From Data to Diagnosis: Biostatistics and Bioinformatics with Dr. Juan Klopper

Hello, everyone. Thank you for joining us. Welcome to another Academic Innovation Support interview. As always, I'm Noah Chartoff, joined by my colleague John McNally. We have backgrounds in education, myself in mathematics, and John in physics. And today we have a guest I'm very excited about and I will let him introduce himself.

Thank you. Thank you very much, Noah. John, thanks for having me. I mean, I've been an avid watcher of everything that comes, at least on the YouTube channel.

Very happy to hear that.

Particularly also love how a CEO can live stream all those conversations. I think that's fantastic. Yes, my name is Jay. My full name is Jean, actually, but I would always joke that my parents couldn't spell the French pronunciation and they spelled it the Spanish way. I have no Spanish connections, but, you know, throughout the decades have had to suffer the consequences of their prowess when it comes to spelling. Anyway, most people call me J. That's fine. I respond to anything that starts vaguely with a J. And my background, once again, thanks for having me. It's fantastic. I think for me, it's always good for me to talk about these things, to talk about Wolfram language specifically. My background, though, is as. I'm a surgeon. I trained to be a surgeon. I went to medical school many, many years ago, went directly to do my residency in surgery. Maybe someone can hear from my accent where I'm really from at the moment. I've been. I'm way, way up there, right in that little corner there in Washington, D.C. but as I say, maybe someone in the audience can know from my accent where I'm from. Studied to become a surgeon at my residency and eventually ended up being being the head of the acute care surgery unit at, I'll now say where I'm from, that was at Grudeskir Hospital. There's always one little pride that we have always had at Grudeskir Hospital, of course, the home of the first human heart transplant with Christian Barnard. So a fantastic place that I ended up in and a fantastic place to be a surgeon. And I was a surge there for, for quite some time establishing that acute care surgery unit. And then the pandemic struck, of course, I think, though we're talking Wolfram language. I was also a keynote speaker at two years ago at JupiterCon and actually did that. You know, we couldn't get together. So, you know, there was a conference, everyone was at home. And I actually gave that talk from the operating room and had to unfortunately, you know, just share the horrible news of the loss of some colleagues, you know, the day before. And that Was truly a horrible time. As a, as a, you know, as a surgeon, what we, what had happened to us. We had closed most of our operating theaters and we would only operate the dire emergencies. And I remember very specifically, sorry to delve on a, you know, pandemic that we all trying to Please but remembering the 25th of December, Christmas Day, walking into the hospital and there was one single non Covid case in the whole hospital that I was covering that day. And that was a patient that I had done some emergency surgery on the day before. And the rest was this Covid. And we all downed our tools. I particularly downed my tools as far as my teaching, academia and research was concerned, you know, just to, to help out where we can. So that was a particularly difficult time. But before that I had slowly transitioned into doing much more research and the support, perhaps more so than anything else of the research efforts before the pandemic. Before the pandemic and as it slowed down, you know, got the call from the side of the, of the Atlantic and you know, was recruited by, by the George Washington University where I've been now for, for the last year in the school of public health, just trying to expand on what I had started prior to the pandemic. But it certainly has been a very interesting, very, very interesting time, a very sad time, very difficult time, of course, for service delivery as far as health is concerned. And I think from there also perhaps opening my eyes wider from just the world of providing care to one human being at a time, which is what you can do as a surgeon, has helped one human being at a time. You know, it's a glacial pace of trying to help society one person at a time. You know, and then the involvement with research just to move then full time into the public health space and the education of public health and research and public health together with that, obviously still having the connections back large research units that we are now building, which is, be honest about it, easier to build from here. There's many more resources here than in a resource constrained environment. And so establishing now, you know, some large research groups with, trying to solve some of the big issues in the. Under resourced environments, you know, that we do have around the world. So yes, that is totally different worlds. Yeah, I'm chattering away now, but perhaps the last thing I should say what got me involved in this is that I think medicine was always my parents. Back to them again, thank you. You know, their dream. I was fully sold on it, of course. I mean, I would not have done so many years of Study and work not to do it. My first love though was always some bend towards mathematics. And after becoming a surgeon I did the typical, what it seems like CEO of a big company, maybe Apple, etc. And just go take all the mathematics courses that you want because I can just walk across campus when I have some time. So I studied mathematics after that and, and the world of mathematics and of course surgery colliding would of course end up being biostatistics. You know, that's how the worlds came together. There's another interesting part about how the Medical Council in South Africa changed the rules as far as registering as a specialist concerned. And perhaps later we can get to that, which really set me on the path of trying to educate people involved in health care as far as using statistics and in the broader sense today data science is concerned.

Before we get into the more educational aspects, I was wondering if you could elaborate a little bit more on some of the ways that you use data in order to make discoveries about ways to help serve these under resourced areas that you were describing.

That's an unsolved problem, of course, and I think there's a lot more that has to go in this. A lot of the research money in the world would necessarily go towards diseases and disease complexes that the better resourced areas in the world. And I think coming from a resource constrained environment where we punch above our weight but still struggle immensely, I think it would be very rare to find someone who's in full time academic clinical practice doing all these other things as well. And as part and parcel of the fact that we do have those resource constraints. So what it is about for me is that there are so many unsolved healthcare problems and how do you empower people to solve their own problems to wait for someone else to give you money, to wait, you know, to have a more colonialist approach to that. You know, there's all sorts of problems with that. And when you encourage people, when you light that fire and show them you know how to, how to do their own research, because they are, they understand their problem way better than anyone else. And if you light that fire. And perhaps now would be a good opportunity to talk about the fact that a couple of years ago the health, the Medical Council in South Africa changed the rules as far as being able to register as a specialist specialist. No longer do you have to do only your residency and, and your, in your exams, but you also have to produce a research project that was enormously, you know, that was.

And this is to be a surgeon.

Yeah, to any kind, any specialist, any medical specialist. And that was a very good decision. The only problem with the decision, of course, is there was no resources available to help. Say, for instance, at the University of Cape Town, there's more than 500 people doing their residency in medicine, gynecology, surgery, orthopedics, you name it, there's more than five. And to put in the resources to help 500 people with their research projects. And these are clinicians that know nothing about statistics that most of them don't know very little about. It put those resources into place. It was an enormous undertaking without any money and support. And that was my, I think, the birth initially of me just starting to help people. People know now that I can help people start knocking on your door. And from there that's, you know, that you suddenly see how many people you can help and what they start producing after that. You know, if you look a year, two years down the line now they've got three, four papers out, they know what to do. You know, you've empowered them. And I think there's so much that can be done in that sense that you don't need, you don't need much to. If you have an enthusiastic person, you know, with intelligence, which, you know, all of us, all of our students have, it was such a pleasure to light all those fires. And what empowered me to light all those fires, of course, was the Wolfram language. You know, that was my main use case of the Wolfram language at the time, outside of, you know, just enjoying its use and just pursuing new knowledge in mathematics.

That of course makes us happy to hear. So when now, I mean, you've produced course materials and courses and many, many things that are educational resour. Can you talk to us a little bit about what were the early experiments that you had in this area after you saw this need for helping people to be able to solve their own problems using the tools of mathematics and statistics. So what were the early experiments and projects that you did in terms of helping people discover these superpowers?

Well, I mean, it was purely needs driven in the beginning. In as much as here's someone who has done their residency, they want to register as a specialist and the Medical Council says no, where is your research project? Where is your paper? Suddenly you have this massive amount of pressure from people who really need to get out and do their service delivery as a specialist to register with a council and they just, it's impossible for them to do so. Now I'm talking a specific case in a specific country. But remember by that time, through other means, what we had done, for instance, through the College of Surgeons, we reached out to many institutions throughout west and East Africa trying to establish, establish residencies in their own countries. So by that time I had done an enormous amount of work of working with surgeons there to establish how do you set up exams for people who want to do residency in your home country? How do you set that up? And we had done that help some using South Africa as a specific use case, but this extended well beyond South Africa by that time. So, yeah, solving a real world problem, you have someone who wants to, to register and they struggling, they don't understand, you know, they, they, they never took mathematics as a medical school, certainly was not interested in mathematics and had no time to study mathematics or biostatistics while being a resident. You know, the, the workload, you know, depending on where in the world you do your residency is enormous. And of course in South Africa, many people visit South Africa to do a, you know, to do a rotation there because the workload, as far as, as healthcare, there's concerns just enormous. So if you, if you want to learn how to operate on someone's AB from a gunshot wound, unfortunately, you know, you go to South Africa, that's where you'll, you know, you'll get your experience. So there's no time for these people. And so how do you onboard them in a short amount of time, you know, with enormous amount of other stress, stresses in their life, enormous workload in their life. How do you onboard them quickly to understand the data that they're dealing with, how to analyze it, how to view it. Luckily, you know, fortunately some of them you catch beforehand. So you can say, well, let's set up your research project before you collect any data. Don't come to me. And now your data is already collected. That's just a nightmare usually. And most people, you know, work with data, would know that that's where you spend your time, cleaning up data.

Yeah.

And so how do you onboard someone under those sort of circumstances? And how do you, how do you grow that? You know, you're one person trying to achieve this most of the time, you know, so how do you replicate all of this? And, and you know, there's languages that make this easy and I would count the Wolfram language amongst that. And I'll tell you straight off the bat, my favorite function of all time. I mean, language with over 6,000 functions or keywords is an enormous language, but there's one function that stands head and shoulders above the rest. And Whoever came up with that one, please send them my personal thank you. And that's the manipulate function.

Oh, yes, yeah, we're quite a feminine.

Language should have a manipulate function. How easy to set up and show someone, live how something changes depending on your input. The use cases of the manipulate function are enormous. Yes, just that ability. And maybe later I'll show in a book a little bit about that. But how easy it is just to show someone, this is what happens. This is what happens to your data if you change something, this is what happens. And of course, it extends beyond just biostatistics and research.

Well, sure, absolutely.

Say again?

No, we're just saying that I agree with you completely. In almost every sort of quantitative system, the most interesting question you could be asking is not what things are, but how they could change.

No, absolutely. So hats off to whoever invented the manipulator, added the manipulate function to the Wolfram language. You wrap anything inside that function and the educational opportunity that comes from that's just enormous. So, yes, for me, the Wolfram language and, and I will mention, I mean, I don't think one should only use one language. You know, if you can speak, if you're polyglot, the normal spoken language is such a. A good thing. And same with other languages. So, yes, you know, there was, there were use cases for Python, there's use cases for R specifically where I work now in public health, you know, we sit in dc, we're very closely connected to the federal government, you know, the fda, institutions like that. You know, there's certain expectations as far as language is concerned in these instances. So, yes, other languages, you know, are used, and I certainly use them. But as far as my experience, just that rapid onboarding was so much easier in the Wolfram, it came easier with the Wolfram language. And by that time I was well versed with the Wolfram language. Now, let me preface this by saying a small subset of the Wolfram language. Well, sure, which, which, by the way, if you think about it, you need to know so little about the Wolfram language to do so much. So forgive me for being a salesperson as fast, but let's be honest about this. You look at that documentation page and you see all the nice colors of all the different areas that are covered. You know, is there anything in, you know, that you can think about computationally that's not, you know, on that page?

