 I'm Lynne O'Brien. I'm Director for Academic Technology and Instructional Services in Perkins Library at Duke University. I'm Director of our Center for Instructional Technology, which is a unit within the library that serves all of the schools at Duke in helping them to incorporate technology into their teaching. This year, I'm also working directly with the Provost on a number of new online initiatives at Duke, one of which are the massive open online courses. So I have a number of different hats this year. Today I'm going to talk about the things that you see on this slide, some context for why we're doing things with MOOCs at Duke, not only the context at Duke, but the context in higher education. And although I'm going to be talking about Duke and using examples from Duke, my talk really is not a talk about Duke. It's more focused on what I see as the ways in which this new educational format is driving changes in our institutions, not only at Duke, but I think across higher education. So I'll use Duke more as a backdrop, but the focus is more on how MOOCs are a driver for change. I will talk a little bit about our current activities and how they relate to our goals for participating in some online education experiments. And then given the short time frame, I'm very cautious about describing results or making predictions, but I'm going to do a little bit of each anyway. So I'll tell you what we've learned so far in less than six months of participating in this space. And I'll also venture some predictions. I'm very nervous about doing that because things are changing so quickly. I'll probably look back in 12 months and chuckle at myself saying that I could have even imagined that I had any predictions. But I do have some ideas, and I'd like to get your feedback about all of this as well. All right, so as I said, I'm going to start with some context about why Duke is getting involved with MOOCs and higher ed contexts. So I'm going to start with last year, but the context really goes back to 2004, which was a year of either incredible innovation or infamy depending on how you think about it. My partner in crime is up here in the front row, Samantha Earp, who was working at Duke and is now leading the HarvardX MOOC efforts. And that year was the year that Duke distributed iPods to everybody in the entering freshman class. And you can remember probably that there was a big storm of attention around that, both pro and con. But I mention it because it reflects something that I think is true about Duke, which is that it is an entrepreneurial kind of school and one in which we are willing to experiment with technology and say, you know, we don't have a complete plan, we don't know exactly where this is headed. We don't even have really sharp, clear-cut goals necessarily, but we see something happening, and the way to understand that and the way to develop goals and processes is to immerse ourselves in it. And that's what we did with iPods, and that turned out to be, in retrospect, a great project because it created this atmosphere of open-ended experimentation. It got faculty thinking about the development of materials and the use of materials that included video and images and audio and sources from outside traditional things and PR clips and sports broadcasts and things that were not just textbooks and lectures. So I really see our work today going back to that. Last year we had these two projects, as well as many others, and the reason I'm putting up the sample of the chemistry project with one of our terrific faculty members, Steve Craig and Dave Johnson from the Marine Lab and an aphid he's developed is I think these set the stage for some of the things that are happening now with MOOCs. In the chemistry project, the faculty member wanted to see whether he could put together an introductory course using almost entirely already created open-access materials. So rather than creating every single thing from scratch himself, he went out and he curated examples from open courseware, from open materials on iTunes U, open access textbooks from Flat World Knowledge and other places, and then he did create some materials himself. He wanted to find out a couple of things. One, do you have to reinvent the wheel every single time you teach a course or might things be reusable? Second was, could the curation and assemblage of multimedia online materials help to replace the need for a proprietary textbook? The textbook in his course costs something like $230 and they only use half of it in the introductory 10 class. So price was partly a factor, but partly just wanted to think about whether other kinds of materials could substitute for that. And then the third thing he wanted to know is by creating and assembling these materials and putting them online, could he flip his classroom, stop lecturing in class, have students do things that were content driven outside of class and use this course in new ways. And the answer to all of those things was mostly yes with some cautions. Dave Johnson in our Marine Lab, which is a wonderful program, but it's not on the Duke campus. And one of the things they struggle with at the Marine Lab is both getting Duke students as well as prospective graduate students just to know about the program and the wonderful things that they do. So he created an app working with students that collected materials that he and his students had created through the years, images and video. And then because of his own professional network, he was able to connect with national geographic photographers and oceanographers at other schools and get them to agree to let him use all of