to aggregate all of the events on campus. And this has been something that other students in this same course have tackled on their own. And, we've working with other entities on campus to really make this happen. But the point to CS50 this term is that all of these little programs we've been working on whether this one or whether one for aggregating all of Harvard's news in one place, all the Google News, all of these things are going to have what are called API's, application programming interfaces. And this is just fancy speak for the capability of letting other people, people in this room, actually doing something with that same data. So, by terms end, if you decide to implement the final project related to news or events or maps, well we, at least as a course, will make that all the more easy this year by providing you with an interface; a sort of function, so to speak, software that you can use in your own projects to go above and beyond where students, perhaps, have gone in times past because this framework has not been in place for them. And again, this all began, for me at least, some 15 years ago implementing Shuttleboy. I then emailed my friend Ken in Pfoho. I said hey what do you think? Is this useful? He mailed it out on the Pfoho open list and walla, all these years later I'm apparently still milking this project. And the neat thing about it, and this is where too a course like this in engineering and computer science, I think, is a little special, is that, when you do your home works and when you do your projects in this course, you know, many of them, yes, you'll put aside and not really use again. But, some of it, you will. And there's this really powerful thing, at least for me, this really exciting thing, when you make something and other people actually care, other people don't just smile and pat your on the back and then sort of go about their own lives. They actually use what you made in their own lives even if it's for something as silly or as simple as just finding when they can get out of here to catch the next shuttle to the Quad. So, by terms end, that's one of the goals for this course; to empower you guys to just identify some problem whether it's on campus or hopefully well beyond campus, and actually use your own skills, your own mind and solve that problem in some part, technologically. So, this is also a course as we say in the syllabus and the courses official description that teaches you how to think and it teaches you how to solve problems more efficiently. Well, what does that mean? Well, when you've got 600 megahertz in your pocket or three gigahertz under your desk at home, you can solve a lot of problems efficiently just by shear brute force. But, there's a lot of problems in this world, whether it is searching, say, Facebook's network or the internet via Google, that it doesn't matter how fast your particular computer is, there's no way in hell you're going to search billions of web pages quickly unless you actually approach this problem intelligently. You don't just search, for instance, a huge list of all the web pages on earth from top to bottom looking for something related to whatever search terms you're looking for hopefully Google and presumably Google and Facebook and any of these large entities have a lot of smart people thinking about how they can find data more quickly, how they can process information more efficiently, and, how they can just generally return results more quickly so that you can do yet, more interesting things with these devices. And so, what do we mean, though, by how, teaching people how to think more carefully, more algorithmically. Well, this little visual always seems to go over well. And, it seems to be memorable. And I was even asked just yesterday by a former student, oh are you going to do the phonebook thing again. And I pretty much on the spot decided alright sure we'll build a phonebook thing. And he asked me when you tore it in half right, I was like well yeah technically we tore it in half but not in the way, in the computer science way will I be able to tear this thing in half. And so, here was the problem that we presented for some time. So, here's the phonebook. It's got at least a thousand or so pages. And the simple goal at hand very real world is to find, say, a person in here. So, Mike Smith, last name starting with S. So, I'm a typical human, I pick up this phone book. And anyone out there, what would you, a typical person, do to start finding Mike Smith, obviously not knowing in advance what page he is actually on. Right, so you go roughly in the middle, right. And at this point in the story, I'm probably in the N's or M's, you know, roughly half way through the phonebook. Turns out, the last time I did this example I somehow found myself in the escort section. So, it's actually not equally balanced between A through M and N through Z. But, today, we are, in fact, in the M's, alright so, now I'm at the M's. But what's my takeaway now just as a normal human off of the street? Where do I go next for Mike Smith. 

[ Inaudible] And so, just probably on this half, right, because S comes after M. And so, herein lies the visual drama. That's not really tearing it in half, right. I kind of cheated down the center. But, we now know that Mike is at least not in that half. We can literally throw half of the problem away. And I'm left with the problem that's fundamentally still the same thing, find Mike Smith in a really big book. But, the problem is now half as large. If it had a thousand pages before, now it's got 500. You know what, I can do the same thing again. I can kind of recursively or repeatedly do the same thing. Now I'm not quite at S. I'm at T. And so, oh I went a little too far. But, I know now that Mike is not to the right, there's going to be some class someday where I just can't tear the dam thing, I bet. But now, I know he's not to the right. So, now the problem has been quartered. So, I've gone from a thousand to 500 to 250 pages. And again, if you continue the logic, continue the mathematics, I'm chopping this problem in half, in half, in half until finally I'm either going to not find any Smith's at all, unlikely, or I'm going to find the one I'm looking for. But, that, then, begs the question, is this any better than the simple approach of just saying, you know, A, nope, B, no, starting from left to right going linearly through the book? Well, instinctively yes. It's going to be a lot faster. But, how much less? Well, if I have a thousand page phone book or let's say 1,024 for those of you who like powers of two, how many time might I have to split this problem in half before finding Mr. Smith? 

>> Ten. 

>> David: So, ten, right. If you have 1,024 pages and you split them in half, in half, in half, in half, I do that ten times which means I go from 1,000 pages to the person I'm looking for in just ten pages turns. And that's kind of neat, but, you know, if you think about it you've been doing this all of your life. It's not that dramatic. But now, that the phonebook isn't just for Boston, it's for the entire US or the entire world and this thing has billions of pages in it, so imagine a phonebook with four billion pages. How many page turns am I going to have to do maximally to find Mike Smith in a four billion page phonebook? 

>> Log base two. 