So.

But you need very little, actually, to do a lot. And I think for me, that was. I could see that, you know, I felt that I lived that with so little use of that language, I could really help a lot of people. Very quickly we actually have a follow.

Up question from the audience. So you mentioned rapid onboarding of people. So the audience is curious if you could say a little bit more about your own onboarding. Like how did you initially come across the Wolfram language and start using it yourself?

That's I, for the life of me, I can't tell you when the first time I wrote the line of Wolfram code was concerned. What I do remember is walking into a bookshop and seeing just a book and it said Mathematica on the side. Now I think by that time I had heard of Mathematica. That was not the first time I'd ever heard of it. But when we moved across to the U.S. one of the things I had to leave behind temporarily was all my books. Because books are heavy. To bring them across is an enormous expense. It's much easier just to re buy them. Now. My wife will, you know, she says no more, no more books. But anyway, I do remember buying a book and starting to page through it and then immediately have to get the software. And I think that was my first real, you know, I'm getting my hands dirty now. That's what I seem to remember. Had I coded a little bit before, I think I might have used at the university, you know, some other time, had some done. But that's my vivid, you know, the memory of at least now doing it. Seriously. Okay, book that I randomly saw in a bookshop.

I, I had a similar question. You mentioned that mathematics was your first love. Do you remember what got you into mathematics? Because I've, I found a lot of people who describe math as an early love of theirs have very different answers to that question.

Yes, I, I think some of it, you know, do you like, you know, I can state this probably in a couple of ways. Do you like Fanta? Do you like Coca Cola? Or maybe I should say, do you like Merlot or Cabernet Sauvignon? You know what you know. So I think there's a natural. You discover all these things at school when you're at your most curious, you know, when you're at your most almost, let's say, vulnerable in a positive sense. Vulnerable to this new world that you see around.

You say most absorbent perhaps.

Yes, let's put it that way. I want to use all of the terms in the most positive sense here. And so you see things and they speak to you, don't they? So there's absolutely this natural it's going to speak to you. Let's leave it at that. And then, though I think the fortunate part is having a good teacher. I remember her as being the most enthusiastic. And by that I mean not, not, you know, grade one or grade two. I mean middle school, high school, you know, all the systems in the world are all different. So let's stick to high school. My high school teacher, mathematics, being the most interesting, phenomenal person as far as again, lighting that fire of the love of mathematics and making it so enjoyable. You know, looking forward to that double period. So, you know, two classes in a row, you know, I don't know where in the world that if things work like that, but those two periods in a row of mathematics is looking forward to that just so absolutely much and just absorbing everything. So, yes, I think that innate. Well, this is what I like and I don't like that. You know, there's lots of things in school I didn't like. I did not like history, I did not like many other things, but physics, mathematics, specifically mathematics, innate enjoyment of it, finding it interesting. And then just by pure luck, having a person who shows that to you right in the beginning are absorbent, let's put it that way, then being lucky enough, being fortunate enough to have a person like that and who really encouraged every single student in that class. I remember, you know, I remember the effort that she had gone through and if it wasn't for her, maybe my experience would have been different, but, you know, I would have purely had to rely on me just loving the subject. But yeah, I think everyone needs a teacher, you know, such as that.

Yeah. So speaking of this, you've mentioned that manipulate is one way for people to explore. So when you're taking students who are entering this world of biostatistics, how do you help them have this sense of exploration and enthusiasm when they're onboarding into learning really any language? But you mentioned one tool. So what are the sorts of techniques that you use as an educator so that students can have the same attitude and enjoy, you know, they're getting into this, this area that they really need in terms of being able to solve their research problems?

It really depends, you know, on your setting. If, if it was a resident, you know, who just has to get this project done with very finite amount of time. I'm not going to produce someone who understands biostatistics at any kind of depth. You know, I. What I wanted from them is the ability to read a paper critically section to understand, you know, when it says this is The P value, what that really means, what context you put in, where does it come from? You know, the first time that any of us saw P value in a journal article, and if you had by that time, no mathematics background, it's a very mysterious thing, you know, and you, you, you blindly believe because this value is very small, that this must make sense. And everyone tells you that this is important now. And you know, you just go with the flow until you understand what it really means. And, and so that's a bit more difficult if you have a constrained period of time. What you really want there is for someone to appreciate what it means, what the constraints are, you know, how, you know in an easy as possible way how to understand that. And what you want from them is just to enrich their experience of reading a paper and being critical about that paper. So that is something that you can pick up in medical school, of course, in residency, because we have journal clubs. You're going to discuss those, you naturally pick that up. But just with a little bit of background, your understanding of what those results mean is so much better. And for you to inform your practice, you know, to change your practice based on a paper, you really have to look at it critically. And you usually, you know, you cannot see that decision to some, to some other authority. You must make that decision for yourself. Of course, you go to a conference and everyone says, this is now the way we're going to do. There's some new guideline that comes out. You are going to follow that guideline. But in some sense, if you want to use the, the literature to inform your practice, you have to have some understanding of the, that now I get into the privilege, privileged position to have a whole semester to teach this, you know, a subject. Now, there's no ways that you can actually do this all in 15 weeks. You know, you need a lot longer. And the students also have many other courses. So, yeah, your constraints are still there, but now you have a different, you know, you have a different opportunity. You can really build something up from the ground. And to me, education in person, education is, and perhaps no one else will agree with me, but I will not be moved from this. Education is a performance art.

Oh, yeah, absolutely.

I completely agree with you in that as well. Yeah.

Okay.

So at least there are an awful lot of people who either don't seem to think so or if they do think so, put very little effort into that aspect of it, but it absolutely is one.

It is, it is a performance art. It is that, you know, I've done So much work and I've had so much success in the online education space and we can talk about that as well. But to me still, unfortunately, I'm going to lean towards the in person because there's something that I can bring to the table. There's some aspect of my personality, maybe it's because I was a surgeon. I don't know. There's something to me that can come out that I feel that connects. And of course it is that eye contact. Of course it's those subtle cues of body movement and from the student side that you can see the frown on the face, the lifting of the hybrid. You know, that's such a pleasurable experience going through that. And the eureka moments or the understand moments. I think you had mentioned that in one of your previous talks that, that I had watched here on the channel. You know, there's, there's something to be said about that and that performance art and also as I mentioned, my mathematics teacher, in bringing out that love and excitement and enjoyment to say that you hate, you might hate statistics, but let me show you the beauty of it. Let me show you the beauty of what a human mind came up. This, what is this construct called the scientific method? Let's explore this and see the beauty, the power that lies in that. You know, and if I think if you bring out that enthusiasm and you bring people in and now you have these wonderful tools available to you to bring it alive in front of someone as opposed to maybe a dry textbook, not. The textbooks are bad, right?

Sure.

Well, yeah.

I mean, there are bad textbooks, there are bad everything.

Yeah. I was hoping you could elaborate on this question of online education and textbooks.

Right.

Because you've been very, very successful in both the online education sphere and in writing textbooks. Nevertheless, you have an attitude. I agree with that. There's something about the in person interaction which adds an extra little something. So the question is, you know, what are the sorts of things that you find do translate well from the in person space to these other spaces, be they text or online? And then what are the things that you keep in mind when you're designing for this not, you know, synchronous in person space?

Okay. Yeah. I have to just once again say how fortunate I've been in life as far as that aspect is concerned. And if I do forget, please remind me and ask me about the wonderful work that is done by a division called silk, the center for Innovation, Learning and Technology at the University of Cape Town. You know, when MOOCs first came out, there's a Very nice backstory that I have there in the phenomenon. Phenomenal work that those people did and the support they gave me specifically and the help of educational specialists to onboard me on into this online world. The success is not mine, you know, even though the success has been, has been, certainly been in the rewards and recognition has been there. But if it wasn't for these people, you know, helping me, it certainly would never have been what it is. So my first take on this for.

Someone who's not familiar being massive online open courses.

Massive online open courses. Coursera, the old EDX now, you know, of course, bought by 2U, Future Learn in the UK, Etc. Yeah, and so my first, my, my first thought about this is not what translates best, what works and what doesn't work. My first thought goes towards, unfortunately not everyone can meet me at 9:35 in lecture room 400A and some people don't want to meet me in lecture room 40935 in the morning. Because being in a class environment is not the best way to teach. Every single human being might love that. I might enjoy that in that engagement and I'm hoping my students do too, let's say that they do. But there is this, that does not work for everyone and of course not everyone is able to do that. And for me, online education solves those two problems. Those that do not want to be there and those that cannot be there. I mean, just a few years ago, education for me was something that was locked up in a high castle, you know, surrounded by walls and if you could not get there or pay enough money, that was inaccessible to you. How fundamentally has the world changed? We can almost not remember how it was, you know, before knowledge was everywhere, the ability to something new was everywhere. We seem to not remember that this was not so when I was younger, I mean, what did I have? My parents bought some encyclopedias and before the Internet, that was my soul and only, you know, connection to knowledge. And I ate those things up. You know, I read those things from COVID to cover and you know, when they bought more, that's what I did. And how poor was the world then as far as, you know, making knowledge available. So for me on the online education space, I don't want to think about what can be delivered there better. I feel that that is a democratization of spreading knowledge to others. That is the, I want to think about along those lines before I think about and can really think about what works in that space and what doesn't work in that space. Because even things that doesn't work in that space, it's going to reach someone with a cell phone. And that's the only technology they have. Africa, for instance, you know, the cell phones are ubiquitous there. Laptop computers are not. Most people can't afford that. They have lap and there's some of these companies, the cellular telephone companies throughout Africa, for instance, forgive me again for those use cases but you know, that, that make data available at a lower rate for people who are using that data for educational purposes. So that happens in many spaces. When the pandemic came along, so many of the data, the communications companies had zero rated, you know, that kind of data access. So if we put out something from the university that would not count towards any kind of data usage, you know, on the cellular devices. And you know, so how do you the ability just to democratize or bring education to those that could otherwise not have it. That is one of the most phenomenal as far as I'm concerned, one of the most phenomenal advances in human society. And we underestimate totally, I think totally we see the, you know, we want to harp on the negatives of social media, we want to harp on the negatives of any kind of new, you know, new technology. And of course, you know, there's parts of society that drive fear as a form of control over people. And let's not go there, let's not fear this, let's see the positive, the phenomenal impact that this has had, this ability to teach more people. It's a no, I can't overstate it. For me, what a wonderful time to be alive and having, and have had the opportunity to experience it.

