Sal Khan’s Keynote

♪ ♪ Hello, I'm Phil Hill. A consultant in market analysis at MindWires and also a blogger at the PhilOnEdTech blog. Today I have the opportunity to interview Sal Khan, founder of Khan Academy, and a well-known author and speaker. It's been ten years to the month since Khan Academy, as a nonprofit, obtained its first major funding from Google and the Gates Foundation. So I'm looking forward to discussing with Sal perspectives on the past and future decades in EdTech. So, welcome, Sal.

Thanks for having me, Phil. Yes, so given our audience here at Canvas Con, this may sound a bit like a flight attendant giving instructions on how to use a seatbelt, you know, as if people don't know that already, but could you give a quick description of Khan Academy and its scope in education today? Sure, at a very high level, we're a not for profit organization, mission-free world class education for anyone, anywhere. There's kind of three pillars to that. One is we want to make all of the core academic content, material, and practice available from pre-K through elementary, middle, high school, and the core of college to everyone on the planet. So, including in many languages.

And as we give that we want to deliver it, or make it experienced in a way that is optimally engaging and useful for students. So a lot of what we talk about is personalization and mastery learning, meeting students where they are. And there's another pillar that we're really just starting to get started on, which is how do you connect the learning through Khan Academy, that personalized learning to kind of outcomes in the real world, whether it's getting into college, or getting a job or an apprenticeship. You know, where we are -- as you mentioned, it is -- Actually, I didn't even fully appreciate it, it has been almost exactly 10 years since that first real funding that we got. And, as many people know, this started as a family project back in 2004 tutoring cousins, it slowly evolved into a very significant hobby, I eventually quit my job, a year later we got that funding, in 2010.

And the journey ever since is math has been where we've been strongest. That's where we have content from pre-K all the way through calculus and statistics. Students can learn at their own time and pace, we have plenty of efficacy studies, but we also have very strong offerings in science, in the humanities. We have Khan Academy Kids, which is early learning, and that's math, reading, writing, and social-emotional learning. We're the official practice partner for the College Board around the SAT.

And, we continue to just try to drive efficacy and reach more folks -- both as independent learners and in classrooms supporting teachers. And right now I think we just crossed And you have to look back, although I suspect I know the answer to this question, did you have any sense of the scale that this was going to get to within the first year when you were doing the family tutoring, if you will? Or even when you quit and started this as a nonprofit, did you have any sense of the scale of this? It's an interesting question. It depends on the day you would have asked me back in 2006 or 2007. I think I struggle inside my own head, I can't get into other people's heads, so I don't know how much everyone else struggles with this, but my life is a constant struggle between confidence, bordering on delusion, and insecurity slash humility. And, so, in those early days I definitely -- you can see a lot of science fiction books behind me, I read a lot of science fiction books, and I did secretly dream, as I was creating tools for my cousins, and tools that in theory could scale, you know, software that could provide practice and feedback, videos that could scale, that maybe -- if it's useful for my cousins, maybe this could one day reach millions or tens of millions, or billions of folks.

And then I still remember, Sal, get over yourself, that's a delusion. If you can just help your cousins or the people in front of you right now, that by itself makes it worth doing, and I think that's kind of where I was 10 years ago, or even longer, 14 years ago. People could look it up, I wrote a letter to my 2010 Sal last year, and in that letter, it's kind of like writing to a younger brother or something, but I think if 2010 Sal saw 2020 Sal, or 2020 Khan Academy definitely would have been very pleased and would have said: Wow, some of those delusions weren't delusions. This actually can reach tens or hundreds of millions. You can actually can create a personalized practice and support narrative from pre-K through college, but what I also wrote to 2010 Sal is that there's something about where we are now where I feel like the stress level is even higher because we've kind of proven that these things can work, they be efficacious, they can scale, but if you look nationwide or globally, there's a lot more work to do.

If all of a sudden we could snap our fingers and everyone got 20 minutes or 30 minutes a day of personalized practice on Khan Academy, we would see test scores improve, we would see more people being able to reach their potential. So I think that's the challenge in front of us. Sure, but if you had to look back on it and, going back to the delusional Sal, what is sort of the secret? How did this actually get to be so widely adopted? Not just in the United States, but globally. So what's like the one secret of how that came to be? Yeah. I can't claim that I know exactly what the secret is.