>> David: So yeah, if you're kind of the math type this is log base two. But, if you think four billion to two billion to one billion, so a half a billion. I mean, that actually goes whittles itself down pretty darn fast. In fact with 32 halvings of the phonebook even from four billions, I'll get down to one page. And that's when this stuff gets powerful, I think. That's when these ideas get compelling when you can have a four billion page problem and in 32 steps you can find the person you're looking for. And so, that's what we mean when we say that you'll learn how to think more carefully, more algorithmically, more intelligently about solving problems. And the returns are huge when you can actually do this, and again, to this point of the Googles of the world the Facebooks who have the luxury of having huge datasets. You've got to approach these problems in this way otherwise it's simply not going to work. Well, what's what did the student actually recall? I, we like to do things that are a little bit silly sometimes to sort of take the edge off of what is otherwise a lot of new material. And so, he actually, this former student sent me this link which I went ahead, just for kicks and then regretted it at the end, pulled up on uTube here. So, I thought I would share what a non-computer scientist means when he tears a phonebook in half. 

[ Laughter ] 

[ Silence ] 

[ Inaudible ] 

>> You know over 200 pages, nice clean line, legitimate tear. 

[ Laughter ] 

>> David: so, I suppose, his way works too. So, what I got out of this course all those years ago is yes, Shuttleboy, and then yes a degree in computer science, and then yes a PHD in computer science. And that's not what we're trying to turn most students in this course into. Most of you in this course are not going to go on an major in computer science. Some of you might minor, but for some of you this will be a terminal course. And that's perfectly fine because this course is, in fact, structured in a way where even if you just pay your dues for several months and you really focus in and you really take as much out of this course as you can, you will, in fact, be well equipped to go back to your world of economics, biology, psychology, applied math, and actually make use of what is ultimately as interesting a course as we hope to conceptually as it is fundamentally practical. And that too is something that's perhaps a bit rare to find. I thought I'd share a few comments that students from last fall put in this past years cue guide when asked about what would you like to tell future students. Well, if you haven't already looked at these comments, one student said this course will take a fair amount of your time up. but, if you put your work into a whole new world, a whole, if you put the work, if you put the work into, alright, so this is his grammar not mine, so if you put, put in the work, a whole new world will open up to you. Amazing how much I learnt in only 12 weeks. Another student wrote, you better take this or you're going to regret it. Another wrote, a little more bluntly, it will kick your ass and leave you feeling great. And then finally, this one is perhaps something I can only show in writing, but, second from the bottom here. 

[ Laughter ] So, I'm not sure if you have many friends who took this course but we kind of did a lot of damage to last year's students. We introduced, at the start, several lectures, a few quizzes, a couple problems sets, this notion of lolcats. You can't see what I can see here but in a moment you will see a young man named happy cat who is the leader of the so called lolcat. This is one of the stupid internet forwards that I think frankly the dichotomy between these two gentlemen here and what we now have on the screen is perhaps. 

[ Laughter ] One of the most inappropriate uses of this theater to date. 

And, why don't we make a little more real a little more personal this notion of actually solving a problem more efficiently. A simple problem on day one of any school is to sit down or stand up and start counting the students, right roll call of sorts. Well, the most naive but nonetheless, very correct approach that a teacher could take is one, two, three, four, five, six, seven, and we'd be here quite some time. So, what's a marginal improvement on this even that you might have done as a kid? How do you count people faster? And, don't start reading off my own instructions? What would you have done as a kid? Start counting twos, right, like so I'm pretty good at two, four, six, eight, ten, 12. So, that's something interesting right. So, how much of an improvement is that there? So, it's obviously a factor of two. It takes me half as much time. But, you know what, as we'll find throughout this semester, big deal, like speeding up your running time by just a factor of two is not all that interesting because the analog here in the computer world is that I'm a CPU I'm a computer, well heck, if I just wait 12 months, 18 months, I the computer will just be able by nature, by the number of gigahertz I have in me to be able to count people in this room faster. So, these kinds of, these kinds of linear speedups like factor of two factor of three, even factor of ten, who cares? Right, we can throw hardware at the problem. We can wait things out. We can develop faster machines. But, intellectually, yeah that was kind of clever but it didn't really fundamentally change the problem. I'm still going from left to right, front to back. It's really the same algorithm. But, I offer that we can do better. We can do smarter than this and do fundamentally a faster implementation. But, to demonstrate this, we need you to humor us for just a moment. And it's at this point where it'll be me talking less and you talking more. But I do need you to execute step one together, which is if you would humor us, please stand up. The directions here on the board, you know are, okay, a couple of you have to sit down. Those of you in back, there we go, alright. Thank you. So, the program here is before you. You are now the computers executing this program. It's fairly self explanatory at this point. Yes, you should be number one. Get it? Right, many of you are probably thinking this already anyway, first freshman year. So, think to yourself, you're the number one. So, step three here, pair off with someone standing, add your two numbers together and adopt that sum as your new number and we'll do a sanity check. 

[ Noises from crowd ] Do you know how many we have? 

>> [Inaudible] 

>> No [inaudible]. 

>> We need to have an answer key. 

[ Silence ] 

>> Yeah. 

[ Noises from crowd ] 

>> David: At this point the algorithm should be winding down. 

[ Noises from crowd ] Maybe we should have gone with my approach after all. 

>> 43. 

>> 163. 

>> David: Ooh arithmetic in front of 300 plus people. 

[ Laughter ] What's our final tally going to be? 

>> 180. 

[ Laughter ] 

>> 1,200. 

>> 386. 

>> David: 386, and are you the last one standing? Is that everyone? Do you know the answer? 

>> What? 

>> David: Do you know what the answer is? We didn't bring an answer key with us so, 386 is in fact precisely the right answer for the first time. 

[ Cheering ] 

[ Applause ] That's okay. 