The democratization of education is certainly a very exciting, you know, reason to be into online education and platforms. For example, like Coursera. Are there any other aspects of this that really excite you aside from the democratization aspect?

Well, I was interested in, initially interested in how do you deal with assessment in this space. By that time, I think Coursera had, you know, the. So the behind the scenes story of that is when I saw massive open online courses for the first time, I was sold. I thought this was the greatest thing because I suddenly took a course and I took another one and another one, another and that's just me. I need to learn new things all the time. The reason perhaps where it was possible for me to leave surgery behind is, you know, you need some intellectual challenge all the time. So yes, I took many courses and I thought, well, we have to be part of this. Of course, if you look at the universities by that time who had signed up to edx, who had signed up to Coursera. You know, there's a bar to entry. For instance, the University of Cape Town is the highest ranked university in the African continent. I think times higher new rankings came out and still about 170 somewhere in the world. So by far the leading university of the African continent. But then you look at the maps of universities on EDX and Corcyra and Africa, for instance, there's nothing, nothing, nothing. A whole continent, you know, with absolutely no contribution to this exciting new field of massive open online education. Education, massive open online courses. That to me was wrong, so wrong and had knocked on the door of the Vice Chancellor, you know, which is like a president or provost kind of thing. The leader of the universities in, in South Africa would be the Vice Chancellor. Knocked on his door, banged on his door, had so many meetings started, you know, inviting more people to these meetings. I was really the fly in the ointment there. I wouldn't, you know, the kick it all. I really pushed and fortunately had people at that time that joined sort of that push. And he was very generous at that time to, you know, to take some of the Vice Chancellor's fund and to the tune of a couple of million rand, you know, with the rand dollar exchange rate not being that great, it's not as much as you would think it was, but make that money available for us to sign up as a university to Coursera. So we had meetings with Coursera, with edX, with future learn and you know, cost was really an issue for us and we ended up with, with Coursera. And then we had put out a call for people to develop courses of course for Coursera. And I put in my own personal proposal which was accepted. And then my course on biostatistics was then the first course from an African university on Coursera. So and from that led to many other things. By that time though, I'm at a tangent, as I usually do at that time was how do we solve this problem of assessment? And Coursera's solution at that time was peer based assessment. So you have to answer your question. Two other people have to mark that question. And I think the system would flag will let it through if those two marks are generally close to each other, but would flag if the marks are very different from each other, the grades are different from your peers marking this. And then a human would intervene. And that was certainly some sort of elegant solution. And of course then became came all the electronic grading systems in big grader, etc. You know, if you do have multiple choice type questions, these are easy for us for machine to answer and then also the ability then to give feedback to the student, to then provide extra material to the student should they get this problem wrong. And I couldn't do it all. So there's obviously people at Coursera who dealt with this on my behalf. I just generated the content. So once again being so fortunate to be surrounded by, you know, experts in their fields making a great success, success of this. So assessment, I think is one big thing in the online space which can get better, but is in certain fields at least a solvable and solved problem.

This is very interesting to me because it's sort of another form of democratization, isn't it? It's not only letting new and different students in who otherwise wouldn't have been able to get in, you're letting more and different perspectives of instructors into this field.

I think how important does that, you know, bad political history in a country like South Africa. Horrible, horrible history. And living myself through a change in the faces of the students sat in front of me, you know, through horrible history and you know, historical, you know, depriving people of education, what a horrible time. And I was also very fortunate, you know, to from a very early age being active in that space and being surrounded by people active in that space and university active in that space, trying to, to fight that old system. And fortunately that old system fell and then coming to the realization, and perhaps this is not the, you know, the best forum to discuss, you know, these issues because these are deep political human issues and important issues. And I'm not, certainly not an expert in this. I know how I feel about it and how I want to empower people. But as you see the faces of students change, I mean, that's an enormous change for a country to have gone through. Not many other countries, you know, have had that opportunity. And now to also to bring on board the faces, the education, the faces of the educators that speak more directly to some of those students. What another opportunity that online education unlocked. You know, there was a situation while you have to wait for someone to retire before you can get their post, if someone is an academic post for the next 40 years, you know, how do you onboard the next generation, what opportunity as a brilliant educator do you give to them? And once again, online education helped at least, you know, be part of that solution, bringing new faces of education that understand their students so much better than I could ever understand the students. You know, there's students who understand me. This I understand some Students. Someone else understands another student better than I can ever understand. And how better is it? You know, I've had these conversations. Well, that professor is a leader in that field. Of course, he usually a. He isn't it, you know, again, I have to shake my head, you know, at how bad we are and used to be and still so many problems we have, you know, we have to do better on. But, well, it's an expert in the field. They must be the only person who can teach that. Nonsense, absolute nonsense. You know, just to bring so many more people into the field of education just speaks to so many more students in such a richer way.

Absolutely.

Oh, no. But someone else is already teaching this. Why must we have a second version of this? Of course you must have a second version. You must have a tenth version of it. You must have 100th version of.

To have one version is to say that there's only one right way to think about this.

Oh, no. You know, there's again that. That online space. How, you know, how much of my life been enriched by seeing an educator on YouTube. This is what have I learned on YouTube from someone who's totally outside of my understanding of their cultural background, their country, and yet how much have I learned from them, perhaps coming in a different way that I would never have connected. I'm just talking about me as a person, you know, fields of study that I have learned something new on that I would not have understood in a certain way. Now, I think, John, I heard you in the last talk you had talk about the constructivist approach or a theory, at least.

Sure.

Which I'm absolutely sold on, of course. And I don't think there's a, you know, there's an argument there. But how did, how did that person allow me to construct my particular version of this knowledge? Because they said it in a way. And if I had not connected with that from a totally different country on another side of the world that I know nothing about, this person just said something in a way that I. That spoke to me. How many times has that happened to me? And here we are, you know, 20, 23. Look, what is possible.

Yes, that's a fantastic. That's a fantastic point that, you know, people's ability to relate to students really helps the students in terms of understanding how the subject, whatever it may be, is going to be useful to them, is going to be relevant to them in their community.

Absolutely. I've talked about before the. The paradox of education, as I refer to it, that sometimes the more of an expert, someone is on a topic, the less they can remember what it's like to be a student who doesn't understand aspect aspects of it, to remember what parts of this are hard to understand for a newcomer. Sorry to interrupt you there, John, but I just Very emphatically, that is absolutely.

I know, I say, where am I to. No, that's absolutely correct. You know, to put yourself back in those shoes and to remember how long it took you to get to where you are now. Suddenly you think, well, no, no, you should do that in a semester. Why don't you understand that that's a very difficult position to put yourself, you know, to put yourself back in those shoes. What. Be honest with yourself. You know, what was my knowledge at that time? How much did I understand at that time? Let me look at my grades again.

Yeah.

And even if they were good, you know, maybe that was just the assessment that was, you know, smiling on you that day. I don't know. But it's very difficult to put yourself back in the position. And, and no, I, I totally agree with. Perhaps the more, you know, the more, you know, the, the more the difficulty, more difficult that becomes. Not always true. Of course. You get the most brilliant people that are just the most brilliant educators as well.

Absolutely, absolutely.

But that's, that's just a lovely, you know, lovely when that happens. And of course, today we have a brand new teacher in the, in the classroom. And of course that's generative AI models. You know, I suppose we need to go there.

Very true. Real quickly before transferring to generative AI models. I'm curious, so since you've had these experiences in helping to create online course materials, what advice would you give for, let's say, a young professor or maybe aspiring educator? They don't have to be professors because as you're just mentioning, you know, we want to democratize this, the spread of knowledge. So what. What are the sort of useful things that you would tell someone who is, you know, really excited now about the sorts of things that we've just been discussing and the opportunities that this affords for democratizing knowledge and education? What. So how does one go about getting into these sorts of things? Or what advice would you give a person who's just starting to think about creating these opportunities?

John? I think I might say a funny thing. Become a YouTuber. Great.

Okay, well, what do you mean by that?

How many excellent educational YouTube channels are there? Many, many, many, many. How much have I learned from educators on YouTube?

YouTube?

To me, the University of YouTube is a totally underestimated YouTube should rank somewhere on, shouldn't it? There's not, you know, why, why are there no metrics for YouTube? So I'm just using YouTube as a, you know, the other platforms of course as well. If you want to look for innovation, look around what people are successful and how are they doing it. You know, the examples are everywhere today. You can go on LinkedIn learning, you can, you know, go take a free course on a MOOC and see which ones are the one successful. What are they doing? There's so many examples out there, there. So for me, that's number one. Look around you. You will find a great way of doing it and you will find a way that speaks to you because not everyone has the ability to teach in every kind of way. I sit in a classroom and no matter how good I am, I know there's a student who's not going to connect with me. That's just numbers. We're just playing numbers here. That's, that's being possible. Which is perhaps another subject which approaches the fact that. How much should you tell your students about other resources? You know, there are people know what I teach you and my notes. Nothing, nothing other than, than that. That's absolute nonsense. But, and again, I'm generalizing maybe perhaps, you know, there are the cases that it is not so. So forgive me for dramatizing in a.

Way, but in my personal opinion, I, as a generalization would agree. Right. That educator. I think all of our philosophies is to make people aware of as many resources as possible so that they can find the ones that, that really allow them to connect to the material.

But my favorite quote about this is saying the worst kind of education is the transferring of information from the notes of the instructor to the notes of the student without passing through the minds of either.

No, no, no, no, no, no. That, that creating a carbon copy of your knowledge in someone else's head, that's, that's a fairy tale. Right. As you say, that bypasses the head of that other person completely. So yeah, I'm glad that I, I, you know, I have learned people like yourself agreeing with me. Thank you. I suppose. I. We've talked, we've spoken about so many things now, I've forgotten where we are at the moment. We onboard me again.

We were talking about advice for people who are interested in starting out in online education and with an intention to talk about AI after that.

I'm naturally just going to, I'm naturally just going to go there. But yes. So there are examples out there and again, just as there is for the students, so there is for the teacher ample advice out there. And then the other beautiful thing, if you have that ability. And now I'm going to turn into a salesman again and I'm going to sing lyrical songs about the University of George Washington University and specifically the Milken Institute School of Public Health. I mean, the public health school here, one of the leading schools of public health in the world. I have landed in amongst the most phenomenal people that I can ever imagine. And I've worked with phenomenal people, with phenomenal surgeons. You know, in a place with a phenomenal history. I have for I. This is the greatest luck I've ever had in my life is learning amongst these people. These are just phenomenal people. The doors that you can knock on with the greatest enthusiasm, experience, knowledge, capabilities is just enormous. You can knock on so many doors and people will just help you. It's just the most incredible thing, you know, and that goes for a teacher, for a researcher and for a student. It's just, it's not discovered. So if anyone's ever interested, come to the. Come to the School of Public Health at George Washington University. So forgive my salesmanship on that one, but I speak from the heart and saying, what a, you know, what a phenomenal place. So if you're in that fortunate position and you know, your early career, wanting, other than research, want to do education as well, just to reach out. There is just so many people around you, if you're that fortunate, that will help you along the way, that will give you those opportunities, you know, that will really go out of their way, despite all the stresses in their own life, to help you succeed in that field. Because in a place like this, there's an enormous sense of let's make the students succeed. These are important students. They are the leaders. They are going to share healthcare policy in the future. They are going to contribute. Let's put them in the best possible position. It's so important what is happening here. We realize it's important. Let's do the best job that we possibly could. And perhaps, you know, at wherever you are, you are going to find those people. So it's all, it's all around you if you really want to look. And if you're in the fortunate position like here at gwu, there's just so many doors to knock on, you know, it's just phenomenal.