I have some theories. You know, I wasn't the first person to imagine that kind of adaptive practice software could really help kids, that they have gaps in their knowledge. I think there might have been something around the timing of where the tools became easier, even one cousin on his own could be building it for their cousins, or putting it on a website so other people could access it. I wasn't the first person, by any stretch of the imagination, to imagine that video based instruction, even bite-sized video based instruction could be useful, even on YouTube I wasn't the first person to do it, but once again, that coupled with the software practice side might have been interesting. I think there was a lot that had to do with, if I'm really honest, about the how more than the what, which is it was a blessing that the first content I was making it for family.

I think in a strange way people can sense the love in it, that it's coming from -- The intent is pure and when you watch those first few videos, and hopefully Khan Academy, I still make a lot of the content. Personally, you feel that personal touch, you see that there's personality in it. You see that there's passion in it. And a lot of what I tried to do is I saw my cousins in the early days and then the early Khan Academy users, they were trying their best. They were fully capable of learning the material, but the way a lot of traditional academics happens is that it's very siloed, you don't connect the dots, the inspiration of like this is telling us something profound about the universe, that people dedicated their lives to discovering, you as a student get to learn this now, and that is often times lost.

And so I think, you know, I can sometimes be a little infectious in my delusion or in my curiosity, so hopefully there is a little bit of that. And I also found in my own academic career, I was the kid that wasn't satisfied memorizing a formula, I always wanted to ponder why that formula worked or didn't work, how it connected to what's happening in other classes, what it's telling us about the universe. And, so, on one level Khan Academy has been very pragmatic. It'll do the work examples, it will help you solve, but it also has that element of it helps you connect and really build intuition, which I think people connect with. Sure.

I like that you mentioned that because a lot of my observations in EdTech is there's a natural tendency to go towards magic algorithms, or like ultra-innovation from technology being the primary factor involved in how successful some change initiative is, but then we tend to overlook just simple aspects, such as what I've observed is just the simple aspect that I can, as a student now, I can work without any shame. If I have to rewind, if I have to play it slower, and then I get the tone that you're mentioning, that that's very significant. So, it certainly, as an outsider observation, I think a lot of the human element gets overlooked in terms of success of platforms such as these. So, if you look, now that you've done a decade. Khan Academy has hundreds of millions of users, obviously you've had your fair share of praise, but also of criticism.

And also noting that I think a lot of your approach has been to put things out and to learn from it, and adjust. What would you say -- looking back over the past decade -- what has been your biggest lessons learned? What's worked, surprised you the most in terms of being successful? But also, what are the biggest mistakes that were made? Something that either you would do over again or you're just saying, this was the biggest lesson learned of what I had to change. Yeah, how many hours do we have? [chuckles] We could just go through -- We're trying to pressure you. Yeah. Well, I think most of my mistakes -- I'm sure if I dig deeper there's even more fundamental mistakes, but most of my mistakes have been on the side -- Well, I mean, there's probably been advantages and mistakes.

You know, I've taken some of these personality tests and the one I took, this thing called Colors, and we don't have to go into depth on it, but I come out very red and blue, and what that means is people who are red like to just get stuff done, like tomorrow, by hook or by crook, like they want to get it done. And people who are blue are the big picture dreamers, so to speak, and it's not uncommon for a lot of entrepreneurs to be red and blue. The other colors are green is someone who is very process oriented. And yellow is someone who can really bring along other people for the ride. And, you know, my bias to action, which is red, and my bias to dreaming big, which is blue, has served Khan Academy in so many powerful ways.

On the positive side, just even having a mission statement back in 2008, 2009 of a free world class education for anyone, anywhere, most people would have thought that's kind of crazy. I was literally sitting in the same, I'm in a walk-in closet, in the same walk-in closet back then, this was the headquarters of Khan Academy. It's a nice walk-in closet, it has a window, I don't want you to feel bad for me. But so that blueness is very important, but obviously that bias to action: My cousins need help. Let me tutor them.

Let me create stuff. Let me see if it's helpful to other people. Let me start writing code. Very good. But as you start to grow as an organization, some of my biggest mistakes have been on me not necessarily communicating as well as possible, or me not driving process as well as possible.

The communication side, a lot of people are surprised by it, they see me giving talks, TED Talks, whatever else, writing. Like, Sal? He's into communication. He's made thousands of videos. That's what he does. But, that's a different type of communication than when you're trying to bring a lot of folks together as a team.