[ Applause ] Very, very nicely done. So, what was the takeaway because that definitely felt a little slower than my very simple but nonetheless very correct approach of just going onesies and twosies and just getting it done with from left to right. But, what did we really just do there? Well, if there were roughly, let's say, 400 of you standing at one point, and then half of you, very quickly, from what I could tell, wanted to sit down. So, half of you sat down, we went from 400 to 200 then to 100 then to 50 then to 25. So again, there's that same divide and conquer idea we had with the phonebook. Again, with every iteration we halved the size of the problem, we halved it, we halved it. So, in theory, your algorithm should've been much faster than my linear approach because my approach would've taken say 400 steps or maybe if I'm slightly clever 200 steps. But, yours would've taken far fewer, right from 400 to 200, 100, 50, 25, 13, six; so fewer than ten steps in the end versus my 200 or 400. And there's the germ of a really interesting idea. And yes, mathematically what you just did was something that's a little logarithmic in nature by log base two, we divided in two and two again. And we'll revisit this later in this semester. But therein lies again a very intelligent implementation. And, we are just scratching the surface of the very clever the very intelligent efficient approaches to problems that we might ultimately solve. So, who are the people you just chatted with if briefly? A common concern in a course like this is that everyone to the left of you and everyone to the right of you must clearly know more than you. Well, that's absolutely not the case. And in fact, in recent years, this course has made a concerted effort to reach out to those less comfortable, so to speak. And by that, we mean not people who've never touched a computer before because that's really hard to find around here these days, but folks who use email, use Facebook, maybe even right now but are not necessarily quite sure what to do when something goes wrong or when they are asked about anything technological, you know, the hands go up or they kind of push the hardware to a friend who's perhaps more versed in it than they. Well, these are some statistics from last year's semester. So, when asked, when we asked our 400 so students last year, you know, what is your level of comfort coming into this class, well 34% wrote themselves down as among those less comfortable. So, if you have come into this room and are still thinking now, that, you know what, this term less comfortable describes me, know that you are in very good company. But, realize too that, if you're coming in this course with a background in computer science and you were really that geek in middle school and high school who was writing code and teaching yourselves languages, well that's fine too because there's definitely a nontrivial sized audience with that background as well, about 13%, last term. And then, there's everybody else, sort of someone who doesn't think they're less comfortable, they definitely don't think they're more comfortable but they're just in a course to get something out of it. So, realize this course does draw quite the spectrum of students. And, to put this in more concrete terms, 71% of students, last year, had zero courses behind them. So, if you're entering this course thinking oh everyone who takes CS50 was a computer science whiz in high school, it's not the case. Most people are presumably just interested in it. And a few, yes, absolutely have taken some courses, some, or at least one person last year had taken five courses previously. So, we definitely have some outliers but again the meat of the curve is in those lower values altogether. And another interesting statistic we shared over the past couple of years is the gender ratio. So, historically, at least in computer science, there is a bit of a male female skew. I couple of years ago it was much more marked than last year. So, we're almost now at the %40 level. And if I can throw down a gauntlet just for the sake of having a third pie chart next year, that's even more equitable. I would point out that there's absolutely no reason from the courses perspective that males and females shouldn't be enrolling in this kind of course. In relatively equal numbers and realize to 38 percent, 48 percent, we're actually quite pleased to see that we're almost at that, at least, statistically even value of 50%. So, if you are, in fact, female, do rest assure too that this is not, in fact, a male dominated class. Well, this just so happens to be CS50's 20th birthday. I, well thank you. 

[ Cheering ] 

[ Applause ] That's very nice of you because we brought cake for you actually. And so, when you exit here later today, you'll actually get a chance to meet many of the courses teaching fellows who'll be lingering in the hallway. And yes it'll be a nice sort of incentivized way of meeting the teaching fellows. But, we do take a trip up to a little bakery and again consistent with the course is playful nature while nonetheless is expecting nontrivial amount of work while we're having that fun. But, we do have one of these cakes for you. 

[ Laughter ] So, if you're laughing that's actually good because even if you're among one of those less comfortable at least you're aware of it. Those of you in the 14 percent might prefer this flavor. 

[ Laughter ] And then, if you don't know what the hell is going on right now, we also have this for the rest of you. 

[ Cheering ] 