Great. So you brought up the topic of new instructors that are now part of the classroom, which are large language models. So I'm Curious. How has the advent of this new way of interacting with computers influenced and affected teaching where you are now?

Let's be honest with ourselves. Too early to tell. Too early to tell. Of course, there's the initial inertia, there's the initial fear. And I think you have already spoken about that in previous discussions about, you know, about the fear. Perhaps that's human nature. We were afraid of the first car, you know, and some of us thought the car was going to go away because this thing is uncomfortable and it bounces and it makes noise. So perhaps next year the car will go away and we'll be back in our elegant wagons drawn by a horse. You know, and there's so many examples of this. I. I am not a reader of the future. I don't have a crystal ball, but I can't imagine this genie ever being put back in its lamp. It is out this. The horses have bolted. AI, Generative AI. You know, let's say Star Trek and saying computer, what is this? We have. We have that, you know, in our culture as far as television entertainment is concerned. But I cannot see that generative AI is suddenly going to not be important. I think we have, we had a beautiful example in South Africa where the prime minister in 1975 said that TV is just a fad and hence television was introduced very late in South Africa. You know, what a way not to see the future. But it is there. We have to embrace it. But it's so young, it is so new. How can we ever, I think, be slightly arrogant to think where this is going, going? So for me, I want to see, I want to explore the ways that it can benefit the student and me, you know, as the teacher, how can we bring it together? So, yes, we have started some projects. I have an unofficial blog that I put most of my stuff on at the university. So I cannot speak on behalf of gw The. This is all my own personal, you know, everything that I say is my own personal point of view. But. So there are some experiments that we have started in our department that I'm working on, and the first one will be on linear algebra, of all things. Linear algebra, of course, if you're serious about data science, you know, linear algebra, my belief is that that should be your first bit of port of call. And so how can I introduce topics of linear algebra but making it dynamic through the use of a large language model? So the problem for me at the moment is. And. And it's being overstated. No one is expecting you just to learn from A large language model or ChatGPT.

Right.

By one of the, one of the products, the Coca Cola of, of. Of generative AI.

Yep.

Right.

In fact, they're usually discouraging that.

Yeah, well, you know, modesty. But you, you know, you, you cannot just rely on it if you, at the moment, if you don't have some background knowledge. And unfortunately for those of us with the background knowledge, it's very difficult for us to see the use of it in someone who's encountering a subject for the first time. How would they act with ChatGPT? So that. Or a large language model, generative model. You can't expect that thing just to teach you if you don't know. Now, you know, we talk about hallucinations, but we talk about the fact that it makes mistakes. I think we over emphasize it in as much as where's this belief that it has to be 100% correct?

Because people certainly aren't.

Because people aren't, you know, so why do we make the bar for AI so, so high? You know, why do we set the bar so high? So yes, it's going to make mistakes, going to hallucinate. You have to have at the moment some background knowledge just to, you know, to verify what it is saying. So there is a problem there that we have to deal with. So I cannot just relinquish my teaching to ChatGPT or large language model to do all of that. I certainly cannot do that. So I have to somehow guide the use. But it's by making use of it actively in the classroom that I feel the students now are learning two things. They're learning about the subject matter and they're learning the interaction. With this model as it stands in 2023, it might be very different in a couple of years time, you know, maybe one day we'll have the brain implants and have to, you know, have the ability to talk.

Who knows?

May it come before, before I leave this planet. But at the moment it's, it's clunky. You have to type. Of course, if you have it on your cell phone, your cell phone, you can speak to it. But yes, let's, let's use it. It is unfortunately an extra level of, you know, something else that you're adding to the burden of, of the topic that you're actually trying to teach. But yes, for me is the important thing at the moment. If, if you are a student and you're reading your textbook and start on page one, flip over to page two, you have a thought that textbook is so rigid it will only go to the next page. There's nothing else it can do.

It can't answer your question, can it?

Page three. Here's something that I'm curious now. I wonder, maybe I can ask something that the large language model is not rigid like a textbook. It is not rigid like a prerecorded lecture. And perhaps it's even less rigid than a teacher who now has one hour, ten minutes, or perhaps three hours. Our post grade lectures are three hours long. You know, how do you, perhaps you don't have enough time in those three hours to go where every student wants to go. So perhaps the easiest way to incorporate this is to say, let me show you how it works by legitimate legitimizing it in the classroom because your teacher is using it. It's legitimized in your head now as something official somehow. Let me show you actively. Instead of me following a lesson plan, we're just going to let go where our minds go. So if I start producing vectors, what, what, what? Of course you know the subject, so you know what has to come out in that lecture. But somehow can you let there be this freedom of thought? Get the answers from live from ChatGPT. Of course, as the teacher, you're guiding this whole process. But perhaps that is a good skill to teach a student that when they do then use this independently, you know, that they've seen how it works, that they are understand what the limitations are. And now it's not a substitute for you as the teacher. It is not this thing that you're inflicting on the student and hoping for closing your eyes, hoping for the best, that the answers they get is the right answer. So my first experimentation in this field is can I somehow surf this wave of curiosity that I'm trying to instill in the student? They ask a curious question about what we've just discovered. Can I use a large language model and then help them, you know, do proper prompt engineering, which is becoming a big thing. I've written about that as well. You know, just a sort of a guideline for our students how you know what prompt engineering is about. And hopefully, you know, that was put together with, you know, would help them a little bit. But can I demonstrate that in class so that I could know that I'm at least getting them to, or, you know, hopefully getting better answers for them on their behalf when they do experiment on their own. And I'm hoping that the less rigidity that's in that system versus a textbook or some other resources that they have, say, YouTube with a prerecorded, you know, lecture by someone that their natural curiosity.

Yes.

You know, can be. Can be fueled. I'm hoping that that should happen, but I don't want to be arrogant right now and saying, you know, this is the solution. This is where it's going to.

To go.

This is what's going to happen. But for us to embrace it, absolutely. We have to.

Wonderful.

We have to explore how this. We. How we can use this new thing called a motor car with an engine and four wheels. We have to explain, takes us and look where it's taken us, you know, and that's where we. For me, that's where we at.

Now. There's only four minutes left in our official time, but thank you so much for telling us before this started that you can go a little bit over. We look forward to taking this conversation a little further because this is all fascinating. I want to ask before we go off of AI, which I'm not sure, John, you might have more to ask about that, but before we have a chance to. I wanted to tie it back into something that you mentioned that I connected to very strongly, that teaching is so much about expressing and nurturing enthusiasm for the topic. And I definitely connect with that statement. I have long believed that's essential for teaching. I believe that students can smell fake enthusiasm, that they may not know what they're reacting to, but they react badly to it. And that if you don't think you can get enthusiastic about a topic that you're teaching, that you're not exploring it deeply enough, that there is always something really beautiful about anything. That's true. And so it's. It's absolutely something that teachers should be doing as much as possible. Well, I wanted to ask, with all of that information there, do you think AI can encourage enthusiasm? Do you think there is anything about its inherent artificiality that would prevent it from doing that, or do you think it's possible?

Well, if the first part of your. Well, not your question, but perhaps just your statement is, I think we're in the fortunate position, if you think about higher education, that you have someone who selected that subject, actually, there are some subjects that they absolutely have to take. You know, that's part of that. And perhaps the enthusiasm is not there. Our first. Our first priority is do not kill that enthusiasm, because most people walk into that classroom with enthusiasm. And the worst thing you can do is to stand in the way of that enthusiasm. And some of our education systems throughout the world are geared absolutely towards killing enthusiasm. I Want to be. You know, honestly, a lot has changed. You know, a lot has been improved. Let's, you know, let's. Let's tip our hat and say, you know, we. We are not in the 60s or 70s or earlier. You know, maybe it was great then, I wasn't around, but. But let's not kill that enthusiasm. Let's nurture that enthusiasm. Can an AI model do that? I don't know where this is going to go, Noah. I think. I think. Of course. Now, let's look at these. They trick you into thinking that, you know, there's some thought behind that. It's our human nature to be tricked by, you know, there's some kind of passing of a Turing Test here to some extent. And it tricks you into thinking that it is, understanding what you're saying. It's. And who knows what the next version is going to look like? Who knows what the version after that is going to look like? And perhaps, you know, we have the ability now. You do. Even if you do the simplest prompt engineering, it's going to answer you in a different way. And perhaps there are settings that will answer you in a way that particular to you because you've interacted with it a lot. Perhaps next time around it knows more about you. You know, there's so much of our data out there now, but maybe, perhaps in your interaction in the future, it will get better in understanding what you connect with better, because maybe, you know, there's some measurement or some metric about your update, you know, your response to its response. Maybe there's something in there to explore, and perhaps it can get to a stage, you know, where this becomes a bespoke interaction designed specifically for you in learning what benefits you as an individual. Now, we tried to do that, I think Sal Khan from Khan Academy, the wonderful work, you know, that they had done, that he had done initially, and the projects that they still run, whereby the responses are somehow tailored towards where are you struggling? You know, as far as I understand, their system is set up that way. It's going to give you extra work in the areas that you're struggling with. And perhaps the AI could just be that perhaps the AI, generative AI, you know, can be a much better teacher than any one of us can by that ability to give 100% individual attention to you instead of 30, 40 or more students, you know, set in front of you, which. Which one human being can definitely not do. I'm curious to see where this is going. I am. I am very enthusiastic to see.

So the answer is that whatever it can do, it will surely get better at it.

I cannot see that, you know, anything but that statement being true.

I'm likewise optimistic about the ability to use tools like this for helping students to ride that wave of curiosity as you described it. Right. Because that. A very good metaphor. Right. Where the students interest is sort of peaking as they're getting into a subject. And anything that lets them explore where their own questions are taking them is a potentially really great way of going about letting them find their way to a new subject. So you also mentioned one of the supports that you put in place in your classroom, which is this guide to how do I prompt this thing to get better sort of results. What are the other supports that either you already put into place or you, through your experiments are starting to suspect might be useful in terms of allowing students to sort of ride this wave? Because one of the other paradoxical things sometimes is that in order for the students to have a very free exploration about where their mind wants to take them, it actually does require a fair amount of support structures in place to allow that to be as successful as it can be. And so what sorts of supports you mentioned, the guide to prompting did you find is helpful in this area?

