Teaching for the Future: Computational Thinking in Education with Gregory Benedis-Grab

Hello everyone. Thank you for joining us today at another academic Innovation support live stream. We are joined today by Gregory Benedis Grab, who is an expert in education and computational thinking and a lot of fields that we at Wolfram are extremely interested in. So thank you Greg for joining us today.

Oh, thank you for having me. I'm excited to be here.

Our sincere pleasure. And as always, I know Shartoff am joined by John McNally.

One reason why we're particularly looking to hearing what you have to say, Greg, is that we both have a background in education before joining Wolfram Research. And so this is an area very near and dear to our heart. So I'm very much looking forward to this conversation.

So, as someone who has studied technology and education and computational thinking, I wanted to ask you what's your personal history with technology and education? You know, the, the experiences you've had perhaps that led you to these fields or experiences in studying other things that brought these sort of ideas into the forefront for you. So.

Great. Well, I think I'm going to share. I. I took the opportunity to make a little Mathematica notebook here.

Fantastic. And by the way, as per usual, anything in the notebook will be made available to everyone who's watching through a link later on once we have a permanent version of this video up. So if anything moves fast enough that people would like a second look at it, it will be coming up.

Great. So I. The first thing I put here in the notebook is just a picture of my wife and I, we just recently moved to Boulder. And later in the talk I want to talk about some of the work I'm really excited about here. But we love it here in the mountains and all of that.

Some of the most naturally beautiful landscapes in the country, of which there are quite a few.

Oh yeah. And the next thing I put in here is a little seal. It's from. So I. A number of years ago I received the Presidential award for excellence in math and science teaching. I got to meet President Obama, which was highlight of my career.

This was in 2010, correct?

Yes. And I. The reason I bring this up is because I think a lot of my perspective on computational thinking and education come from my long career in the classroom and done various things. But I, I feel like some of the best experiences in my life have happened in the classroom and it gives me a lot of perspective about the educational system and kind of what's needed. And I feel like, like one of the things I want to talk about today is just sort of the need to really rethink some things about education. And I say that with some trepidation because I feel like everyone's saying that these days. Like everyone's talking about disrupting it and changing it. And I think I'm. I'm going to say something slightly different. I don't know if it's so much that we need to just disrupt it as we need to kind of understand more deeply changes that have been asked for for a long time. Maybe going back. Like, there's a lot of stuff I might draw on from the 60s and 70s, which. Which is when computers were first starting to kind of become part of. Of the. Of the global social consciousness. I would even go further back. And in fact, I put in a picture here of John Dewey. I feel like a lot of his writing about education has inspired me. And I think a lot of it is just about kind of understanding more deeply what education is about. And so another picture I put in, kind of connecting to this personal story. This was my foray into the computational space. I don't know if you guys recognize this artifact. It's a Commodore 64.

Ah, I recognize the name more than the appearance there.

You can see there's a little logo here. But that was. My parents got me a Commodore 64. And that's sort of what launched me into kind of getting excited about computers and thinking about things. But I think, like, the reason I put the picture of Dewey up here is because I think, like, what I read in a lot of Dewey's writing is just about kind of making. Putting kids at the center and letting students really drive the process and empowering them in this process and making education about kind of figuring out the world and figuring it out with guidance from a teacher, but ultimately about creating this space of discovery. And I think some of that discovery has been lost somewhere along the way. And my wonder is not. I feel like a lot of. A lot of the stuff I see around technology, it almost feels like the. The essence of it is we're almost like programming the children like. And I even interesting the AI stuff. It's like we're going to give them the right guidance so that they learn this concept, or we're going to set up the environment so it's optimal in terms of, you know, their progress. But I feel like. And you know, this, I. I felt this in Dewey. I feel this in some of the stuff from the 60s and 70s, like Seymour Papper with Mindstorms, maybe even more current stuff as well. But it's like the. The real answer is the Kids programming the future. And I think what enables that is that there are that technology makes available these tools that empower children. And in fact, the next thing, I just made this little grid from 1 to 25. And I was thinking the reason I was playing around with the interface, but I was thinking about the spreadsheet, because spreadsheet, which I think was invented in some class at mit, like some student was watching some professor like a race and update all these values on this huge chalkboard and they're like, this is just ridiculous. There's got to be a better way to do this. But the spreadsheet, which is the, which.