[ Applause ] So, those are being chopped up into pieces outside as we speak. So again, let's take things up a notch and at least introduce some basic tenant that we'll then take for granted in the remaining of this semester. So, we all probably know that computers ultimately boil down somehow to zeros and ones, these things called bits. What does that mean? Well you can think of it frankly as a light bulb, a bit is a light bulb that's either on or it's off. It's a switch, it's either on or it's off. Now, computer scientists generally think of off as the number zero arbitrarily and the, and the notion of on as one. So, if you throw a light switch on that's one turn it off that's a zero. Or, equivalently, turn the light on that's the value of true turn it off that's the value of false. So, these are just synonyms. At the end of the day, yes, all computation, all stuff that goes on underneath the hood in a computer does ultimately boil down to zeros and ones. And, when you stop to think about it, it's pretty dam remarkable what the world has done with these basic building blocks including some of the devices you're holding in your hand. Well, why are zeros and ones useful? My God, even in the real world we at least have ten digits, zero through nine. Why so limited? Well in hardware terms it's very easy. It's much easier to represent two states, right. If computers are plugged into the wall, usually, you have this easy source of binary nature. It's either on or off. Electricity is flowing or it's not. So, that's kind of one of the initial reasons that the world went with just zeros and ones in computers. But, you can use them as building blocks. So, this actually is a little small on the screen here but we can count in binary or we can represent anything we want in binary just as we can, say, decimal, just as we can, even in fact, with alphabetical letters. And so, here's the quick system. In just a couple minutes time you'll now know how to count, how to write, how to speak binary much like your own laptop. Well, you have, in the world of binary, sort of just like grade school columns or values. So, in grade school we learn this is the ones column this is the tens column, the hundreds column and so forth. In other words, why is the number we know as 123 the number 123? Well, my teacher taught me this is the ones column, this was the tens, this was the hundreds. Well what did that mean? Well, to figure out what the total value is of what is otherwise the string of digits, just hash marks on the board. Well, we have one 100, so that's 1 times 100 and then plus what two 10's so that's 2 times 10 plus three ones, so that's 3 times 1, okay, so I seem to have gotten back to where I started. But, this very simple process of just multiplying the digits by the value of the column is exactly what binary is all about. But, in binary you use powers of two not powers of ten. So, here's a series of eight light bulbs. Computer scientists like to use values like two and eight and 16 and 24 and 32 because they're all generally convenient values to use. So, when we talk about eight values in sequence this is like called the byte. And these days we rarely talk in just bytes, we talk megabytes gigabytes and terabytes these days. But it's all ultimately the same thing. Here's eight bits AKA one byte, just simple jargon. So, all of these light bulbs are off. So, the math is actually really easy. So there on the right, though it's small to see, is the one's column, the twos, the fours, the eights, the 16s, 32s, 64s, 128s. Alright, maybe it's hard to memorize at first but it's just multiplied by two from right to left. So, you can always recreate this. Well, they're all off so that's 0 times placeholder, 0 times placeholder. So, this is the number 0 represented in binary. In other words, how do you write the number we know as 0 in binary? Well, one, two, three, four, five, six, seven, eight, and technically just like in decimal, all of these leading zeros, they don't add anything to the formula they're just there because we arbitrarily decided we want eight of these things in a row, right. If they're physical objects, we have eight so they've got to all be off. So, we could've just gotten away with one zero. Well how do you represent the number one? Well, instinctively you're going to throw one of the switches. You're going to turn a bulb on. So, there is the number one because we now put a one, a light bulb that's on, in the one's position. How are we going to represent two? Well, we're actually going to throw two switches now at least versus this state, we have to turn on the two's column turn off the ones column and now we have the value of one in the twos column. So, that's the total value of two. Now, a quick mental exercise; what's going to happen with the next sequence? How do I present the number three? Well, I need one from the twos column, one from the threes column and ah ha. So, now I've incremented that one. And now, this same basic pattern repeats. What's the number four? Well let's see, I just need a one in the fours column and then zeros in the twos and ones. So then, it continues that way. And, it's just this sequence of steps. And in the notion of binary is really not all that complicated. At the top there we have eight zeros. Below that we have seven zeros and a one. Below that we have six zeros, a one, and a zero and that's binary. Now, thankfully, we computer scientists, we programmers by terms end are not going to often be writing in binary, talking in binary, thinking in binary because we've built many layers of the traction on top of that. In fact, people rarely talk about bits except in a networking context they talk about bytes because one bit, not that useful, eight bits these days still not that useful, but at least eight bits, one byte is at least getting us toward more useful units of measure. But, when we start looking underneath the hood, when we do certain problems such as look at files and look perhaps at network traffic, it will become interesting to know how you're representing data because you're going to have to figure out how to display that same information. And, in fact, we, if we only have zeros and ones, how do we go about representing, say alphabetical letters? Well, we're going to have to have some mapping between binary and letters. But that's as of yet unclear. But to hammer this point home and to give me a break from speaking at, I thought I'd pause for a moment and see if we can't recruit eight volunteers, one byte eight people. Can I get eight people for whom it is not too awkward to come on up. good, good one, two, three, four, five, six, seven, eight very enthusiastic eight, come on down. So, you all, for just a moment here, oh and you have to be willing to not be on the internet because there is in fact a camera in the back of this room. So, turn back now if you don't want to be on the, hello mom, in perpetuity. So, let's see, you here will be the 128th column. And let me ask that you simply hold this before you. And actually, if you could scooch this way just a little bit, then I will get out of your way. You'll be the 64's column, you shall be the 32's column and the 16's. 32's, 16s, 8's, 4's, 2's and 1's. All right. So perfectly lined up with the place holders from before, but we need a little bit of mapping, so a quick set up. We just need an arbitrary way of mapping binary numbers, which really could be converted very easily to decimal numbers. So we need a way of converting numbers to letters if our computers are actually going to show us emails or type URL's or actually display non numeric information. And, in fact, there is a system for that, it's called ASCII, an acronym whose meaning is not so interesting, but whose utility is. This is just a mapping from letters to numbers and there's actually a lot more numbers on the screen, and you guys, don't worry, aren't going to have to test yourselves with this in just a moment. But there's a mapping between numbers and letters and I'm just going to get us started by pointing out a couple of these. The numbers of interest right now are going to be, whoops, are going to be this. Whoops. So in this column here, I can't seem to zoom in simultaneously, we have decimal. Dec for decimal. And then we have the character on the right hand side. So if I hop over into this column, thought it's a bit small on the screen, the number 65 represents what letter? Capital A. And the number 97? Lowercase A. All right. So those two alone, right, so do not go home and try to remember ASCII chart. Come quiz time, do not try memorizing an ASCII chart. Memorize two, 65 is capital A and 97 is lowercase A. And from there you can bootstrap yourself to all of the other alphabetical characters. But all this chart says, and I just downloaded this chart from ASCIItable.com, this is a standard that's been around for many, many, many years, but most computers still use it. What we now have is a means of encoding alphabetical information using what's ultimately numeric. So if these guys are just switches, light bulbs, computers, we can actually convey messages. So each of you has some hints on the back. We're going to have three rounds here where these guys are going to spell out in English, in ASCII, a very short message, a three letter word. It will involve them either raising their hand, thereby representing a one or just standing there awkwardly, thereby representing a zero. So why don't we proceed immediately with round one and you can either raise or not raise your hands. So the audience doesn't get off the hook. What number are they currently representing? 66. Right. Not hard because we have a one, a hand up in the 64s column, a hand up in the 2's column. 64 times one plus two times one is 66. They're representing the number 6, which means the first letter is B. So not too bad, right? Round two. First letter was a B. 

[ Laughter ] 

>> Seventy Nine. 

>> David: Can someone second that? Yes. All right. So 79, which means it's now an O. So B,O. There's a third letter. Actually that is kind of, yeah. Third round. Third and final round. What number is this? I heard an 87. Someone else agree? All right, yeah. This is in fact 87, which maps to, and this is a very clever set up at which point you all B, O, W, it's bow. Very good. So actually besides that little souvenir, that was not a very enthusiastic applause, so I'll take care of you. So we do have a little something for these guys here now that they just appeared on center stage doing perhaps the geekiest thing they could. So maybe if we could humor them if only because I'm feeling a little bit awkward now. A little round of applause for these guys here. 