Yes, definitely. As I say, it's early days. It's very early days. I do not have five years experience and can tell you what works.

No one does.

And so again, what a wonderful time to be part of this. You know, we have to think carefully about this and perhaps the con, you know, people with conservative mindsets, you know, wanting to say just slow down, you know, we have to listen to them, just don't jump in. We have a saying in healthcare and medicine, first do no harm before you try and help another human, first do no harm. So let's try not to harm. Let's be very vigilant about what we're doing and what we're introducing. Very vigilant. Let's be critical, let's come together as academics and perhaps form interest groups, research this. Of course we've got to set this up as research, but, but perhaps put in place committees, working groups, etc, to, to monitor what is going on here, to actively engage with the students. You know, there must be some metric about where this is going, how well it's doing. Let's not just, you know, we can't, we can definitely not have free reign. No matter how enthusiastic some of us are. You know, those usually are, you know, we have, I think there's some constraint.

Is definitely, show me numbers.

Definitely called for the second.

No, I say show me numbers. Show me numbers. Yeah.

No, there's got to be measurement. That's. No, this is an academic pursuit, you know, that's my field in academia, you know, not in corporate. But, you know, the natural inclination is, you know, let's research this. Let's see, let's see what is happening now. It's very difficult. You really don't want to make your student the subject of your, you know, experimentation. That's just wrong.

Although of course, in reality they always are.

Yeah. And there's this other aspect which, you know, is sometimes called action research. Right. Where educators are encouraged to, you know, frame because, you know, there's always you, you think of five different ways of presenting the same thing with every new batch of students that you get. Right. And so, you know, figuring out for yourself which, you know, who is going to resonate do I think best with each of these different ways of presenting it is sort of a real research problem that educators are confronted with every single year.

But we do understand what you're saying, that you don't want to be throwing methods at a student that you think have a high chance of spectacularly failing.

Right.

And killing the enthusiasm most of all and adding to their burden. You know, students have an enormous burden. And I think it's much worse now than when I was a student. And I think there's not enough, perhaps always understanding of that. The stresses in life today. It's different from when I was a student. Absolutely. It is, absolutely. And we have to be cognizant of all of that, which I think.

And when you were a student, they were monumental to as well.

They were, they were looking back at it now. What a bunch of fun. What, you know, best time of your life.

Sure.

Connecting with this new knowledge. But the important thing that you, that you made me think of then, now is onboarding the students. They are equal partners in all of this, are they not? You know, we, we. We should not hold all the reins and make all the decisions here.

Right.

We learned that lesson in health care, I think, quite some, quite some time ago. I think it was a Swedish student, perhaps it was Sweden. I'll have to check, double check on that. Who gave a very interesting talk, you know, it went viral on YouTube of the system of the student being co designers of their curriculum, their healthcare, their medical curriculum, and many places in the world. And I had gone through that experience personally, where we were forced to redesign the medical school curriculum from the ground up, saying that this was so ancient, this was so Ill designed for an era that does not exist anymore. And going through that pain of redesigning a whole curriculum, forcing a whole faculty, a whole body of faculty, the whole school to introduce new material, teach it in a new way. It's a painful process, but one of the biggest things, and one of the former deans, may he rest in peace, unfortunately took his own life, you know, and I think what the system had done to him just to force him to that. The burden that was on that man, but what a fantastic human being that was. But his drive was so absolutely at that time to bring students into that decision making process. So they sit around that table with all the experts from education, the field experts, so the surgeons and the gynecologists, etc. In my example, and the students sat around that same table, equal partners in the design of the new curriculum. They understand what they need, they understand how they need it. They understand much better than you understand. That's my experience of that is 100%. I do not live in their world. How can I understand that? I understand my side of this equation, but they must be equal partners in this. And that's a very exciting other thing. I'm also again in the very fortunate position where students actually just reach out and they want to be part of some experimentation. While we, we hear you're going to do the summer thing, can we, you know, can I sign up? You know, and, and they shape, you know, to some extent what is happening there. So I, I think for me, the student is, is, is absolutely an equal partner in this. They are, they are the ones paying for this. You know, do you walk into a clothing store? You walk into a clothing store and say, you know, here's my $50. And the store manager says, no, that thing is, that's torn. And that's the one I'm going to give you. It's my money. I, I want this better one. No, no, you.

Yeah, I, I should have to figure out what fits me, shouldn't I? Usually I. There was a comment in from our audience right now that I wanted to make sure we got in to the interview somehow. And this seems like a very apt moment. Someone said one advantage of asking an AI versus a human back from when we were talking about you using AI as teaching tools and learning tools is that you don't have to worry about stupid questions that may also be a source of imaginative exploration. And I think that's a great point in and of itself. But I think the fact that that point is being made right now shows that there are students out there who have questions, who have things they don't understand, want to know about, and also think that that means something is wrong with them.

I think we've all, we've all been there at one point or another. Right? I mean, even though we intellectually know as educators that that's not the case. It's. It's a very difficult thing to break the habit of.

Sometimes it is. But. But this sort of idea of opening up the curriculum to design by students, I hope would help to dismantle that. To be able to say, at least at some level what you want to know about is what you should be taught, that it's, that it is not just about the questions that someone else wants you to ask, the information someone else wants you to have. That this sort of approach. I hope, you know, maybe it's pie in the sky, but I hope it could help to break that down. So thank you so much for making steps in that direction.

But anyway, whoever made that comment, thank you very much. I mean, that's an excellent, absolutely excellent comment.

Absolutely. The user's name was Antipas.

So excellent, thank you. And I 100% agree. You know, there's this big, such a big fear in me and perhaps that there's one thing we haven't mentioned before is who in that classroom is not getting what I'm saying now? How can I elicit that person? Where is that person? I look out for that person so much and I know of course by pure numbers I'm going to fail. Somewhere, somewhere along the line, there's someone who's not going to understand. And once again, what steps have we taken from previously, I think before you had spoken about other of these live casts was about formative and summative assessment.

Right.

Once again, I want to bring that back to my own little world of medical education with the residents. You know, we have these high stakes events called the final exam. And the amount of space that, that I come from, a world where.

One.

Part of your final exam is to actually operate on someone and being judged by people watching you operate. And thank goodness we got rid of that. How bad is that for the person being operated upon? Oh my goodness, craziness was that because.

No one's at their best, you are.

Not at your best when you are being judged while you're operating, you know, and we got rid of that, thank goodness. And what silliness was that? But anyway, high stakes event, totally artifact. If you think about the knowledge that you want to impart and the skill that you want to impart or put someone at least on that path. Because I remember when I first, you know, qualified either med school or being a surgeon, I couldn't do everything. I didn't know anything. It took years after that and for most of us, okay, now, now you put someone at that, in that very artificial final exam, that is not what their life is going to look like. That is not what real life is going to. That's not the, you know, the, the punches that are going to be thrown at, thrown at them or how, you know, how they're going to work in most cases. It's a very artificial thing that this final exam and such a high stakes event, if you don't pass it, the consequences are enormous. In the field of medicine, at least where I come from, we really tried to move away from that. As far as residency was concerned, is this continuous assessment of an individual. It's too late to find out at the end of the year or the end of the semester that someone didn't. That's too late. You've lost. You know, to me, always someone must take responsibility for that. And for me, always I feel very strongly that was my failure, my mistake. If a student, student couldn't do well or didn't pass, I failed. That's the first port of call, not the student. Where did you know? That's just a thing for me. And I think many people do the same. Where did I fail? Where did the system fail? And one of those things is just leaving it too late, too late to find out. So again, that thing is, I want to agree with this and maybe now you've lit another fire in me again from that comment is how can I incorporate finding that student, you know, who's now not going to put their hand up and ask a question. Especially, you know, the bigger the class goes, the that goes out the door, you know, everyone being given the opportunity or want to put their hands up, you can't ask anyone, close your eyes and someone can put their hand. What crazy way can I come up with? Just getting people to ask a question that they would otherwise not have. So yes, maybe let's put in more effort and thinking can I do let them explore with AI when I've empowered them to do so on their own. But once again, all boils down to how can I keep the spark alive? They came in there with a roaring fire and by the end it must not just be this little trail of smoke.

You had mentioned co design as a really exciting part of figuring out this new world of incorporating things like ChatGPT and large language models, all sorts of models into the classroom with the appropriate caveat that co design doesn't mean that one group of students co designs and then everybody does this. But with that caveat aside, what are some interesting suggestions that you've had from students that you've worked with in terms of this co design of what these sort of future directions might look like?

Again, not the forum for this. My personal experience had been going through a redesign of a medical curriculum enforced by government and government regulations, just recognizing the fact that was designed. And unfortunately this comes back to politics and the bad politics of apartheid. And so maybe I should go there because we can't hide, you know, you can't hide from the truth whereby, you know, medical students are forced to attend a clinic in, you know, far away from, say, the academic hospital there, all these community clinics that they have to go to, that's designed for a cohort of previously advantaged students who all got.

A car.

And, you know, had the ability to pop out to a restaurant and buy their food. And you've got a brand new code of students whose parents did not buy them a car, who can't pop out to a restaurant and just buy lunch for themselves. And now in, in the, in the early years of, of, of the new democracy, it was, well, you're going to be penalized because you didn't show up. You didn't show up. You know, you're a bad student, you are going to be reported. I'm going to, to, you know, it's going to be a negative afflict your grades. You are a lazy student. Look at you, you lazy. And that's absolutely the truth that happened in the early days. And so that's for me is a visceral example of absolutely enforcing a redesign and a rethink of, you know, that's a very tangible, I think anyone can understand. But it goes much more subtle than that. It goes, you know, there's so much more nuance to, you know, what you go into curriculum, what the background is of someone who comes in the struggles that they have, which you just cannot. Don't fool yourself and think you can understand that. That's arrogance in the extreme for me, the just, you know, having, having gone through that and listening to these voices, listening what they come up with, some things that I could never even fathom or think. And, and if I look at the curriculum now as being part of, you know, having worked towards that new curriculum, is it perfect? No, it's not. Can we ever, you know, should we always just stop because everything is now perfect? No, no, we always carry on. But the changes that were brought in and how it is now, you know, trying to empower a generation of new healthcare workers in that setting. Let's just talk about the successes of, you know, what happened in a country or in a region. Empowering people who had the power to start off with, with. Had the ability to start off with was probably better than the majority of the people who were empowered and were doing it before that, you know, just to unlock. How do you mean?

I'm not sure I understand what you, what you mean by that.