Is the thinking behind so many innovations, isn't it?

And, and the spreadsheet is such a simple concept. Like it's just a, it's a two dimensional array of values. And yet it's been transformative. And I, I was always struck in a school setting. A lot of teachers aren't comfortable using spreadsheets, maybe in math and science, but a lot of, and it's such a powerful tool. Like it, it can help you keep track of things. Like there's all these things that it can do. Even though it's such a simple, almost ridiculously simple tool, it's a cognitively powerful tool that allows you to expand what your brain can do. And I think that's true. Like that's why we invented language. That's why a lot of these tools exist. And now we're in this space where there's really powerful tools that young people can have access to. And that, that's why I was excited to talk to you at Mathematica, because I think that's a very. Aligns with a lot of the thinking at Mathematica and behind the tools that you guys produce.

Yeah, I've already heard so many things which I have a philosophical agreement with. One of the things that you just said that really resonated with me was I remember distinctly a conversation back when I was learning about constructivist theories of learning. And the discussion was around, well, sometimes we call what we're doing scaffolding because the idea is that we're providing experiences that, that eventually will get to certain learning goals. But maybe actually a better word would be trellising in the sense that you, you build a trellis, that a vine grows up. Right. And, and so this I think gets into the difference between, you know, you know, we, we don't want to program the students. Right. That's, that's actually not the goal with any of these tools. You're Talking about. But exactly what you were saying is that we want to provide some infrastructure so that it empowers them to explore and to learn and to grow.

That is just such a great distinction. Because, of course, I know we're stuck with scaffolding at this point. The term gets used enough, and we pretty much just have to go with it. But the people who make the scaffolds are also the ones who stand on the scaffolds and build the building on top of it. Like, the scaffold doesn't allow some secondary person to then put in the work to construct something. Whereas trellis is so much better to say the kids are the vine that will grow up because this structure is here and will be able to themselves explore the environment and find the best way for them to grow. That's. That's such a better metaphor.

No, I like that metaphor as well. And it also emphasizes that the scaffold plays a temporary role. Yes. Meant to be something that's always there. You're kind of giving them this jump start, so then they can kind of grow past the. The. The structure and really, you know, take things on in a real way.

I don't know enough about trellises. I would have thought that they stay there forever, but if that's the way that they're used, that's even better.

Mine doesn't.

Yeah, I wish I could take credit for that metaphor, but it was actually one of my colleagues back when I was in graduate school. But, yeah, I. I've loved it ever since.

Nice. That's awesome. So that sort of. So that kind of emphasis on tools and the other thing I was going to say about the tools, which I think relates back to some of the. Some of the work that you guys are doing at Wolfram is like, these tool. These new tools that really rely on computation and computational thinking, they're not only this powerful model, but it's also how the world works. Like, ultimately, humans are computational creatures. Like, if you think about the process of, like, how our DNA works and how our brain works, like, a lot of these components that construct us are built around a computational approach. And so I think as we embrace these computational tools, we can get. We can go really far in terms of this trellising concept, but we can also get more insight into ourselves and what we need to do. So I think it's sort of like all. Everything's sort of coming together in a perfect way at this particular moment in history that I think is really exciting.

So I have both some specific and Some general questions related to, you know, your thoughts about how, from a constructivist perspective, you know, should we go about some of these things? But I want to also make sure that you get a chance if there's other things in your personal history that you wanted to elaborate on before. We sort of nerd out about educational practice.

Yeah. Because we do. See, you're about a third of the way down the scroll bar of this document here, so.