those materials, share them out with a public audience through an app. And part of what he wanted to know was can you form these coalitions of content creators across the proprietary publishing world and the open access development world. By making things available on an open basis, can you engage a global audience in your course? And then the third thing was could you use these kinds of materials to attract attention to your program and to your research so that you could get a bigger share of the world's attention. And again the answer to all of those turned out to be mostly yes. That's the Duke context. I'll come back to that in a minute. Let's talk about the higher education context. This is a very crammed timeline of the last academic year. And it's not necessarily important that you read every single thing on here. But I think by just glancing at it, if you take a look, you realize that from the experimentation August of a year ago with the faculty out at Stanford saying, what would happen if we opened up our courses to the whole world to very quickly MIT doing a similar kind of course. And then the Sebastian Thrun spent off saying, some of you probably remember his famous thing about a red pill and a blue pill. And I've swallowed the red pill and I can never go back to the blue pill or something. But basically I've taught in this huge open environment. It's incredible. It's amazing. And guess what? I'm leaving Stanford and going out on my own. And that sent shock waves I think for a lot because it meant that a really talented professor could basically say, I don't need higher ed to do what I do. And in fact I've decided I can do it better as an independent operator and I think I can do it better as a corporation, as a startup company. So that was kind of scary. And then over the course of the year, Coursera, a different faculty member from the computer science department said, well guess what? I'm going to go off on my own and I'm going to create a different startup company. That was Coursera. And then not long after that MIT and Harvard said, we're going to get together and form an organization. We're not going to be a company. But now all of a sudden you had Coursera and Udacity and edX and any number of individual players as well saying, I think we can do things in a different and better way. And so on the far right in July, this past July, Duke announced that it would join the Coursera partnership. And in September we launched our first MOOC course. So if you think about the pace at which universities usually move, we went from, I was asked on Memorial Day weekend to please convene a group of faculty to learn about Coursera and some other online possibilities with the intention of reaching a decision by the end of the week as to whether we would join them. And I did it. I got 35 faculty members to show up and we got them to make a decision by the end of the week. So that is not typical of how higher ed usually works. So let's talk a little bit about what it is that we're doing. Coursera, I think most people in this room know what it is, but I decided to throw this in just because every once in a while I would be giving a talk and someone would say, I don't really know what that is. Well, it's a startup company formed by faculty out at the computer science department at Stanford. Is anybody here from Stanford? It's got to be awfully lively in that computer science department. They have 33 schools in their partnership so far. The American schools are all AAU schools, which is a relatively selective group of schools and they have a number of international partners as well. As of last week, they had 2 million unique students who had signed up for Coursera courses. There are about 200 of them that are either underway or in the works. And some things that are very significant, these courses are open access, not in the same way of, say, creative commons license open access, but there is no admissions process, there's no tuition, there's no fees charged for materials. You can mention books that are useful but you can't charge any money for any course materials. And there's no credit from the universities offering them. The courses in some cases have something called a statement of accomplishment that you can achieve by doing things of a level that the instructor sets, but not even all courses have the statement of accomplishment. So by the end of this year, Duke will have 11 courses that they will be offering through Coursera. And this is a graph of the registrations. Now I'm sure as all of you have heard, registration is very different from graduate or even student because we actually think of registration as an expression of interest in the topic, not even the equivalent of an enrollment. And you can imagine the far blue line, which is actually a philosophy course. Who knew a philosophy course would not only be the biggest enrolled course for Duke, but right now it's the biggest enrolled course for any course Coursera has offered. It's called Think Again, How to Reason and Argue, and it's essentially an introductory logic course. The reason things don't match up with the numbers is I made the graph before I went on a trip a week ago, and it was 164,000 enrolled then, and then when I made the slide it was 172, and then I looked yesterday and it was 184,000. So you have to take all these numbers with a grain of salt. Question? Oh, are you in the course? Okay. Correct. Correct. Which partly explains the popularity of this course. Although he did not make that claim until the course started, so that was not why people signed up. But his personality comes through in every aspect of the