I'm saying apartheid was so bad. There were people with such great abilities and they never got the opportunity to that ability. And so let's use that as a horrible example which unfortunately I was born into. I had no choice. I, as the saying goes, I felt the need to be near my mother during my birth. But, you know, there you are. That's where you are born. Not agreeing it from a young age, having parents who didn't agree with the system. You're not agreeing with the system, you know, did what you could and feeling bad because you always feel you never did enough. But living through that experience of unlocking or not unlocking the potential is always there. That ability was always there. That mental power, mental capacity, skill, innate ability to do all of this artificially not being allowed, right. Because of, because of the laws of a country. And then to have lived through the unlocking, not unlocking of that, just to take those shackles off and it was not unlocked. It was always there. You know, just take away the artificial barrier to that and see all these brilliant people suddenly everywhere and again and again. Then that innate need to empower that generation to go out and solve health care issues by being able not only to serve the patient by standing at their bedside, but also answer some questions through research, you know, with diseases that they fully understand because they faced with that every day there's a different burden of disease in one country, from another country. I spoke about trauma. The interpersonal violence is, is on an unfathomable scale in certain parts of the world. And yeah, you know, in the research, the answers to those are not going to come from a country in Europe, you know, with all due respect, it's, it's going to come from someone who, who's in my instance, doing 4, 5, 6 laparotomies for gunshot wounds to the abdomen every night. That's how I spent my residency. I under, I felt Like, I understood that problem, you know, I can research that problem. And that's a, you know, tiny, visceral example. I'm talking tuberculosis, I'm talking about hiv, you know, people who understand not only the disease, but the patients with that disease, they are the ones who must come up with those solutions, you know, through research. And, you know, just having played a tiny little part in making that possible, you know, using the warframe language is what I did. And showing them how, you know, how to start their research career, even though they are clinician scientists. That's what I dealt with, was my cohort of residents. Yeah. So again, for me, very powerful, having gone through that experience, feeling enormously guilty from being born into that certain amount of that that you carry around with you and always will. But we all have our burdens all over the world. And of course today, the wars that we still see today, how one human being can do that, that to another, human beings, unfathomable to me. I have to stay quiet about it because if those doors open, whoever comes in, I have to operate. I can't make those judgments. And other people have the ability to do that much better than I can. But to see people's potential being given the opportunity to shine, I think there can't be anything more beautiful in the world.

Yes.

Something that, if I understand correctly, you care a lot about is open communities and being able to be able to have exchange of information and code, for example. Right. So I was wondering if, if you might be able to talk about both your involvement in terms of open source projects and also you, you've mentioned many times that you found Wolfram language particularly helpful in terms of, you know, starting research questions. What are some of the ways that you use Wolfram language along with some of these other platforms? Because, you know, I, I agree. You know, speaking many languages is good. So being able to write code in several languages is also good.

Yes, yes, I'm going to, just as I did before, equate that to the spoken language. You know, if you are polyglot, you know, empower. I can speak two languages very, very well and the third one a little bit. And I so wish I could speak more languages. That's such an empowering thing to do. And then the same goes for computer languages and, you know, some languages, I mean, there are these. I think. No, we spoke about it before in a previous meeting where we just spoke about the ratings that they are for languages. Yeah, the different ones, they are extremely artificial as far as I'm concerned. We spoke about for instance, the exceedingly good documentation that exists for the Wolfram language. I mean, they're no better than the Wolfram documentation. Let's be honest. I don't think that's, you know, that's, that's a point that I would argue with anyone.

And that documentation is either talking about, show me numbers. I believe it's the world's largest documentation of a software language.

There you go. And so, you know, do you need to reach out to YouTube video? To answer your question, no, you don't. Because the, it's right there, you know, it's right there at your fingertips. Should, should there be more YouTube videos out there? Again, forgive me for, for YouTube. I mean, for me it's the biggest search engines.

It's a shorthand for all these various resources.

Is there a new generation who, you know, I think that the, you know, I'm a visual learner. That's nonsense. We, you know, we've all seen the research on that work. But there's some things that you perhaps like more than something else. Are we seeing a new generation who just expect everything to be in a short video, you know, TikTok and all those? I'm not on TikTok. How would I know? But anyway, you know, are we, are we seeing a generation who perhaps don't want things in documentation wanted in a different way? Of course, now we have, I mean, once again, congratulations. How quickly did the chat driven notebooks come out? Like they weren't there and today they're there. That was fantastic. And a whole new field to explore. You know, do I even now need a YouTube video or the documentation? Well, I just use the chat enabled notebook. It's going to write the code for me, you know, so brilliant, brilliant new things, you know, coming down the pipeline. For me that was, I mean that was impressive. Implemented so well and so quickly and it's not even as I understand. I mean the next version of Mathematica, you know, will be even better. I'm quite sure it will be, and even better integrated. But well done for where it stands right now, just out of nowhere. And again, so to bring back to that as well, the wolfram plugin in ChatGPT, you know, what an initial enormous success and probably onboarded so many more people into the, you know, that otherwise wouldn't have bought that was a fantastic decision to make and to make that, that plugin, by the way. But anyway, I digress. On board, me again with where we were.

Yeah, so I was curious as we mentioned and you were sort of Stating very eloquently that there's value in using all sorts of different tools. So what are some of the ways that you use Wolfram language along with other tools available? You mentioned R or Python or something like this. So what are some tips or ways that you have found that you can use these various tools that you have available together to, you know, have a, a project which has gone from some research question to, you know, doing analysis?

Yeah, so I mean there are two parts, research and education. You've got to split the two of them up. So the, the, the research groups, a brand new one that we just started last week is a, is a doing, doing network analysis for instance, as far as trauma surgery is concerned. So and again driven from the authors of papers out of South Africa and doing network analysis and that what we're trying to build is just to see, you know, through pure graph theory, you know, where are those connections so that who can people reach out to if they want to further their research career and they aren't in Southern Africa, for instance, where are, you know, where are the vertices that make sense for them and do that so suddenly, you know, working with a group of clinicians who've never done graph theory in their whole lives. So just very simply putting together a couple of notebooks, onboarding people, you know, with the basics of, you know, what is graph theory? What is a vertex, what is an edge? You know, what is graph centrality? What you know, what's an incidence matrix? So easy in the Wolfram language? You know, there's, I mean there's network X I suppose in Python you can do that as well. But the implementation of so elegant and I think you can, you know, that's one use case. But there's so many implementations in the Wolfram languages are so elegant with a tiny little bit of knowledge. The beauty of course of just the notebook. I mean of course we have jupyter notebooks, but the long history that we do have with the computational essay with a notebook in the Wolfram language, I think, you know, it's such a nice environment to work, such a non threatening environment for someone who you're not trying to turn them into a coder, you're not trying to turn them into someone who does graph theory for a living science. They're a clinician. But I find it personally it is very easy to come up with a quick resource in a research group and not even mentioning the fact that my research documents when we have our meetings. As far as our research projects is concerned, I am the Meticulous one, putting everything in a notebook. It's such a nice environment, make it look nice. Such a engaging a format to present your work in and say, okay, this is where we stand with analysis and just showcase the work to your fellow, you know, people in your, in your research group with great, you know, some of them are pure clinicians, some of them are, you know, experts in some other field. And so if you're the one who's now also doing the biostats for that group, what an elegant way to, to have those meetings is around a notebook book. It is, it is phenomenal. So if you ask me, and that didn't come up, you know, what is the, what's the worst educational tool that ever existed? And that's PowerPoint.

Oh my goodness. Thank you. I look at, I look at PowerPoint similarly to the way I look at nuclear weapons that as long as everyone else has it, yes, we have to like have it available, but it would be great to invent it. It wouldn't it just completely gone for everybody. That, that would be so wonderful.

It is the, it is the overhead projector of the who. Who came up with it? The military. The US military in the 1930s, 40s came up with overhead projector.

I'm actually not familiar with that history personally.

Yeah, that is education from 1930 still exists today in the form of PowerPoint. And I, I have not used PowerPoint in. I cannot remember when I refused to. All my. Everything I teach is done as a notebook. I do not come to class with a PowerPoint slide. Never in your life. Now I like to write on the board. I would just start three hours later, I'm still writing on the board.

Something where you can change what you're showing and adapt to questions that are asked. Sure.

And again, the terseness of the language, the power of the design of the language. Once again, Stephen, well done. But you know, student asks something very quickly. Can you just come up with brand new code on the spot? That's not a problem. And lo and behold, the manipulate function.

Yes, that is a great one.

Quickly make that, you know what that students asking, you know where they're going, make it come alive on this big projector here above your head. You know, you make it come alive in front of them. And unfortunately, you know, Jupiter, the, the, the widgets, the notebook widgets, they are difficult to work with. They are difficult. The manipulate function, there's nothing close in any other language. The manipulate function. And for me, that is just how many times do I just rely on that. So I would come to class and I have the prepared notebook. Now, this is School of Public Health. So of course Python and R, you know, specifically R.

Of course. Yeah.

Realistic about it. They, I, I, you know, we have two postgraduate courses, Introduction to Python, Introduction to R. I teach both of those because they are important in upskilling or onboarding people for the work that they are going to do. These things are important. So I'm not, will never ever say anything negative about it.

Like you said, as much as possible, you want to make the classroom similar to what they're actually going to be experiencing.

So. Sure, absolutely. But there is the fundamental, you know, postgraduate courses that you teach and there's, for me personally, it's easier to make that come alive with the work.

Thank you. Nothing wrong with the tool, which is very powerful and very complicated to use and takes a long time for you to build what you want to build in order to get it to do what it's doing. Nothing wrong with that. Absolutely. But it really does limit your ability to be dynamic with it.

Right.

And that's such a thing in a class.

One of the other things about the language design that one of our good friends at another university pointed out to me was that if you look at what functions exist and you look at what options exist for those functions, this can actually be a good way. As a new learner, let's say that you are a clinician who's dabbling in graph theory, right. You know, you're not necessarily wanting to dive into, you know, introduction to Graph theory. You just need a working ability to do some things because it's relevant to your practice as a clinician. These research questions, but by seeing the things that are sort of a function and the options that the function has, can help guide your exploration in terms of figuring out what do I need to know. And so I thought that was an interesting observation. I was wondering if you had any other examples like that that had come up in your teaching experience.

So I'm in the fortunate position to. We have a master student at Kent, Master Teacher Academy. I'll get there eventually. I think I should stop talking now. But Master Teacher Academy, which is, you know, they take proposals to solve educational problems not through the, well, let's design a new course. And it takes two years to get the course, you know, through the system and eventually. So this just allows you to solve problems dynamically by, I suppose maybe it's akin to the MIT X kind of thing, which I'm not super familiar with. So it might be quite different. But so one of the problems that we see in the school of public health is mathematics. Of course, you know, many students come into public health and they did not, they were not math majors, you know, so. And suddenly they realize through the different courses that they take, maybe their first postgraduate here that, oh, this is interesting and I want to pursue this, this. But maybe I feel I'm not going to take quantitative methods because, oh, I'm worried about my mathematics, I'm worried about that and, or I have done it so long ago, I feel a little bit unsure. Can we create educational resources for that? So at the moment I'm working basically complete would be one on algebra. Because if we think about an introduction to bias statistics, you know, if you just look at how do you calculate the T statistic, you know, how do you calculate the x squared statistic that follows the chi square distributions? These are inherently, you can teach them as algebraic expressions. We don't have to go to the calculus, you know, for CDFs, etc.