[ Applause ] All yours. All right. Thank you very much. So, so, it turns out that we can take this one step further, but you are now entering a community of people who understand things like this. 

[ Silence ] It's slowly spreading, yes? 

[ Laughter ] It's not cool that you understand this now. So we've already done some damage here today, but yes, there is now such a thing as geek humor that you might just catch a little bit more on to. So what are we going to do throughout this semester? So there's 12 weeks, zero index, so 13 weeks in total. You'll find throughout the semester that computer scientists tend to count starting from zero for reasons that will soon become clear. But what we're going to do this week, starting this Friday, is introduce you to programming. Now you may already be coming at this course with a background in the same, and that's perfectly fine, but we'll also be able to introduce some more sophisticated topics than are typically covered, things like events and threading, so just to toss out a couple of buzz words for those unfamiliar. But what we're going to do is not dive into what the typically traditional language like Java or C++ or C sharp catch phrases that you might have heard in the past, but we're going to use a programming environment, a graphical programming environment that was actually developed by MIT's media lab. And this is an environment that allows you to program, quite literally, by dragging and dropping puzzle pieces that fit together logically that lets you implement these things called loops and conditions and Boolean expressions and you can actually, on day one, after this Friday, write a program that actually some of your friends might care about. So in yester year, say when you've taken a computer science course in high school, odds are one of your first programs was the equivalent of hello world, like wow mom, I just made the computer type some characters on the screen. It's a very underwhelming experience. Much more interesting is it to implement a little something like this, as one of last year's students did. So I give you the work of a former classmate. It's a love story, of sorts, about dancing cookies. 