course. I sometimes wonder what would have happened if this course had been called Introduction to Logic, as opposed to Think Again, How to Reason and Argue. And then if you look at his little course promotional video, it's about two minutes long, you watch that video and you think, I want to take that course, and it's partly because of just who he is. Okay, so 11 courses this year, there's one that is not one here yet, an English composition one course that recently received funding from the Gates Foundation. Gates did a call for proposals and then gave awards for 10 courses to be developed in the MOOC format that would help address introductory level courses that could potentially reach a large audience. The Gates Foundation's focus is on increasing access and increasing completion of college, so they wanted to see if MOOCs could help with that. So that course will start sometime in the spring. Let me mention one last thing before moving on here. Duke really does think about their activities with MOOCs as experiments, much as in the way that we did with the iPod project. We have a set of goals, but really we just want to find out what they're all about. The fact that if we had asked in advance what course would get the most enrollment, it would not have been a philosophy course. If we had even had a really formal selection and course development process, saying, well, let's take our time, think about all the courses we could do, let's come up with a strategic decision to do this one instead of that one, let's look at data about course performance in the past. This is not the mix of courses we would have. And I'm guessing that by the time these courses are all over, we're going to learn some things that will help us in choosing the next round of courses that we would not have learned if we had done it in a really methodical way to start with. That's my way of saying all the chaos makes sense. So who is in these courses? Well, let's look at where people come from. We have four active courses so far out of those 11, and the only way we know where they come from is because of what they say on a pre-course survey. We did have 42,000 people respond, so I think it's a pretty good sample, but it's not the whole sample. In fact, I think it represents about 35% of all the people who were registered in the courses. Our pattern of geographic distribution is fairly typical across all the courses, although there are variations. And it's fairly typical for course error courses at large, but again, not exactly. About a third are from the US, another third from Europe, and then South America and Asia have pretty significant representations as well. I don't conclude that this shows where interests are. I think some of this is related to infrastructure. So for example, Africa, I'm sure is partly low because the infrastructure for taking these kinds of courses is not as strong. I'm guessing that part of why the representation from Brazil is so high is because they've always had quite an emphasis on higher education and have a more stable economy than some of the other countries right now, but we don't know. A couple of things about the student population. Only a third are what we would call the typical campus student age of the 18 to 24 year olds. About two thirds are over 25. 20% are over 45. About a third of them already have a bachelor's degree. 90% of them do not have a degree or significant work experience in the subject that they're taking. I want you to remember that because when we talk about completion rates later on, I think that's important. About 15% have some post-secondary education and somewhere between 10 and 15% have no prior college education. They're either high school students or people who didn't go to college. All right, our first course was a course in quantitative bioelectricity and 11,000 people registered. And so a question on my mind and no doubt many people's minds might be who takes a course in quantitative bioelectricity for no credit. Well, here's a few samples that are representative. I'm going to take a minute and let you just read a few of those. And then what I would like to ask you after you've read through some of these is what do you see as the advantages of having this kind of a group of students and what do you see as the challenges? So just take a minute to look at the slide. I should have asked if you can read from back there. Can most of you read it? The print, I guess, is not as big as it should be. So what do you see as some of the advantages of having this collection of students in a course? Or what do you notice about the differences in the students taking this course? Very diverse. What else? Yes, that's an interesting point. It would be hard for most of these people to come to class and take the course as it's offered at Duke on campus in a room. Good point, right? So what do you see as both some strengths and some difficulties in having the 20-year nurse, the cardiac researcher, the high school student, et cetera? Sure. And not just the high school student who might have trouble completing because it's over her head, but why might the working nurse who's been doing this for 20 years, why might she have trouble completing it? She might say, I'm interested in the topics three and six, but the rest of it, I don't need. Or how about this person? The electrophysiologist, full-time working mother with a young baby. Let me tell you when I had babies. I wasn't taking electrophysiology at night. And she says, I don't expect to be competitive, but just to enjoy whatever I can learn. All right, we'll come back to some of this. We think there are enormous assets