Well, there you go. Well, I was going to jump into. So this is sort of a segue kind of to what you just said. I wanted to talk a little bit about the work that I'm doing here at bowl at CU Boulder. We have a group here called School Wide Labs, and essentially we're building. We've been working together with Denver, Denver Public schools for about 10 years in a research practice partnership. And we've been developing some curriculum units. But really, it's much more than just a set of units. It's a whole way of thinking about what kids can do.

I'd love to hear more about that.

And we've decided on the micro bit because it's this powerful physical computing platform, but it really can engage. We're sort of pitching this to grades three through eight. That's an area where I think there's a need for a lot more kind of thinking and innovation. And this platform is just easy to use. It's easy to get involved with. Here I have a little picture where we have the micro bit and we attach this LED strip to it. And with those two components, you have access to a whole range of sensors on board on the micro bit, like temperature, volume, light level, magnetism. And if you want, you can get more sensors and attach them. There's three extra pins you have down here that you can attach more stuff. But the thinking behind this is we teach students how to program sensors. So they learn some of the make. We use make code, they learn some of the programming. It has a great data logger feature. So they can collect data, they can power it, they can control it.

They.

And we have this opening curriculum where they just start to think about all the possibilities. And we actually have them construct, like, a model of how they think it's working. And they develop that over time. But really the vision of this work is that the students start to find the places where they can use this technology in the world. So we've had teachers, you know, where the kids say, you know, I want to measure the volume in the cafeteria, or I want to see how we're using like how is power being used in the building? By checking the light in different classrooms and tracking it over time. Are we being as environmental as we think we are? Or they might check the temperature, like is this building properly being heated? But the idea is that the students start to build the curriculum and to build the questions that they want to answer and build the investigations and realize that, you know, a, this kind of technology can be transformative. And I'll give an example that's a little different from what I'm doing now. So when I was in, I was teaching at a high school in Brooklyn and one of my colleagues taught a physical computing class. And this used the Arduino, which is a similar technology to the micro, bit more, it's more appropriate to high school. But in his class there were students who created a. This was in Brooklyn and they created a subway board. So they connected so there's like an API where you can get subway data. And they found when are the next three trains coming to the stop that stopped right near our school. And they displayed that in the student center. And you know, this isn't like novel information, like any kid can get this information on their phone. But what they realized is that by having it publicly visible, it had a different impact on the community. And people were excited by the fact that there was this display that had an impact on their lives. And they realized, you know, we can be the impact we can be. And they knew that some, one of their, you know, fellow students had made this thing. And, and I think that's what this work is about, is really getting kids. And what I love about the school wide Labs work is, I think it also allows for teachers and students to learn alongside each other. It provides this almost de facto scale up model because we just, we need to all learn this stuff. There's, there's just not enough people who understand all this work at this point. And we need systems that are going to get us doing this kind of work that are easy to get into and easy to kind of grow.

Interesting.

Absolutely.

As someone who works at Wolfram, my first thing I was wondering about is can the Wolfram language be integrated into this sort of system? But what I love about the system you're describing is how flexible it is that it's, it's approaching things similarly to the way that a human body works, that it can observe its environment and be aware of a whole lot of things. Some things that someone might not even think about ahead of time. But it also got me thinking. I'm a huge fan of problem based learning. And it sounds like this is a lot of the philosophy of that, but a little bit beyond it, because not only are students finding their own ways to solve these problems, they're also proposing the problems. And I wanted to ask, do you think there are any difficulties that arise with having the students that much in control of the process or does it make it even better?

I mean, I think that's where we need to. That's the rethinking part. We need to think. I mean, you don't jump into this work where you tell, you ask, okay, what are the problems you want to solve? You need to create an experience where they can start to see that they're defining these problems and they can engage in this work. Like, I think, you know, for a long time, I think what's driven school reform is curriculum. And I think curriculum is really powerful. But sometimes the way curriculum is manifested has a staticness to it.

Yes.