[ Music ] So this was actually from a student who had never programmed before. And just to give you a teaser of this interface, and we're only going to use this for one week time, really to get the comfort levels up and to really get folks excited about what you can actually make these machines do, this is an environment where you'll be programming by, for at least a week, just dragging and dropping puzzle pieces that fit together, as we'll see, in a logical fashion. And it lets you do fun, silly things quickly. And for any of you who are a little bit familiar with things like Java and C++, it's usually a good semester or two or three before you finally get to start writing something that another person might actually enjoy watching. Well, in week one of the course, or the second week, zero index, where we dive into a little more traditional environment. In this course we'll spend much of our time on a language called C. This is one of the earliest languages that, unlike Java, actually gives us much greater access for better or for worse, to manipulating hardware and actual memory, and we can do a lot of damage with that. In fact, any time you read of web servers being exploited or computers being hacked into, very often to this day it's the result of people having run software in a language called C or C++ without sufficient care given to what are called buffer overruns or lengths of arrays, so some fairly technical concepts that are not that hard to understand, but there's been a lot of humans making a lot of mistakes for years and with this language we'll actually understand some of those security implications. But each week, typically, we'll introduce also an interesting domain that will inspire that week's problem set. So we'll also take a look this week at the field of cryptology, cryptanalysis, the notion that you can encipher information in such a way that people can't read your messages, but if clever enough perhaps you can, in fact, crack what are otherwise secret messages. In week three we'll then look at some basic fundamentals of computer science like searching and sorting and we'll actually look at this notion of recursion, how to debug things. In short, we'll take things up to the level of problem solving and not so much about mere syntax and implementing short little cookie cutter programs. In week four we'll introduce more sophisticated problems still. What does it mean to manage memory? What does it mean to run out of memory? What does it mean to actually segfault, if you've heard this expression, and odds are you haven't, but very bad things happen and frankly by terms end you'll understand terms like my code segfaulted which actually is generally not going to be a good thing. But then again, we'll introduce some domain specific stuff. So actually one of my internships during graduate school was to work for the local Middlesex County District Attorney's office doing forensic investigations in a digital sense. The local Mass state police would bring to our office hard drives and floppy disks and sometimes for no reason mice and keyboards. And we would be tasked with finding evidence. Now it turns out, maybe just an FYI, that the criminals in Cambridge, Massachusetts, at least, not necessarily that clever. Very often our forensic analysis boiled down to booting the computer up and double clicking on the My Documents folder, and therein laid the evidence. But what we'll do this week is give you a sense of what it actually means to store information on a hard drive, on a flash drive, and why can't otherwise private information be disclosed to third parties when you drop off a computer to be repaired or even when you discard it. Well someone might very well be sifting through those hard drives and it's really not that hard to find data if you know how to look for it. Well, in week five we'll look at more sophisticated data structures, tools and techniques that you can use ala phone books and ala searching huge data sets much more intelligently. And the week's problem set will task you competitively to see just how fast the code you can write is and how little memory you can use. Toward terms end though we'll take things up to a higher level in terms of programming languages and put C behind us and introduce a web programming language known as PHP, one of the popular languages these days. Much of Facebook is still written in PHP, at least outward facing, and it's just a very easy language to learn quickly once you already know another. So with PHP we'll be able to solve problems and implement websites that do interesting things. The news site I showed you, the events site, the shuttle boy site, all of those all happen to be written in PHP. But by no means is that prerequisite. What you'll also find throughout this course is that you're not going to exit this course knowing C. You're not going to exit this course knowing PHP. Those are not interesting skills to claim, but knowing how to program, knowing how to use data structures and algorithms in an interesting way, that's a compelling story. Even I, all these years later, throughout college and graduate school, the only time I ever learned a language was probably in CS50 where it was still taught in C and CS51, which is one of the successor courses. Everywhere along the way since has been go learn this language and then work on this problem set because computer science is not supposed to be about here's a language, let's spend seven weeks teaching you where the semi colons are. There are much more interesting ideas out there. Databases, for one. And in week two we'll introduce you to this query language called SQL that'll allow you to store information in databases and actually implement some fairly interesting final projects. The week after we'll look at what's called client side programming, doing interesting things on a user's computer, taking advantage of their CPU cycles, doing things ala Google maps and these websites that actually use JavaScript, which is actually a language we'll introduce to let you do client side things, make more interesting, more sophisticated user interfaces, but at the end of the day have to be really mindful when implemented of a lot of sticky issues like network speeds, network performance and security. So it's not about again just coding something up that's fun to use or pretty to look at, but actually thinking about the design that'll be the interesting process. And then, the end result aesthetically will often be a really neat toy to show yourself or a friend. And then we'll look at hardware. We're going to take for granted that there are these physical boxes throughout much of the semester that will do what we tell them to do. But how are they doing that. So we'll actually take a peak under the hood and talk about what a compiler really is, what a CPU really is and actually consider what's making all of this fanciness from stupid YouTube videos on down possible. So what are the expectations? Attending lectures and sessions, completing nine problem sets assigned roughly weekly, two quizzes, the final project and there is in fact no final exam. Grades, it's always a question, broken down as follows here. You can take the course pass-fail. But let me say it here, and I've said it many times. I personally care relatively little about grades. I think they are a wonderful, well, they are a useful pedagogical tool with which to motivate and unfortunately sometimes de-motivate, but you'll find in this course that at the end of the day it's more about seeing how far you can go from weeks zero to week 12, as we say in the syllabus than it is about absolutely ranking all students in this course. So if you're coming into this course among those less comfortable, though a large demographic do not assume that that's automatically putting you in the lower third of the so called distribution. At terms end do I sit down with the TF's and discuss each and every one of the students. We take into account where a student came from, what kind of trajectory they were actually on and so at the end of the day please focus more on, and I don't mean to sound too grand, the process of learning and less on the numeric's. And you'll actually see that we grade problem sets, in particular, very coarsely, whereby along three axis, you will get a score of 1, 2, 3, 4, or 5, where 3, is in fact good. 3 is not C. 3 is good, 4 is better, 5 is best. But more on that as the problem sets are released. So the course's website is where you will find, as in most cases, a whole lot of useful things, if not silly things, too. And we have the luxury, as computer scientists of implementing it ourselves. So over the course of the semester you will see various new features rolling out. You'll find out over the course of the semester that the bulletin board is a wonderful place, certainly in the course of this size, to turn when you have questions about the problem set, questions about some computer science concept. You'll be able to check your grades, download all handouts in advance of lecture typically will I post the slides, if you're the type who wants to scribble on notes. But as you've seen, usually I shy away from long lists of details and more like pictures of cats and other visual cues. So you'll find, too, that we will provide printouts of code when we actually think it particularly advantageous to write right on the paper. Software, videos will be available, usually around 1 to 2 to 3 days maximally after each lecture. So the courses are being filmed and every lecture will go online in various formats including QuickTime and MP3's for downloads to iPods and such. And also you will find a link to download what we call the virtual terminal room, which as you'll see is related to virtual office hours. Books are expensive and in the computer world also not that useful. They are on the shelves of the coop, although we're told that last year they ordered like 120 books for the course and 30 were actually purchased, but this perhaps speak to the fact that there are places like Amazon.com, so if you are the type who actually enjoys the sort of support structure that a book provides, if you like to read up more than say a course's typical materials offer, by all means, procure one or more of the books that we recommend here and discuss more in detail in the syllabus. But realize there are so many resources that this course provides in printed form and electronic form, and my God, there's the whole internet out there. Books are not something that are required for this particular course. Lectures, in general, will be Monday's and Wednesday's, except for this Friday and next Friday, simply because of shopping period and Labor Day. So we'll meet this Friday, next Wednesday, next Friday, but there after this is just a Monday, Wednesday course. So this little visual cue here is to remind me that yes, we do distribute videos in QuickTime formats, MP3's, fun ways that you can, if you're actually this type, maybe in the 14 percent, watch a computer science lecture on your iPod, but it reminded me that at terms end last year, the course culminates in final projects, in which students can design most anything they want, and we actually had some students tackle some iPhone applications. Some one of these at least is now in the Apple store. It's called Rover. It's been adopted essentially by Harvard agencies and it's the iPhone version of the unofficial guide to life at Harvard. They did a brilliant job. I bring this slide up too, just to point out what other technologies might very well be down the road. Some of you may have seen this before, but just to give you a teaser of apparently what's in the pipeline, I thought I would offer this brief distraction here. 

[ Music ] 

>> With the MacBook wheel, Apple has replaced the keyboard with a sleek, touch sensitive click wheel. Apple CEO Steve Jobs introduced the product yesterday at the annual Mac Expo. Senior product innovator says the MacBook wheel will make typing a thing of the past. 

>> At Apple our philosophy is create products that are simple to use and nothing's more simple than a single giant button. 

>> Gilman showed me how the revolutionary new computer works. 

>> Gilman: Just open the intuit-type alphabet menu here, scroll to the letter you need and center click to select it and click again to capitalize and repeat this process for each new letter. It couldn't be simpler. You could also let the predictive sentence technology complete the sentence for you. 

>> Gilman says the MacBook wheel has also simplified organization of files, so searching your hard drive will be a snap. 

>> Gilman: Just press both sides of the wheel concurrently and center click and there, you have an alphabetical list of every file on your hard drive. Everything is just a few hundred clicks away. 

>> Apple is calling the MacBook wheel the most intuitive product every designed. 

>> Here at Apple we like to think that we're giving customers features they don't even know they want yet. 

>> While the MacBook wheel won't hit the shelves for another 3 to 15 months, many apple users already have it on their wish list. 

>> I'll buy almost anything [inaudible] and made by Apple. 

>> Alex [inaudible] was one of the lucky few to get to try out a MacWheel and spent 45 minutes typing an email to his friend. 

>> I never really realized how much I hated keyboards until I saw this thing. 