to having this kind of variety in campus courses and bringing this back to our campus courses, whether it's somebody who's taking this course for the first time and wondering, what in the world do you do with bioelectricity? All kinds of stuff. Or someone who may make connections for research, somebody who might end up in an internship somewhere, someone who might just end up with a colleague to talk to you someday and would never have encountered if they took the course on campus at Duke. All right, so let's talk a little bit about what we've seen so far. Now, I'm going to keep doing my cautionary tales. We have completed one eight-week course. We are just about to complete a second 12-week course, and we have two other courses that are underway that started the week of Thanksgiving. So very, very limited information from which to draw these conclusions. But one of our goals in starting with the MOOCs was to drive innovation in our campus classes. These are the things that faculty have been talking about doing just in the last three or four months. Flip classrooms where you use the materials that are developed for the MOOC class so that you can do something different during class. Flexible class meetings and course length. When we started asking faculty if they wanted to teach these courses, they would say, well, when does the course start and how long is it? Well, you can start whenever you want. It can be however long you want it to be. Our classroom, our campus-based courses are 14 weeks long. Not a single faculty member chose to make his or her course 14 weeks long. They range from six weeks to 12 weeks. One person wanted to do a 20-week course. I talked him out of that and suggested breaking it into two 10-week courses. But one thing that's happened on campus that's been fascinating to me, Duke has a... We don't do credit hours. Every course counts as one. And we have a certain number of courses you need for graduation. So whether you're taking a lab course or a seminar course counts as one course. We are actually going to rethink the way we count courses and maybe move to a credit hour system so that we can have more variety in our courses because what faculty are telling us is, look, the first 10 weeks really hang together. And then the last four weeks I just put a bunch of stuff in there because it's got to be 14 weeks long. Or I always thought it would make more sense as two six-week courses because some of the advanced students, they don't need the first half and some of the people in the first half really don't get it. They're not ready to do the second half. But I just squish them together because we have to do a 14-week class. So the fact that we don't have to do a 14-week class means that we can start thinking about things like modular content. We can think about materials that are used to prepare someone to be ready to take a course. We can think about materials that you can refer a student to for more advanced work. So all of a sudden there's a great deal of flexibility and variety. The last one on the bottom right, adapting to change as students come to campus having taken MOOCs. We think this is going to be really interesting because we know from the pre-course survey that a number of people who are in these courses are either prospective Duke students or current Duke students. We also know that 10% of the people in these courses are either faculty or academics. So probably people just trying to learn what they're all about. So what's it going to mean when someone has already taken the bioelectricity course and then they're getting their engineering degree and that's a course they need in their curriculum. Those of you who work in engineering schools know that the engineering curriculum is not very flexible at all. Accreditation tribes it. There is a very rigid sequence of courses and you've got to take the courses on campus. You can't substitute courses from somewhere else. I'm pretty sure we're going to have students saying, you can't really tell me that I have to spend Duke tuition and take that course all over again when I've already done it. And if you don't believe I did it, tell me what I have to do to prove it to you. We don't do competency-based education at Duke. We give you your credit because you came to the course. I think that's going to be tough though. I think we're going to have to think about that. The other thing that may happen, we may have students coming into the class and gladly taking it for credit, thinking this could be my gut course. Who would think that bioelectricity would ever be a gut course? And I'm not proposing really that it would be. But what would a seminar, usually that faculty member teaches a seminar of 20 students, suppose four of them have already taken it online? Is that going to change the dynamic of that class? And if so, is he obligated to do anything about that or not? Don't know. All right. Impact so far on tech support and planning and faculty governance and policy setting and all sorts of things, we have chosen not to put a new support structure in place or any really very formal support structure because all of this is changing so fast that we feel like we don't know where it's going to be in six months, much less two or three years. So we want a very loose organizational structure for right now. And we have a long history of doing collaborative projects across different organizational units that do. So our Center for Instructional Technology, our Office of Information Technology, our Provost Office, the individual schools and departments, everybody's playing some kind of role in the development and support of these