At every stage of growth of an organization, I've just seen that you can't over communicate, and I think that's where a lot of my mistakes have been around the team gets a little bit, they think is the goal and then that requires course corrections and rework, which is very frustrating. I think there's just a lot of tactical things, that if I knew in hindsight, I think my blueness and my redness tends to make me try out a lot of things, but if you try out a lot of things, you also have to refocus a lot, and you also have to prune, which is frankly more painful than starting things. So, you know, that's a cycle that I'm constantly going through. So, it's kind of like the same things are double-edge swords, they're both positive for our growth, but they are also things that can be painful and frustrating sometimes. Sure.

But given where we are today, and you have a different model than most -- Well, I wouldn't say than most, but there's a model of being a nonprofit and going for scale of impact, but then also working within the context of EdTech companies that are for-profit. But given where you are in the ecosystem, where does Khan Academy end and other EdTech systems begin? Not just in terms of technology, but also in terms of pedagogy. So, in other words, how should we view Khan Academy and where it fits in? Is it a supplement? Where does it begin and end compared to other systems and other parts of the teaching experience? Yeah, that's a great question. Talking about the previous question around needing clarity and proper communication, this is one of those things that we have also, like, as you grow as an organization, you have to get really clear on your communication here. The way we describe it is: We are a strategic supplement.

Because supplement feels like it's a nice to have, strategic supplement is like, no, this could be really powerful. It can really move the dial. It's not just a nice to have, it could be somewhat essential. And, so, I see this as a strategic supplement for personalized practice. When you look at traditional curricula, they do some things reasonably well: These are the standards, these are the lesson plans.

But if you think about the real gap that exists, no matter how good a given curriculum is, they tend to be pretty light on practice. Whatever practice they have, not particularly interactive, kids don't get immediate feedback, teachers don't get dashboards of where they are, and even where you do have, I guess you could say kind of digital worksheets that do some of what I just described, it's very seldom in a way that supports mastery learning and personalization. There's a ton of research that mastery learning and personalization are the best practices. Teachers learn differentiation in ed. school, but it's very hard to implement without some of type technological support in the classroom.

So I see Khan Academy's role as -- which is a very important one because arguably, most of the learning happens through practice and feedback in pretty much any domain, is being that place where students can get the practice, the feedback they need, and teachers can get the information and the data on how kids are fairing. So where do you draw those boundaries? I think there's another space on curricula providers. Those are things that we're increasingly trying to align more closely with and integrate with. I think you have assessment providers. These are folks that we have partnered with, whether it's the College Board around the SAT, or whether it's NWEA, around the MAP Growth Assessment.

And then there's learning management systems, and this is kind of the central way that teachers communicate with their classroom, and assign things, and understand where kids are. That's where I think there's some really interesting partnerships that we could forge. Because the problem at Ed. Tech right now is that it is so fragmented, we didn't even talk about rostering solutions and student information systems, and it is a very -- I actually paid them to not talk about those subjects. [laughter] Right, right.

So I think we have our lane, and being that place for personalized practice, or just the place for practice that we can hopefully guide people towards personalized practice. And the dream is if we can integrate much better with all of these other parties, so that it's a much more coherent experience for teachers and students. Well, as a follow-up though, are there differences in where those appropriate lines are, where the strategic supplement should be, and how to integrate within these other systems or -- Not just or, but in particular, teaching styles and how things are handled in the classroom? Does that change much by age level or grade level or even geographic? You know, first world versus developing world, does that present a significant difference of what the appropriate role of Khan Academy is? So, it depends on what layer. If we just take it at a high level, this notion of kids needing functionally unlimited, personalized practice and feedback, and teachers need to know where those kids are, that I think is a universal problem slash opportunity. Whether you're talking about four year-olds in Brazil or whether you're talking about thirty year-olds going back to college in Oklahoma, that's a universal problem.

I think how you implement it is different depending on the subject matter, the grade level, and the geography, and that's what starts introducing a lot of complexity. So, obviously for early -- You know, we have Khan Academy Kids, which is for ages four to six, we're hopefully going into second and third grade standards soon, that's holistic math, reading, writing, social-emotional learning, tablet based, it's manipulative based, but it has some of the same principles around personalization, mastery learning. Teachers can both assign and understand kids' mastery states. And then obviously, as you get older you can put a little bit more on the student, expect a little bit more independence, they don't want to see as much of the cartoons and the animations. And it's different from subject to subject.