>> I like how it automatically says, sent with a MacBook Wheel, so people automatically know you have one. 

>> With a price tag of just under 2600 dollars for the lowest end MacBook Wheel, it is an investment. But the super thin laptop features numerous innovations, like the new ultra thin hummingbird battery, which can power the MacBook Wheel for a fill 19 minutes before needing to be recharged and the computer is virtually unbreakable unless dropped or hit. But Apple isn't resting on its laurels, Bryan Gilman says they're already hard at work on the next generation of the MacBook Wheel, which will be four ounces lighter due to its lack of streaming hard drive or wheel. For the Ugly news Network, I'm Jeff Tate. 

>> Thank you for that, Jeff. It remains to be seen if the wheel will catch on in the business world where people use computers for actual work and not just dicking around. Moving on, police warn. 

>> David: So I've seen that thing at least a half a dozen times now, including once at like two a.m. last night and I was scrolling down on YouTube and literally like an hour prior, some guy in all seriousness posted, just a wheel? That's the stupidest product I've ever seen. 

[ Laughter ] And then the rants ensued in the thread thereafter. So besides lectures there are also sections and it is your teaching fellow and the course's course assistants that you will get to know very well over the course of this semester. Sectioning won't begin for a week or so's time. More on that in the lectures to come. But know that the course will have several different tracks of sections, much like First Knight's has for music. Sections targeted more at those less comfortable, those more comfortable and those somewhere in between. And we also offer most Sunday nights before a problem set is due on the following Friday is what we call code walkthrough's where one of the course's teaching fellows will lead a session targeted specifically at the week's problem set and will help you answer the question where do I begin, how should I approach this? And that too will be filmed and made available online and always linked on the course's website. So this course has a huge staff. Our staff is larger than Harvard's average class size. We have over 50 people involved in the production of this course on a weekly basis. Some of them teaching fellows who hold sections, grade work, hold office hours and more. Course assistants who are former students, who are offering on a volunteer basis in fact multiple hours of office hours in the computer lab per week. System administrators, folks doing all of our AV and video production. Someone who will actually be taking notes for you. So it is actually my opinion that scribbling down almost every word that someone like me says is not a very good use of one's time. It's much more interesting I think to actually engage with what's on the screen, what's being presented. And so what we will do as a course is post PDF's online within a couple of days of lectures of scribe notes, which is essentially an authoritative version of course notes, unedited, but documenting what went on in lecture that day, thereby hopefully freeing your minds and pencils from jotting information down. We have some of our team here. Most of our team is actually shopping courses. But if a few of your guys would like to come on stage for a moment with our head teaching fellow and assistant head teaching fellow. Allow me to ask the latter to say a quick hello. And as these guys come up in absentia allow me to say that also on staff this year besides the folks you see here are Alex, Alice, Andrew, Batul, Charlotte, Chris, Dan, Dan, Daniel, David, David, David, Derek, Dev, Didi, Doug, Drew, Erica, Fil, Glen, Greg, Janice, Jason, Jean, Jeff, Jesse, John, John, Jonathan, Jordan, Josh, Karim, Kelly, Ken, Kent, Kyu Bok, Lee, Madhura, Marta, Matthew, Michael, Michelle, Mike, Mike, Nathan, Patrick, Peter, Peyton, Rose, Saba, Sanjay, Shannon, Stacey, Ted, Wellie, Yair and me. So our course is teaching fellows and course assistants, Cansu and Yuhki, would you like to say a quick hello? 

>> Hi. It's definitely a great honor to be back here for the third time. And if you're one of those who are like less comfortable. 

[ Laughter ] I think I can [inaudible]. Okay. 

[ Laughter ] If you're one of those [inaudible]. There's no need to be afraid. I walked in here two years ago with no experience what so ever, and I thought that I was like dumb and I was sure that I was going to fail the class. And I came back for the third time. I did not fail the class. This course is great because it has such a good community. As you can tell by the names listed by David. All the CA's and TF's are here to help you. So if you have any questions at all, just come up to us. Not taking this class because it's hard is not a good excuse. So, hi. 

>> [Inaudible]. 

>> Hi, so well both as a student and TF, I've really enjoyed being a part of this course. And even though it's a course that requires a lot of work from you, we've put a lot of work in to it for you as well to make it as enjoyable and as fun as possible. So, we really hope that all of you stay and take part in this very special experience. 

>> Thanks. 

>> David: So you will find, thank you, all of you, you will meet them in the hallway in just a few minutes as well. So you'll find that this course, perhaps more so than many is very much a shared experience. This is not a competitive course. This is not your premed course. This is a course which ultimately yes is about doing independent work, as you'll see each week by way of the problem sets. But you can't help but to get close to someone when you're fighting with some bug and banging your head against the computer screen proverbially to try to track down some bug. And I think you'll find that past students will concur that you exit this course with a lot of friends among students and also on staff. So where does that happen? Well, within office hours in the basement of the science center there's some large computer labs and we pretty much take them over five nights a week. These won't start for some time, and this, to be honest, is perhaps the most beautiful picture of the science center I've ever seen. 