courses. I will mention that the little picture up there, this was taken from the landing page of the think again course and that is the professor who has promised to shave his head if at least 28,000 people finish. And the videos alone have been a disruptive force. They're about a minute and a half to three minutes long and many faculty said this was one of the hardest parts of putting one of these courses together because they have to be able to explain in a minute and a half what the course is about, what the requirements are and why you should be interested in taking it. That is really hard. But what students have told us is, that's great! You're going to do that for all the courses, right? And then when we say, well, not necessarily, I actually had a student who said, you mean a course that I have to pay dupe tuition for? I just have to guess what that course is going to be like. But if I take a free course, I can get detailed information and samples of what it's going to be like. I'm not so sure about that. So that's been very interesting. And then we've also had faculty coming to us and saying, well, I just realized I've always wanted to get the right kinds of students in my class or wanted to reach a broader audience. So maybe this is a good way for me to promote my campus course to campus students. So anyway, we're using a collaborative across campus structure. Let me just mention one last thing. One of the things that we're finding already is that faculty who are teaching in the MOOC courses would like to use the Coursera platform for some of their campus courses. And a number of them are beginning to combine their campus course with the Coursera course. So all of a sudden, the kinds of planning that we've done for our learning management system, we just converted from Blackboard to Sakai. We did it at what was record-breaking speed in the pre-MOOC world, which is we basically did that whole thing in about a year. And people were worried because we only moved over four years worth of materials. It was a very complicated process. And they've got the Coursera platform where people are saying, oh, I don't care if it doesn't have a grade book. That's fine. I'll use it anyway. Oh, I don't care if the feature set changes, you know, randomly in the middle of the week. The quiz tools, you know, don't really work well yet. That's okay. I don't mind. Where were you people a year ago? Who do you mean that's okay? But what it's telling me is that there is going to be a loss of control from the center. The idea that we will choose a course management system, we will make it rock solid, we will upgrade it at very careful times. We will move every piece of data for you. With our Coursera faculty, we've told them you can store copies of everything that you're producing locally and we'll help you do that. We're not getting anything out of Coursera for you if you decide to not continue using it or if the platform goes away and the company goes out of business, we're not going to retreat anything. So you've got to keep local copies of it. We would never say that about a campus stool. Well, maybe we will in the future. All right. The library and publishing, this has been a very interesting area of impact already, too. If any of you came to the HarvardX talk this morning, you heard a really great discussion of copyright issues and publishing issues. We have a scholarly communications office within the library. I work within the library. We hired an intern to help us with this, to help review materials and decide whether they needed any kind of copyright clearance or negotiation. We have been able to negotiate with a number of textbook publishers. The person with the wild hair, he's in the ninth edition of his textbook. And there's another one who's written one of the, who's going to be doing one of the neuroscience courses. And when we talked about, well, remember, you can't charge any money and you can't have any materials that are required that would charge money. And some of their responses have been, well, you know what? I've got the best-selling textbook in the field. I can mention to my publisher that he can either let me, he and she, can let me use those materials or significant chunks of them in my course at no cost, in which case I will mention to the 178,000 students that this is a really great book and put a link in there in case they want to buy it. Or if you won't do that, I can tell them you don't need this book at all. There's no need for you to buy this for any reason whatsoever. So I know every word of that book and I can teach it without the book if I need to. You'd be surprised at the reactions you get with those conversations. It's not that simple, though, because in the cases where people have given clearance, sometimes it's only for one semester, publishers really don't know how to think about this very well yet. So they say things like, well, what's the enrollment? And I say, well, it's 65,000 today, but by March, when the course is being taught, it could be 150,000. Well, what's the limit? I don't know if there's any limit. So that's been a difficult thing for them to think about. The other thing we've discovered is that publishers are willing to let us use some of the textbook materials at no cost when they went and got clearances for pictures and graphs and the kinds of things that are in there. If they thought that book was going to be for a U.S. audience, which in the case of the philosophy book, they assumed that there was no audience for that beyond the U.S. So now they're having difficulty because the eversion of that book, they can't make it available in other