Math is where we're strongest right now in mastery learning, we're planning on making a big investment in science, we already have a lot of science. Where you can, skill specific, give someone practice and then give them mastery challenges, or spaced repetition and multiple context switching modes to make sure that they really grasp things, while English to language arts, you have a fewer set of skills that repeatedly get exercised in multiple reading comprehension levels and multiple contexts. And so we have worked to kind of -- How do we do personalized practice in that domain? And then as you go into other countries, I think the really tricky thing, we have been doing the of Khan Academy around the world, we are trying to create regional mappings to standards and we are also creating some new content for the various regions, but then the EdTech spaces in these countries are completely -- the rostering systems - That's another -- The fragmentation makes it a little bit difficult, as well. One thing we have to acknowledge, it's obviously not just the past decade of Khan Academy, and looking forward, we're in this unique year, hopefully unique year of 2020, what if you took this similar approach of what have you actually -- What does Khan Academy experience over the past 6 months and what is that change, in particular for what you expect to see moving forward in education? What is the impact of Covid-19 pandemic, both on you and your organization, but also on how you see EdTech playing out to serve these students? And I realize that's a loaded question, but we have to acknowledge it. Right, right.

No, we -- like everyone -- we first caught wind that something was going on in February when we saw our Asian traffic increase, and I remember reading this one letter from a teacher in South Korea saying, "They're using Khan Academy to get them through school closures. " And I was like that's wild, a whole country has shutdown their schools physically. And then it was a few weeks later that it started to become clear that it was going to happen in the U. S. and much of the rest of the world.

That's where we kind of did a full court press, we kind of had a war room situation where we made sure our servers, our infrastructure could handle the increased load. We started planning webinars for parents and teachers to support them on how to do distanced-learning, putting out schedules for what homeschooling could look like in this world. And then that first week, we saw our traffic get to about 250 to 300% of normal. On a normal school day in the spring we would have seen about 30 million learning minutes a day, we were seeing about 80, 85, 90 million learning minutes per day during the spring. Teacher registrations were 10x normal, parent registrations 20x.

So we were just trying to keep up with it, and as soon as it became clear that these closures were not just through spring break, which a lot of people were talking about in those first few weeks, that it was going to go through the end of the school year and possibly longer, we realized we would have to create more supports, create a whole effort to keep people learning through the summer, which frankly, is always a problem slash opportunity, but even more important this past summer. And we created things like Get Ready for Grade Level courses, which is a way for teachers and students and parents to evaluate how ready a student is for grade level work by assessing them on the prerequisites. And if a student does fine on that, that means they're ready for grade level. If you they don't so well, if they do 40 or 50% on the course challenges on the Get Ready for Grade Level course then they have an opportunity, in a personalized mastery framework to work on those prerequisite skills. That could happen in parallel as their classroom goes into the grade level work, or the classroom can really spend two or three weeks at the beginning of the school year making sure everyone has those strong foundations.

Because we see over and over again, that's always debilitating in any school year, kids have all sorts of gaps all over the place, this year it's that much worse. And on top of that, we've just been trying to provide as much thought leadership as possible, and what does good distance learning look like? Because most schools had to spend most of the last six months just trying to figure out, can they open, how to open, how to put desks six feet apart, what's the policy? And not a lot of folks have had been able to develop a strong view of what good or even adequate distance learning should look like. So you mentioned at the beginning of your answer about the servers. I mean, so we're talking not just a unique moment in demand, and I'm also hearing you describe sort of unique usage as they get us through the shutdown, which has implications about how supplemental you truly are in this transition, but at the same time server costs were going up. So it sort of raises a question of sustainability.

How does this affect a nonprofit in terms of sustainability where you're having to pay a lot more, particularly for server costs at the same time that there's such a great demand? You know, I think the good thing is -- Folks, I think have always intellectually understood our social return on investment is kind of off the charts. We're the budget of a large high school and we reach a good chunk of kids around the world in meaningful ways, and ways that have been proven to be effective. But Covid, to your point, Khan Academy -- I think for a lot of people they were already dependent on this type of resource, but I think they're even more so. Khan Academy, especially in math, but in other subjects as well has been kind of the go-to for practice, assessment, feedback on a regular basis for much of many classrooms around the world right now. So I think hopefully the philanthropic community, the corporate community continues to realize that this is a very powerful social return on investment.

We do have some elements of what we do as a not for profit that creates revenue for us. Like the College Board, they give us resources so we can do the free SAT practice, LSAT gives us resources so we can do the LSAT, and so that I guess you would call earned revenue. We have a few other enterprise partnerships with districts where we co-resource that, so that helps, as well. It's a constant -- I think any model, whether for-profit or nonprofit, has its positives and negatives. And my aspiration for Khan Academy has always been to have, to try to do the best of both, to have the talent and the nimbleness, and the innovation of the best for-profits of the world, while having this very deep mission focus and being able to kind of be like full-throated effort on the mission of a nonprofit.