That just operate in today's space. Like when you think like, you know, there's certain areas, like if you're studying Newtonian mechanics, it hasn't, I mean, there's things that have changed, but there's a. Some things stay the same. Yeah, these more tech infused spaces changing too rapidly for a static curriculum. And so what we're proposing here is something that's a little bit more dynamic, where kids are creating things. And ultimately, you know, all of the literature is saying that we need to give students, you know, we engage them in professional practices. Well, what, what do professionals do? They don't get assigned problems.

Yeah, well, to a certain extent they do.

To a certain extent they have to define the problems. So like architect goes into a space, they look at the landscape and then they think, how am I going to frame and construct this problem? When an engineer goes into a space, they think, how am I going to construct? And like there might be some things that they have to solve, but they're doing a lot of the construction of the problem space. And those are key skills as well that we need to start to get kids on board with in some way. Now, is it going to look the same for a fifth grader and, you know, a professional person? No, it's. We're going to have to use our knowledge of kids and how to engage them and make it appropriate. But I've always been inspired by. There's a book by Jerome Bruner called the Process of Education and he made this bold claim that you can, you can teach any concept at every, at any level. If you adjust the Framing and the, the nature of the concept to be appropriate for that age student.

Yeah.

I believe, like, I think, and I think technology puts us in even more stronger position to do that.

Yeah.

So if I'm paraphrasing you correctly, you're saying that this rethinking that we need to do is not at the level of curriculum, it's at the level of practice and of the community atmosphere in our classrooms. That, you know, what the students expect is possible is really the thing that we need to rethink how we, how we help them reframe what's possible rather than, oh, I'm going to learn some shiny thing. But what's possible is that I can actually define the problems that are relevant to me and learn by solving them.

Yeah. And really giving more agency to all these different actors, whether it's the students, the teachers. I mean, I think there's a real need to make this a collaborative effort where we're sharing resources between schools, we're sharing resources between educational resource researchers and practitioners. Like, I think we need to bring everyone together on this more and make it more of a. It's almost like we're engaging in social change. This is a.

Now maybe we could think of this.

As a revolution in a way.

Obviously, the idea of everyone contributing and collaborating is one that objectively sounds good. But let me ask, I don't think I really heard in your description of these sorts of materials or philosophy of teaching where the integration of a whole lot of other people would come into it. So, so let me ask, what, what, what is the, the part of the process that, that you see that improving?

Sure. So in terms of the, the materials that we've created at School Wide Labs, all of them have been created as part of this research practice partnership. So they were developed and refined and thought about in classrooms in Denver with teachers working together. I think the, the place we're trying to figure out now is how do we scale that up? Right. How do we. Because it's more than just materials. Like, I think what made them really successful in Denver is that people know. They know. They'll. They'll know the teacher that worked on them and they'll have this direct connection to the process of the creation of this. And I think within this framework of like, you know, having students driving the inquiry of finding the things that are of interest to them, I'm hoping we can create a space where this process can go on and students can start to suggest, maybe students can share data, maybe students can share ideas, maybe, you know, we can really make this truly place based where they're really thinking, how, how is this different in my community than what happened in Denver? And making all of those things. So that's.

It's also worth acknowledging. I'm sorry. It's also worth acknowledging the extent to which students already are doing those things. I read at one point that a conversation around the fact that a professor's idea of cheating is often complicated by the expectations of students in the digital age, that everyone understands, you shouldn't be saying this is something I did when someone else did it, but that students expect at this point, and maybe rightly so, that information should be freely available and able to be shared at any given time, because that's the world they live in, that's the world they've always lived in. And, and it's, it's possible that some of these avenues of conversation, of sharing information are ones that could become more part of the educational process. I mean, we've already seen things like social media groups being used for classes, having particular pages made for them, something like that. Who knows how far this is going to go.