countries because they don't have clearance for it or worldwide use. By the way, one of the things we're hearing from students is they want any materials to be either available on the open web or easily downloadable. They do not want e-textbooks. They don't want proprietary readers. They don't want licenses that expire. I'm not saying their requests are reasonable, but their philosophy is I don't want to hear that I've got to open it up inside of the e-reader and do my annotations in there just give me the stop, especially if I pay for it. If I pay for it, I want it. I don't want any of that stuff to go with it. So that'll be interesting. We have been able to get some free or heavily discounted software. The license resources from the library, of course, are trickier. One of our faculty members found that he was not allowed to assign his own article for his class because when he signed his contract, he signed away his rights to be able to reuse it that way. And we are hoping that this will be a wake-up call for some of our faculty to look at those contracts more carefully and think about retaining access rights to them for the future. All right, early findings, meaning what do we know based on one class for eight weeks. Completion is always a big topic of conversation with these courses because the completion rates are always way lower than the registration rates. Now remember, I said that from our perspective registration is merely an expression of interest. And that's partly because across all of our classes and I think across most of the course era courses so far, about 50% of the people who register never click on the course one single time. So I would not consider the half that never click on the course once to actually be students. The sort of beating heart that we're going by is this one. The people who complete the quizzes or homework assignments for the first week, that's what we consider the starting point because if you at least took the quiz or did the homework for week one, you've shown some expression of interest in the course. Now think back to that slide about who takes the bioelectricity course. Even with that, it doesn't mean that everybody who clicks on the week one intends to complete the course in the sense of doing all the work or getting the certificate. The cardiac researcher doesn't need a stinking certificate. The grad student is already going to med school. That's the certificate that matters to that person, not the bioelectricity certificate. The high school student doesn't need a certificate. So there are many ways that people are active in the course without getting the certificate or taking all the tests and quizzes and one of the challenges for assessment is figuring out what is meaningful. What is a meaningful number? And the answer is we don't know yet. We just don't know. We do know that about a quarter of everyone who got at least one point on one quiz during the first week finished the course. And we actually think that's pretty good. This is a hard course. It's actually almost a graduate level course. Another thing we discovered from the student comments is that describing the courses accurately is really important. The faculty member for this course put down nothing more than average math is needed. If you're an engineering student at Duke, that's true. But not everybody in that course was an engineering student at Duke. His understanding of an average math background and the actual math background of many students in the course did not match. So I'm sure that if we had had a much clearer description up front, we have an interesting one underway right now. It's an astronomy course. And he put in his course description the course is going to do this. There's going to be a lot of math in it. If you can't do this problem and he put the actual problem and get this answer, you're not going to do well in this course. I thought that was great. However, over the last week there's been intense controversy in that course that it was not fairly described. And people are still saying, well, I knew how to do that problem. But you actually have to know all of calculus. And I haven't done that for 20 years. So you should go back and review it. Or I have the math background, but you didn't say it was going to take 25 hours a week. And I work. And so you should take that into account. And I talked with the professor a couple of days ago. And he said, well, am I supposed to change the course so that everybody do I need to go back and review calculus with these people? Do I have to give longer deadlines because somebody has a big project at work? And my answer so far has been, no, you don't. I think our challenge is going to be how do we let people know this is what the course takes? And if you're not really prepared to take the time to do it or you don't have the math background, then that's not something we can help you with. On the other hand, and well, I'll come back to that minute. All right, lots of controversy over cheating. Can't people cheat? Of course they can cheat. They could and some of them probably do. It has been very interesting, however, to see how powerful the norms against cheating are across the students in these courses. The one on the left is actually a Facebook course for the biology course that's underway, not bioelectricity. And somebody put up a thing for a paper mill there and people immediately, within 10 minutes people are writing in what's wrong with you? You get a free course from a university. You're not getting any credit. You didn't pay for it. Why would you cheat? And you hear that a lot with people saying, you know, sad the need