I will say, from my perspective, that a lot of the challenges you're talking about are common in for-profits, as well, particularly just the core challenge of rising demand, a need to be able to adapt to the students, and serve them as best possible, but at the same time the cost structure is getting just more and more difficult, including the budgets from schools. So I think it is a common challenge that we're seeing across the space in this area. One of the other things that we have to look at, and it's not just because of this year, and a lot of the social unrest, but it's a general challenge of how do you reach and serve underserved student populations? So from your perspective, what is your role for underserved students, whether it's minority, first generation going to college, students without adequate resources, where does Khan Academy fit in there? Should we think of Khan Academy as serving, needing to have a minimum level of socioeconomic capability to use it, or is it viewed as rising them up? So, what's your role in helping underserved student populations? Yeah, so clearly the intent of Khan Academy, even in our founding mission statement: Free world class education for anyone, anywhere. Implicit in there is that the world needs this because a lot of folks don't have access to world-class education, and even if it's there, they might not be able to pay for it, so that's kind of the free element of it. But obviously in order to use what we're creating, you do need some baseline device access, internet access.

We try to make it as accessible as possible, we have phone apps and things like that where you can get a pretty -- the full experience of Khan Academy, and we partner with other nonprofits who are taking some of our content and putting it on offline servers, those are less optimal experiences, but they're better than nothing for a lot of kids. Going back to the communication conversation, we've become much more clear that a lot of what we do helps everyone, but implicitly the middle class or upper-middle class or affluent kids already have something, and we want to be world class, so we want it, so that if we give them something, it's actually better than what they already had, so they get a benefit. But the lower income kids, the kids that have been historically under-resourced or under-served, their status quo is here, so if we can provide them this, it is a massive improvement. But one of the reasons that we've started even more deeply engaging with school districts over the last three or four years is we can create great content and just hope people come, and we get a lot of folks, and in fact a lot of kids from underrepresented communities coming, but these are pretty motivated kids that are coming on their own. So if we really want to reach the kids that are most vulnerable, most on the fence, you really have to go in a formal through districts and through schools, and so that's why we've been doing this Khan for Districts partnership.

We've been partnering with folks like NWEA around assessment so that we can really reach, hopefully, all kids eventually. Now, looking forward, and I guess this question is not just Khan Academy, but more just your view on EdTech, in a succinct way, hopefully, what is the role of technology in the coming years of addressing educational challenges we have? And that obviously includes coming out of the pandemic, it includes economic turmoil coming from so much of the shutdown, so how should we think of technology in the education space and where it has the greatest ability to transform student learning? Yeah. I like to reframe it a little bit because many of us are technologists, I'm a technologist, and as a technologist you often have a bias to saying: I made something cool. Let me figure out how to make people use it. The real solutions happen the other way around.

You identify a real issue, you figure out, you have a thesis on how to fix it, and then you say, what are the tools that I need to be able to execute on that thesis? And, so, the way I think about it, let's put all technology aside, let's think about what does an ideal education system look like or an ideal learning experience look like? An ideal learning experience, I think number one, I'm a big proponent of competency based learning. That if we can clarify the outcomes that folks need to get to and ways that they can prove their capability at that outcome, that has huge positives to it. One, it levels the playing field on the credentialing side because regardless of who you are, where you're from, what school you did or didn't go to, you can show that you're as good as anybody else. It levels the playing field in terms of cadence. Some folks it might take two months to get to that level, some folks it might take two years, but as long as you get there, that's all that matters.

While right now our system tends to judge people if they take a little bit more time. Then it just completely opens up how we get people to those outcomes, it kind of allows more diversity of approaches, and then as you open up those diversity of approaches, that's where I'm a big proponent that allowing for personalization is a big deal, and it's not just personalization on kind of a screen or on technology, it's personalization when human beings get together. And it's also increasing human interaction. So, are there ways that kids can get a lot of practice and feedback at their own time and pace? Are there ways that teachers can know in real time where kids are? And then use that information when they get in the room together, or I guess when they get on the Zoom together, and to do more focused interventions. Are there ways that teachers can facilitate the students to help each other? So, that's the kind of the school, the education experience that I think most of us, that I aspire to, and so then you say, what are the tools there? Some of it can be pretty low tech.