And I can also say this is part of the philosophy behind having things like a data repository, the ability publish notebooks and to share URL. Right. So this idea of promoting open science, which is really what we want to have students replicating in terms of their classroom practice, as you mentioned, that having actual professional practices as part of their learning process is really the goal. And you correctly point out all the literature points toward this. And if you look at any of the standards that are a result of these sorts of studies, this all has this in common. So I also wanted to ask as a sort of a follow up to this, about the, the challenges that we have where the technologies are new. You know, we as teachers grew up without having some of these technologies available. I'm thinking specifically of things like ChatGPT now. So what, what are some ways that you've found effective, where you think will be effective in terms of helping instructors be comfortable using these technologies in their classroom? Because there's this, there's also a line of research where teachers, when they're not comfortable using things, they're less comfortable teaching students how to use these things. And so how do we circumvent these sorts of problems and empower the teachers as well as the students?

Yeah, and actually I have a lot of different thoughts on that. One thought is just thinking about the design of the, of the tools themselves. I think this low floor, high ceiling is really important.

I'm not familiar with that terminology.

So this is something, I think I first heard it come from the scratch group at the MIT Media Lab. But the idea of low, it's easy to get started using something and it.

Has a lot of potential.

Make the barrier to entry very low so that people can start using it. And you know, this is not necessarily unique to that. Like, I think that's what makes, you know, saleable products work. Like if it takes you too long to learn it, you're not, you're not going to use it. So Shameless Plug.

But Wolf from Technologies has always struck me as something that has an unusually accessible opening point for something as powerful as it is. So yeah, I absolutely am worthy of that.

Of most things that have been successful. Like you think about Google Docs, like the reason they've been very successful is they're very intuitive and some of that intuition stuff is also built on understanding the landscape. Like Google understood that a lot of people use the Microsoft products, so they copied some of the workflows of that because they, that's, that's a good way to have the floor low. The other thing is that high ceiling. Can you do a lot with it? Can you take it so that students can run with it and take it far? Because otherwise it feels like you're being limited by the tool. And so I think you need both of those pieces. So I think tool design is important and it's, it's important that we're really thinking about the design of tools because these tools are here to like enhance our cognitions, enhance our learning. And so we need to be talking about the tools that are being made available in the public space. And maybe that's one of the goal, one of the things that these governmental structures and things like that can help like support and make sure that the, you know, the design is meeting the needs of the education system of this kind of approach.

So in your opinion, which do you think is preferable? Is it building tools with something like a large language model at their heart or is it building classroom practice around some existing thing like chat GPT? Right. So you know, for the difference that I'm trying to make, the distinction I'm trying to make there is that, you know, certain tools, you know they exist. And so then it's a question of how you're going to use them as part of your practice. Whereas you're also talking about this, this co creation of tools, right, where you can take, you know, a technology, let's say a large language model and you can incorporate that into the design of something which, okay, maybe it's actually narrower in scope than, you know, everyone, but it's being designed by teachers and researchers and their students as well for some purpose. So, you know, is there, what, what can you say about the difference between those two kind of baselines of the existence of the tool? Right. Something which is kind of out there and you're, you're adapting or something that you're, you're really co creating more from before it's even released.

I mean, I think you're sort of doing both. And some of it, I would say, is almost a reverse engineering. Like one of the things we do in the opening unit with school wide labs is we create this thing called the classroom display. And typically we let the teacher like we have sort of a base model and then we let teachers can take it where they want, but one version that a teacher created it, the display has this LED strip and it cycles through. It first shows the noise level in the room, then it shows the light level, then it shows the temperature and it cycles through. But you present this to the students and you ask them what they notice and what they wonder and they start to make inferences about what's going on, which, like, is it cycling through things? They're trying to make sense of it. And we actually take them through a whole activity where they develop a model of this and then as they start to learn to program the different sensors and their model becomes richer and more informed. So I think ultimately we have to. There's things in the world and we can't be in denial about that. Like, we're not like ChatGPT out there. It's there. We as educators, we don't control the world. Yeah, we have to work with the world that's there. But I think we can. I think ultimately what this work is about is, is engaging with some of those tools that are there in an inquiry way, but also stepping back and saying, how did they construct this? What are some of the liabilities of the. Like, maybe we can have kids, like maybe not in middle school, but in high school. I've definitely walked kids through the digit classifier and getting kids to think like, what does it mean to have a language model? Or just any, any kind of machine learning model. And there's, there are, there are challenges in creating a machine learning model.