Sure. And so the calculus of like the normal curve or whatever is incredibly unfriendly.

Yeah.

As an equation.

So how do you, you know, how do you upskill that algebra just, you know, to make everyone still happy with a basic algebraic manipulation. And then I spoke about the linear algebra, of course, and so, and then both multi variable, single variable and multi variable calculus. So designing those educational resources for our students to tap into and they, I've done them all in the, in the Wolfram language. And so I'm, I'm just abstracting away from the biostats a little bit. And so we're now talking the language of biostatistics, which is mathematics. And so can I empower and onboard someone very quickly as, as rapidly as I can, once again solving that, that, that old problem that I've spoken about before and, and making people feel comfortable again with the level of mathematics skills. And the hope in this is that people now feel, oh, it's going to be very easy for me now to take an advanced course in biostats. I'm, I remember integration, I remember, you know, my derivatives for a cost function and, and using gradient assumptions. All those things, of course, this is what they are. And so yes, let's put those resources out there for our students. I'm very fortunate to be able to work on that, you know, creating those resources. And they are done. And now I can divorce myself really from R, because now I'm not teaching categorical data analysis. Categorical data analysis I teach as well. And that is purely R based because there's so many of the packages and we're just preparing students, you know, there's work with the, with the different governmental agencies, etc. So we're doing this in Russia. That, that class, you know, before I came was done in R. I'm continuing doing this in R. Right.

Because even if you can do this in the Wolfram research, Wolfram language, having all of these lessons written up, I, I completely understand that it's a, it's an important resource to use.

Yeah. But then as I say, this, this new level of the support for our students that we are building out, just updating their mathematics skills again, again, I don't think you can use R for me will not solve that problem. Python with Sympy, with symbolic Python can to some extent get you there. I think it's quite an elegant implementation, but it's never going to reach the level of a computer algebra system the same way as would the Wolfram language be. I think that's a point. And no one can argue that, you know, there's a subsection of the Wolfram languages that's just geared towards, you know, that fundamental mathematics, linear algebra and calculus.

You were describing the various languages that you speak and some of them being computer languages. And I think I am going to be adding to your list when I suggest that mathematics is one of the languages that you speak. That it is its own language, that it has its own grammar, its own rules, its own poetry, its own prose. That it is a language, one that discusses a specific topics usually, but it's a language. And of the various computer languages, I've run into Mathematica's language, Wolfram languages, syntax, grammar, the way that it runs as a language is closest to maths language of anything that I've run into. It's, you know, how there are certain languages, if you learn, if you know English, that it's easier to learn, say French or Italian than it would be to learn Chinese. There are languages which are separate and ones which are closer to each other. And that, that's, that's just always been my perception.

Well, let's put mathematics in its proper sense. It is the language of the whole universe. Let's go, let's go. I, I know, you know, it's the fundamental construct, I think, of the universe. If I put my, forgive me, you know, for the experts in the, in the, in the chat that are physicists that have learned physics, Physics, There's something about mathematics that is just the, you know, the unreasonable effectiveness of mathematics. You know, there's all these nice sayings, you know, that that have been uttered. But so, yes, it is a language. It's also the language of data science. It is the language of statistics. You have to be able to speak that language. Now, I want to fully agree with you. There's nothing also saying the manipulate function is my favorite, but very close to that is the. Is the mathematics palette, you know, to make an integral encode look like an integral. Come on. You know, how beautiful is that? One of the things that's the thing of beauty.

Yeah, One of the things I'm curious whether or not you've had occasion to use. So I personally, in things that I have done, have not had occasion to do all that much R personally, but there is like an R link for being able to take working R code and run it in a notebook. So I'm curious if that's something that you have experimented with, with much in your practice or if that's something that you had anything to say about.

I'm going to say flat out, no.

Okay, haven't tried that one.

Keep these things very, very separate. And perhaps that's the wrong thing to do because, yes, you can run kernels of different languages in different ides, et cetera, and it's something that probably we should do more of. I am unfortunately, always been a purist. So if we're now talking R, we are talking R, if we're talking Python with Python, but maybe, maybe that's a shortcoming on my side.

One of the foundational ideas of Mathematica is that there's more than one right way to do anything. So we're not going to be prescribing that. If it works, then it's one of the right ways. So absolutely use them separately, use them together, you know, whatever it is that's going to work for you.

Look, yeah, they are important tools in certain fields. Sass, of course, we mentioned, we haven't mentioned here, of course, with the fda, if you want to, you know, if you want to present to the fda, you have, you know, it's SASS only.

I got some opinions on sas.

We won't go there. It's. Look, it is very, very important. You know, if that's what the FDA expects and, you know, that's the way that new medication, everything is introduced by the work done with sas, then so be it. It's worked out now we've got wonderful medications out there. The world of healthcare, you know, is part of the success story of as far as Humanity is concerned. And yes, so that language has played a role there because that's what the FDA has accepted now very recently, Novo Nordisk, they've put out a podcast on Posit website where they've made their first submission to the FDA successfully using R. And so I think we might see a crack in that war. And what at the end of that, at the end of that podcast, what the, the statisticians at Novo Nordisk was saying is what they were surprised about and of course it should not have been a surprise, but the curiosity of the people at the fda, you know, there's this perception perhaps, and it's my position, I might be completely wrong, that there's some level of rigidity that. But these are curious human beings and they were just curious to see how this would work in R. And they, you know, they met with people at Novo Nordisk and Novo Nordisk. I think from what I understand from that talk is they were a bit nervous about this. Oh, they want to come see us in some punitive manner. But it was not. So it turned out these were curious human beings and they, you know, reached out to Novo Nordisk, not in a punitive sense, but in. Let's explore this. Let's see, we want to learn as much as you want to learn and perhaps, you know, there's, there's that track in other languages will be allowed and of course, if we talk about other Food and Drug administration type institutions all over the world. Not everyone is connected to SAS in that way. You know, in many other parts of the world you can, you can use other languages. So let's see, let's see where this goes. But of particular interest to us with the fda. Let's see where that goes. But you know, R lends itself to that world. It is the world of, of public health. If you look just at the method sections of papers, you know, R is growing there rapidly. You know, as you know, the king of that world used to be spss etc, which is another little, you know, I'll just remain positive about it. But there again I, I much prefer writing code to do my analysis and not push a button and, and have no control over what's happening. But you know, you can see the open source language is starting to make a big impact when you just purely look at those metrics as far as healthcare and public health is concerned. But there's, you know, there's an equal place as far as I'm concerned for the Wolfram language, which perhaps has never been pushed in this field that Much, you know, it's not the language that you would perhaps in many places in the world when you have a school of public health, be mentioned that much. But I think we've spoken now about, you know, so many beautiful use cases in the instruction. And of course, my experience on the research side, again, of the Wolfram language is being able to do all of this.

So there's a. There is a specific question coming from the chat. I'd feel remiss if I didn't ask it. I definitely want to get back to talking about the more general things as soon as possible. But there was a question. Does Dr. Klopper have a recommendation for mathematic packlets specialized for circular statistics? Their example is the assessment of biorhythms like circadian rhythms.

Okay, let me throw that right back at the company and the gentleman representing.

The company, and let me say that I am going to look into this and post the answer, both from Dr. Klopper and ourselves in the comments under this video when, if and when I have something specific to recommend.

So, yes, so, yeah, please watch the space. I'm pretty sure that exists, but unfortunately I have not used it and I'm not aware of the existence. But if that's a question, I'm sure someone has solved it. We'll look for it.

It's interesting that you brought up this use of integrating large language models into the notebook environment, because actually, we've spoken at least one school of public health recently about the new functionality. To be able to have something that is, you know, automated come up with code. And maybe, you know, the code might not be exactly right the first time every time it does it, but, you know, neither will it be if a human did it. And so this is a great way to take people who are not aspiring to be, you know, experts in software engineering. Right. They're clinicians, they're public health researchers. They're people who, you know, have research questions that drive them. But the actual coming up with the code is probably not the most interesting part of their job as far as they're concerned. Right. And so, so having this ability to have not only as a teaching tool, but even to suggest new ways of doing something with a language that you may or may not have great facility with at your current level of learning is a great new opportunity that's available to everyone.

Let's watch the space once again. What is version 14? What's version 15 going to look like as far as how are we going to interact with coding or computational thinking or data Analysis in the future. We don't know where this is going. If I, I, if I can venture a guess, it is again going to democratize people's ability to do this work because they are now not stuck in the, you know, sometimes esoteric construct of a language. You know, R is specifically not the easiest thing to learn when you see it the first time, you know, or with a tidy verse dialect, you know, which you know, some people love and some people don't love, you know, complicates, you know, when you don't have a single design philosophy as far as your language is concerned, you know, you have base R and you have the tidy verse and these things are different. And so the onboarding there, you know, you almost have to double your efforts now because, you know, there's, you know, do you do this in basa? Do you do this with tidyverse principles? And so there's something to be said about a coherent language design, I think, which again is one of the success stories of the Wolfram language being built from the ground up, an extremely elegant and proficient way. But who knows where this is going to lead and how is that going to broaden the ability for a domain expert in their field to solve their problem easier in the future and not have to take up their precious and valuable time to learn the minutiae of a new language and something that they don't want to do and they don't like something. So let's see where this goes. As far as I'm concerned at the moment, it's already brilliant. I have now personally, as you say, we all have bugs in our code. I didn't just sit down and write beautiful code and all the analysis just works. I have used now the chat driven notebooks and I have now asked the large language model to write my Wolfram language code quite a few times. And you know what? It works.

Yeah.

Made fewer mistakes than I have and I did not have to go to the documentation.

Yes, yes. I mean, this is, this is the next step of the philosophy that Stephen Wolfram was espousing when he first designed this language, these programs that he designed, that he was looking at a computing environment where you would have to define addition in order to do it or download a packet of someone else's definition of that in order to do it in a computer. And he just thought that was ridiculous that you should be able to have a program that does all these things you already understand in a computational way and that you, you shouldn't have to be going outside of your Understanding of the math that you're doing, of the science that you're doing in order to use it with a computer. And now it's going even further in exactly that direction, saying, yes, we have a great language, we have a great coding language that people can use. It's much more intuitive, it's much more complete. But also, if you have no experience with coding and don't and find the idea of starting that experience daunting, you can also jump in with this tool that is going to help you with that stage of it. I love the idea of being able to say, you are intelligent, you know what you know, you know what you want to know. That should be enough for you to use a computer to go from there. And I think that's just a wonderful place to be.