[ Laughter ] It does not look like that on the inside. But it is a place where a good percentage of students do choose to spend time to get one on one assistance with each of the week's problem sets. Each week we will offer more than 100 hours of assistance one on one in the science center per the schedule that will eventually go online. And for those of you who live, say in the quad, or in Mather, or who just don't want to trek all the way to the science center to ask that quick question, we also offer what are called virtual office hours, much like remote technical support where you'll be able to log into a glorified chat room, in which yes, you can type back and forth with a TF or a CA, but you can also share your screen or share control of your mouse with them subject to your consent so that they can help you troubleshoot any problem or any question you have remotely. To give you a sense of what's involved in the course, these are statistics from last year. So you will, I think it's quite fair to say, work your ass off in this course. But per my promise from the start, this is one of those experiences that I don't think a typical student regrets. If you flip through, for instance, the Q Guide forms that are all on the course's website, you'll find that only by putting in the time to a course like this do you really exit having felt like you got something out of it and frankly might very well even have emerged even smarter. So what does it mean to actually write programs and implement problem sets? Well, the first problem set last year, a quick retrospective here, was to implement your very own programming scratch. And we spend just the first week there, and the goal is to implement your own game or animation or art or anything that appeals to you. In week one, then do you implement some basic C programs to get acclimated to some more traditional programming environments at the command line, at Linux so you get comfortable with some of these basics that will certainly serve you well in the long run. Last year's Problem Set 2 was about cryptography. We gave students the opportunity to implement some ciphers, some algorithms that let them encrypt information. And we also offered some students the opportunity to do the opposite. Each problem set, many problem sets, as you'll see are implemented in two editions. A standard edition that we expect almost everyone to do, say 90 percent of the class. And then a hacker edition, which presents the material from a more sophisticated angle, expects perhaps a bit more clever work for those more experienced, and for this one did those students get handed in the hacker edition pre existing user names and passwords in encrypted form and they had to figure out what those passwords for. We actually got into trouble for this a few years ago because all of the sudden FAS Computer Services saw like 50 people running a crack program on nice.fas.Harvard.edu, so they now have been given a heads up for programs like that. Last year students implemented the game of 15. Not all that much fun to play as a party favorite, moving the numbers up down, left, right, but a real stepping stone to implementing something interactive that involves a human and the hacker edition for this one had students implementing God mode, which actually solved this thing for the human, maybe cheat mode is the more apt term, in Sudoku, which was a step up toward that, where we provided the students, you, with a framework with a whole bunch of code that you needed to first understand before taking things the final mile and implementing some of the remaining functionality. And then we dabbled in forensics and this grew out of a real world experience between me and a friend. For this problem set last year, this buddy and I, who he being a particularly good photographer, strolled around campus and we shot photographs of unique, non obvious, but identifiable places on campus. And this was using his digital camera. All of these JPEGs got stored on a digital flash card or what not. And then I accidentally formatting the thing, or erased it. Well, it turns out formatting something or erasing something, usually means neither. It just means doing a little bit of work to make the human think their data has been erased, when in reality it hasn't. It's just been forgotten. And so we tasked last year's students with writing a program in C to actually recover these photos for my friend. We gave everyone a forensic image, so to speak, a big file representing that compact flash card, and they had to recover the several dozen photographs. And they also had to solve this little murder mystery. We provided folks with this image here. Those of you who've seen this before might be doing that squinting thing right now where you're trying to look through the picture, well you look pretty awkward, because this is not a magic eye puzzle. This is just random red noise that actually has behind it an actual message. And it turns out, I think it was Colonel, let's see, I've got my little solution here. Last year it was Colonel Mustard in the study with the candlestick. And this was the murder message recovered by students who wrote code to do so. And then in Problem Set 6 last year. We challenged students with implementing the fastest spell checker possible. What do we mean by that? Well, we handed the students a big text file containing 140 plus thousand English words, and we said write a program that loads all of these words into memory, but very quickly and very efficiently can tell us yea or nay, is this document perfectly spelled or are there errors? And what we provided was a testing framework, a harness, that students, if they wanted to compete, could run their code against some benchmarks, against some testing programs that we wrote. And then, ranked on the course's website from fastest to slowest were the running times, the amount of time it took for each student's code to actually execute, complete opt in. It's by no means required. And we took a bullet for the team and made sure that one of our own teaching fellows intentionally, I think, was at the very bottom of the list. So, there was definitely a lower bound. But the odd, the funny thing was we saw this in the cue guide evaluation, this problem set of all things, was apparently the one that people sunk the most time into. Not because it took any longer than all the other p sets, but they would come back from like dinner, look at the board, and realize dam my friend, my roommate just ever so slightly inch higher than me and so it incentivized people if only to one up their friends and roommates. In Problem Set 7, last year, we asked students to implement their own E*Trade like website, whereby you had to create a website using that language call PHP a database called mysql that actually lets people log in and then manage a portfolio of stocks, buy stocks quote unquote, sell stocks quote unquote, get stock quotes. And from where? Well, we have the students; we had the student patch into Yahoo Finance where they grabbed merely the Real time stock quotes to actually simulate the process of buying and trading stocks. Unfortunately some students, perhaps some economics majors or just folks with too much free time figured out last year how to gain the system. So, even though we gave everyone 10,000 virtual dollars, I think our winner turned teaching fellow this year actually, right don't push them away make them one of your own, for a billion dollars something like that ridiculous. Turns out we weren't really checking whether or not people were trading after hours. And it turns out that also, Yahoo stalk quotes are not in fact perfectly Real time which means if you just have a real E*Trade account you can see the future and see where the stocks are going. 

[ Laughter ] So, apparently you can exploit that if you put your mind to it. and then, in Problem Set 8, the course's final problem set last year, we had students implement a mash up where they take Google News and Google Maps and mash them together using JavaScript, PHP and SQL in such a way that they had a website that they could find all of the local news from whether it's zip code 02138 or 90210. And, you would actually pull the local current even articles by this thing called XML and integrate it into the view of a Google Map. And then, at very course's end was the true climax of the course. This course culminates in final projects, which ultimately will be your opportunity to design your very own piece of software, designed by you or two or three friends. And the course then climaxes with the new and improved, with the second annual computer science 50 fair which last year was an exhibition across the street in one of the engineering buildings with 300 laptops atop tables, popcorn was popping, music was playing, stress balls with CS50's logo were being thrown around. And this was an opportunity for our 300 plus students to exhibit their final projects not only to each other but also to some 900 attendees in total. And we also invited some recruiters as well. So, that is what you have to look forward in this course. You also have cake to look forward to in the hallway. This is CS50 and we'll see you on Friday. 

[ Applause ] 

==== Transcribed by Automatic Sync Technologies ====