for approval is bigger than the need of self-betterment. I mean, these people are really active in policing that. Does it mean that nobody cheats? No, of course it doesn't mean that. There are also some technology things that are being added to the platforms to make it more difficult to cheat. Typing recognition patterns proctoring through webcams and then a variety of things. And some of you may have seen the New York Times article about the Princeton sociology professor who actually had an army of TAs hand grade the same things that were turned in for the peer assessment. And he found a relatively low rate of plagiarism. Now what you might say is, well, how does that compare to campus courses? That's one of the things I think is very interesting about this whole movement because we have many faculty saying, well, how's the dropout rate compare? How's the cheating rate compare? How does this compare? And I say, well, what cheating data about our campus courses would you like me to use? Oh, we don't have that. I actually had a faculty member in a meeting a couple of weeks ago say, people were talking about the completion rates being low. He said, well, you know, if our students, our Duke students could drop a course at any time without any financial penalty and without any impact on their grade point average, how do we know 90% of them wouldn't drop our courses? I'm glad he said that and not me. But the truth is, we just don't know. We don't have a lot of data to compare some of this to. And, of course, the best solution for cheating is to have assignments where it's really not easy to copy or paste or to use a paper mill. All right, looking ahead, here are some of the things that I think are going to be features of learning at Duke and I think at other places, too, that are partly driven by the MOOC environment. Lifespan learning, you saw that with the people in there, a global audience, a diverse audience. Of course, it's being much more iterative and data-driven. For example, in the astronomy course, when students said this is too hard, that's too hard, he made some adjustments immediately in other cases that he would not. By embedded, I think that the materials and the teaching are going to become so intertwined that you really can't separate them. So the idea right now that you've got a teacher, you've got a class, you've got a textbook, you have library readings, these are all sort of in their own little silos, I think it's going to be a much more mixed kind of thing where the materials and the teacher are going to be much harder to separate. I've already mentioned a number of things that I think are drivers for openness, openness in publishing, just the fact that these courses are wide open for people to see and critique. Boy, I give a lot of credit to our faculty who are doing these right now to have a wide open 50,000 person discussion of whether you are or are not a fair grater, or whether your course is or is not any good, or whether that last question was or was not correct. I think it's a pretty thick skin, but there's going to be demand for more of that. We've already had some push it due from students saying we want all kinds of things about our new courses to be open and visible before we sign up for them. That's pretty scary for some people and not for others. And then experimental, I think I've tried to say throughout this, we don't know where all of this is headed, we don't know what's going to happen and we're comfortable with that, but it also means that some things are not going to go well. You can have an experiment where there's no chance of failure. And so we haven't had it yet and it won't be fun when it happens, but I would be surprised if we get through this whole year without some public difficult thing that we have to deal with and what that will be. I just don't know yet. So let me just wrap up. I had a wonderful opportunity last week. I got to go to Tokyo for a business meeting and while I was there, I put an announcement in the four courses that we have running and I said, I'm going to be in Tokyo if you're taking a new course and you want to come talk about it? Stop by this restaurant and I'd love to hear your ideas. So about a dozen people came and they were the perfect reflection of everything I've said. There were people who said, you know, I ran into, they were so earnest and sincere about the courses that really touched me. One person said, oh, I'm so sorry I didn't finish the bioelectricity course as if I would know that. I said, I was doing really well and I had a big project come up at work and then my baby was sick. But don't worry because I've downloaded all the materials and I'm going to finish. I am going to finish and several of them said, yeah, I wasn't able to finish this one or that one but don't worry. I will finish it and I think many of them will. So that raises some really interesting questions. We've been talking a lot in the whole MOOC world about big data and all this stuff that we're going to be able to do because thousands and thousands of data clicks out of all the stuff. That's true but the people who finish the course after it's over, we're not going to have any data about that. The people who did most of their studying over on Facebook, I'm not going to have any of that data. There are many things that we won't know upfront that we'll have to figure ways to collect later on. So I'm going to stop there and just again credit all the different offices that are working on this because we have a really fabulous team. But let me find out at this point what your questions and thoughts and feedback are. Yes. Thank you very much.