We have to create a mechanism for folks to prove what they know. And we can brainstorm there. We have to create a mechanism for students to learn at their own time and pace, and get practice and feedback, and report to teachers. There's a little bit more nuts and bolts. We either make ways that all of this integrates, maybe certain people are going to be -- If we create this -- Many different pathways, you need a way that those pathways can integrate in some way, whether this is learning management systems, or various platforms that make it easy for people to say, okay, this is what I'm going to use for math, this is what I'm going to use for foreign language, etc.

, etc. So, I think EdTech is going to have a -- A lot of what I just described is going to be very hard to do without EdTech, but I think that is the lens is that we need to speak towards the real pedagogical goal we are doing, and we have to be accountable that it is actually working. And then think about how do we make it as engaging and as useful as possible? And I remind the team at Khan Academy of this all the time, because I think any organization -- for profit or nonprofit -- it's very easy to make it about you. I mean, look, I can be competitive sometimes. We all have our egos and I want Khan Academy to win, sometimes.

I've put a lot of work into this, but I remind everyone that it isn't about success of Khan Academy or any one organization, it is: what can we do so that the world becomes better? If someone actually out-innovates us to create a better solution, that's good. We should actually cheer that on and we should learn from that, or we should learn to complement that, or we should learn to integrate with that. So I think if we collectively have that type of a mindset, I think it's going to create much better experiences for students and teachers, and I think if we do that, then EdTech itself will thrive. And are you optimistic or pessimistic in regards to the direction we're going? Are we going the right direction to get there? Yeah. I think right now there's a lot of good innovation on particular solutions across the board, but I think the fragmentation is a real problem in EdTech.

If you talk to any teacher during Covid, the biggest pain point is they found -- They're using Khan Academy for this, they're using Raz-Kids for that, they're using Canvas for this, you know, they have this rostering system. It's a lot and it doesn't have to be that way. So, I do hope that we do get -- That we have a little bit more clarity in the space over the next five years. And I think it's going to happen. I think you're definitely seeing some go-to players in these different domains, whether it's on the learning management side, whether it's on the personalized practice side, whether it's on the curricular side, and on the assessment side.

So, I think if some of us get together, we can help simplify things dramatically and make it a much better experience for everybody. Great. Final question. You had said earlier that if you don't get excited by Euler's identity then you have no emotion. Do you still feel that way? Is there anything else that gets you more jazzed than Euler's identity right now? For those of you who don't know Euler's identity, it's that there's a very strong mathematical justification for why E to the I pi should be equal to -1, and if that still has not made you excited, and famously I said in that video, "If that doesn't blow your mind, you have no emotion.

" E comes from notion of continuous compound interest. You also see E in many forms of exponential growth, for those of you who remember your calculus, E to the X is the only -- F of X equals -- X is the only function that if you take its derivative, it is itself. So, E is the kind of magical number that doesn't repeat, very cool. pi is also, most people are more familiar with pi. Ratio of the circumference and the diameter of a circle, this number that never repeats, it has all of these interesting properties on it.

So it's coming out of this completely different geometric part of our mathematical, you know, the mathematical universe. I is something that we kind of invented in order to understand solutions to polynomials, and understand some phenomena, the square root of negative 1, but it has unlocked a whole domain in mathematics. And the fact that these three numbers are connected in this way, this very beautiful way. E to the I pi power is equal to -1, not negative 2. 5, negative 1, or it can also be written as either E to the I pi + 1 is equal to 0.

So it has all of the fundamental numbers in one simple equation, some people call the proof of the existence of God, it gives you a hint, just makes me saying this right now gets me a little tingly, because there's something deep about the universe that's being communicated in math. We're just glimpsing, but what I tell everyone and I tell everyone at Khan Academy, we shouldn't teach anything unless we're excited about it. We shouldn't write code unless we are excited about it because the end users going to feel that. And if we're excited about it then it's going to be infectious. That captures so much.

So much of education -- Sorry, you got me riled up! [chuckles] Well, you've got the emotion. But so much of education gets down to the human element and motivation, and support of students, and connecting, so I'm glad that you feel that way, and in particular being an engineer myself, it's good to hear this. But I really appreciate your time with us, and I appreciate your honesty, and your ability to just be very transparent, and hold your organization out, and what you're doing, and congratulations on all the success in the past decade of doing this work funded, but even 14 or more years overall. But on behalf of everyone here at Canvas Con, thank you very much for your time, Sal. Great, thanks, Phil.