Absolutely. And let's not forget Wolfram Alpha. We've existed for quite some time. And how many times in my education material that I prepare for class, I would just have Wolfram Alpha, you know, just bring up, you know, I want to talk about, you know, now suddenly I can't think of an example. But instead of me, oh, let's say we, we talking about the development of the understanding of something very simple, the dot product, as far as the angle is constant. So I want to bring that in. How do I derive that? Oh, I just want to remember the cosine rule from trigonometry. Quickly, just put that in your notebook. You know, Wolfram Alpha, there's a beautiful piece written for you. Remind everyone quickly of the cosine, you know, rule. And okay, from that, very easy. Now we derive an equation for the dot product as far as the angle is concerned. So, yeah, many times I just slip in Wolfram Alpha just in my educational material, you know. And again, Wolfram Alpha has been with us for quite some time before the large language models of gender AI we have. And how elegant a solution, you know, has that always been, that making knowledge computable in some sense and understanding what you're trying to type into this text box, that's been a brilliant success, hasn't it?

So, yeah, it has. And so we've been talking to you about so many different things that, you know, I want to make sure that we have a chance to circle back to anything that might have come up earlier in this conversation that you're excited to talk about. So, you know, in terms of what else do you really want our viewers that are with us to, you know, come taking, taking away from this conversation, what are the exciting new resources and things that you would want them check out. What are those sorts of things that we could share?

Well, absolutely. As far as the warframe language concerned, are the chat driven notebooks. I think that should be your first new chat driven or chat enabled, whichever one's going to work. For your specific use case, enable those and explore those. Yes, you need your API key and you're going to pay for those tokens. And I think as far as those things are concerned, sometimes we perhaps oversensitive with the fact that you actually have to pay for something. No one, you know, the good things in life unfortunately cost money because they effort, they take people's time. You know, someone worked hard on this and, and that's just the way, you know, works. So yes, you know, you're going to pay for those tokens and so be it. But you know, you, for me, that's, that's. Let's just watch this and, and we'll play with it number one and see, let's see where this is going.

And.

It'S so difficult for me, I want to say this for me to answer the question because once again, look at that documentation, the front page of the documentation and look at the breadth of human scientific endeavor that is covered by this language. My use case on that page is this small and biology, finance, physics, engineering, you know, you name it, it's there. If you haven't, you know, if you're watching this and from my point of view, if you haven't explored these things, go and go and do it.

Yeah. Another, you know, quick tip that I'll give for people, if you are new to in particular Wolfram language, is that there's a couple of things that are good to understand about the way that it works in general and then you don't have to know all 6,000, you know, functions to be able to do what you want to do. Right. Oh, of course, it's very much not designed that way. It's designed in such a way that as you're just describing, you know, there's going to be a subset that's relevant for the domain that you care about. There's some things that are sort of very general that are nice to know and, you know, that's enough to get started.

Absolutely. And once you know, it's one of the successes of the design of this language and also once you understand a little bit, you can probably guess at what, what a certain new function is going to do and how it works. It is not that there's this divergent approach to how these things are Put together, I think it's a very easy language to extend your use of it without having to put in an enormous amount because there's this hodgepodge approach to designing it and the different dialects and the design of it. You really have to put an effort to learn the language. I think there's some elegance to this language in as much it's easy, in my experience, to extend your knowledge of the language. But that being said, I'm going to fall back on saying that you need so little to do so much because it is so powerful. I can't oversell that in my personal experience.

And isn't that the goal of the tool, to let you do more with less?

Yeah. It must not stand in your way because, look, I mean, they are completely computer scientists and, you know, now this particularly is their field. And of course, you know, I'm sure that that knowledge exists in the Wolfram, in the company itself, you know, people whose interest is the design of the language. But for people who are domain experts outside of computer science, and you need a tool to solve your problem, you know, you can't, you can't, you can't do much better than the Wolfram language. And that's, as I say, I use Python a lot. I use R a lot, lot. And so it's not degrading any of that. Oh, no, but. And do all of them, you know, please, please go ahead and do all of them. You'll be, you know, the rewards are there. You'll be richer for it, for doing it. But should the Warframe language be there? I'm hoping just through my, you know, one person's experience, you know, that does not data analysis make sure. But if I just look back and what the language has meant for me and, and from what I can see, what it has meant to others whose paths I've trusted and who I've started on some journey. It's worked beautifully. It has worked beautifully and I'll continue to use it.

Speaking of others that you've sort of helped along this path of learning to use in particular, Wolfram language to solve problems that they're having. So, as I understand, you have new versions of some of the educational materials that you've created. Am I correct in understanding that?

That.

Yes. So, for instance, all the, all the new material for this Master Teacher Academy project, the, you know, onboarding again, or just sharpening those mathematical skills again for our students, those who feel so, you know, this is, this is going to be a resource that exists for our students and that they can tap into that is all done. You know, that's all done in the Wolfram language. There is a version that I am putting out as far as the algebra is concerned using Python, I'm putting that one out as well. So definitely that is. So now the formal lecture notes for some of the courses. You know, I joined a department. It's a. It's a newish department. Biostats and Bioinformatics as a. One of the newest departments in the school of Public Health. You know, having evolved out of a previous setup up. So there are these courses that I am now just teaching that previously were taught by someone else and, and you know there's a large infrastructure of. Are already existing as far as the education resources for those are concerned. But when it comes to me interacting in person with someone and making it come alive for them, even though I will put my hand up and saying I rewrote all of the education material that that existed for those courses rewrote all my own, I cannot do some. I unfortunately cannot use someone else's notes. I. I have to remake my own.

I relate to that.

I. I just cannot. And so yeah, they all redone and I did them in R because they existed previously in R. So they're all rewritten and they all in R. But when I pitch up to in person and I have to make it come alive, those notes in class are all the Wolfram language. Because I can really. I can. I can light that fire. Once again it's back to being a performance art. And what a beautiful friend or tool or whatever it is. Co educator in my class. The Wolfram language is my partner in that making it come alive. I can't make it come alive in another language. Someone else might. My specific use case is I. I can use. I can make it come alive. Alive with. Yeah.

To make the metaphor very clumsy. Perhaps you. You use manipulate to make the figures dance as you're doing the performance of educating.

Absolutely. There's nothing. There's just absolutely nothing like it.

Jay, speaking of making things alive, thank you for every single word that you said in this interview today. This is, this has all just been enlightening in. In every sense of the word. And I'm. I'm so glad that everyone got to hear us.

Oh yeah. Sorry for everything in my little off a tangent so I can speak about this. I really appreciate you listening.

No, I mean I think the audience definitely appreciates not just hearing about any particular topic. Right. I mean of course we have sort of a primary focus. But as we were just discussing all of the various things that students are bringing into their context. As educators, we also bring our own personal experiences and our sense of community into the context. And so this is, of course, relevant to everything that you're saying.

Absolutely. I love a quote from a comic book of all things. That said, anyone who thinks one book has all the answers hasn't read enough books. And I'm a big fan of that sort of perspective of saying that even if one tool is capable of doing everything, it doesn't mean that it's always the tool that is the right for everyone to be used. There could be someone who connects with something else. I mean, so much of what you've been talking about today has been about removing artificial barriers and allowing people with different perspective, different experiences to gain the sort of benefits that should be universal to everyone. And I, I love everything about every perspective that you were sharing with us today.

Now, we live in a wonderful time. And as I say, for me personally, I'm in a wonderful place as far as George Washington University is concerned and being part of the educational system, being part of a research system, you know, what, what a wonderful time to be, to be alive and being, you know, have the abilities and the tools that we do have today. So my suggestions is to everyone else, it's all around you. Just, Just reach out and use it. It's. It's never been as good as it is right now. And from everything I think we can see at the moment, we're just at the precipice of making it even. Even better. So, yeah, I can't but be enthusiastic about, you know, these tools in research and, and, and in education.

Have you ever heard of ameliorism?

Tell me about it.

Ameliorism is a perspective which is offered as distinct from optimism or pessimism, which is the belief that the world can be improved and that perhaps on some level it trends in that direction overall. And I don't always have that perspective. When I'm looking at a lot of things, it's hard to maintain that perspective. But I hear it in what you're saying to us, and I think as much of that as possible is good in the world.

Well, thank you for this opportunity. I mean, people must speak out about the positive in the world. It's so easy to talk about the negative and see the negative. And unfortunately, the negative is a reality, and you have to feel bad for what is going on in so many places.

And you should not ignore it.

You cannot ignore it. You cannot do everything Yourself, you know, you have to seed the work and the responsibility to others. Our modern society is divided. In as much as you are a specialist in one field, someone else is a specialist in another field. You cannot solve all the problems, even though it makes you feel extremely incompetent and, you know, you feel at a loss. You want to contribute, but you just cannot. And perhaps the only way, you know, the only way is to speak out. We each have a voice and concentrate on the positives. I've certainly, you know, being at the end of life for many people, being, you know, faced with, you know, the worst of what can happen and eventually will happen to each of us, some in a better way than others, if you do not concentrate on the positive, if we will never be driven forward. And maybe perhaps what you said now is so absolutely true because we have moved forward and perhaps if we look at it critically, yes, we have tended towards the positive side, otherwise we would not have been here, you know, where we are today. And there are checks and balances and they are important as well. Well, being conservative in certain circumstances is obviously, you know, is, is, is good. We had, yesterday, I had an expert come to my class as far as the, the review boards are concerned, the ethics board is concerned, to come and speak to my students, you know, and how important that is. And that was a wonderful talk given by Paul Hindabele at the university as far as the IRB is concerned. And, you know, so many, so many experts, so many wonderful things, but yeah, so many things to be positive about. Absolutely. And this has been a wonderful experience for me to be, be able to talk about these positive things and hopefully it helps one single person.

I could say it's helped me. So there's that.

Thank you.

And I believe you've helped quite a few and then we'll continue to. So thank you very much for being with us today, Jay. Thank you everyone who joined us online. Thank you everyone who's watching this later as a recorded interview. And as always, it's been a pleasure to share the time with you.

Thank you very much. Thank you. I really thank you for the opportunity. And I'm obviously subscribed to the channel, have been watching for so long and looking forward to your next one.

Our listeners should also subscribe to you as well because you've got some great content out there.

There.

I've got a tiny little YouTube channel. Yes, I, yeah, I, I, I need to do some more work there. It's been, it's, it's been a while. I don't know where, how many subscribers. I have around about 12,000, thereabouts. Tiny little.

I have a feeling you're about to get a few more. So.

Thank you very much. Thank you very much. No, it's, it's wonderful. Wonderful what, what you're doing, wonderful what the company is doing. And, and.

But.

When it comes to research and teaching, I think there's so many people doing wonderful things. And thank you for highlighting some of them and even my tiniest little contribution. Thank you for that.

Thank you so much.