Sure.

You need to start to think about and then you can start to think like, what is ChatGpt producing and what is it not? Like, what is it not doing that? Our brain tricks us into thinking that it might be doing. And I think we need to let, we need to get more comfortable. So I think in addition to the kind of low floor, high ceiling, I think there's a lot of work to do around kind of building up that comfort level with teachers. And maybe that involves some professional learning, maybe that involves having certain experiences. But I think it also connects back to this community piece of really kind of working. Like, I think we need to connect with each other. And I think often teachers, like I've noticed in professional learning sessions, when you create this community and you start to work with materials like the micro bit, if you create a space where there's safety in making mistakes.

Right?

Yes.

People can be much more comfortable and we need to build those structures in. I think one of the features of education is it can be very isolating for the teacher.

Yes.

That works against some of this community building safety piece that I'm talking about.

Yes. No, I, I couldn't agree more. I've read pieces to, to this effect. So I'm curious, can we go into more detail about the sorts of experiences in community building you think would be most useful in telling the story of how a tool like Chat GPT came into existence? You mentioned that you had taught in high schools showing digit classifiers like example. I think it's fair to say that in most high schools, students actually don't get that experience today.

Right.

That if you look at the vast majority of schools, that entire story of how we got to this piece of technology is kind of absent from a lot.

So at the best, you get someone describing how an end user might be able to use it, but very little beyond that.

So what can we specifically say about bringing that story back into people's learning in terms of both the teacher's professional development and in terms of what that looks like for students too?

I mean, to be fair, the students that I walked through were definitely like students who chose that route. And I don't think that's something that's not necessarily the right low floor place to start this work. But one thing that I have done with students of all backgrounds is use some of the. Was it the Google AI playground, I believe it's called, they have a website where they have some very simple AI tools that you can play with. Like you can create an image classifier by feeding it photos and seeing what the confidence level is based on the data that you've given it. And I've done that activity with middle school students. And so I think, I think we need to Face the reality of what's out there. So I don't think, I think we can, we can experiment with ChatGPT. One of my colleagues who is an English teacher, she gave an assignment where she, she had a question on a book that they read. I don't remember the book, but she get, she took the book, she took out the question and she said, okay, I want you to feed this into Chat GPT, get an answer and then give me an analysis of ChatGPT's answer. Like how good do you think that answer is? And then the follow up piece was try to prompt it to produce a better answer. And what prompts did you use to. And, and how successful? So I think there's room even within these things that are creating a lot of fear.

Right. I completely agree.

Same inquiry approach of just, well, what do we think about this model? And do I, you know, there's, it's an amazing thing, but it also has a lot of limitations that people aren't seeing. Seeing as well.

Oh yeah, yeah.

Not fully understanding what's going on.

Now there's an interesting wrinkle that I want to get your perspective on. So these kinds of exercises where it's a, it's an excellent, I agree. Critical thinking exercise for, you know, high school, even middle school students to say, okay, I'm going to take what this thing says and I'm going to evaluate it in the context of what I know and go from there. But I think some people perhaps rightly have less confidence that that's exactly the right way to go with, let's say a young elementary school student where you know when, to use the example of when you're learning a times table. Right. It's not commonly thought good practice to, you know, hand the students an incorrect times table. And then at the end of the lesson, say, by the way, can you critically think and figure out where the errors in this are? Right. So in other words, what do you see about ways to helped students transition as they grow older and their critical thinking skills develop to the point where, you know, these exercises that I think we agree are excellent ways for older students to be able to use the generative models that are out there today? What are the sorts of things you feel are important for when we're dealing with elementary age students in terms of helping them reach that level of critical thinking where then you know that that is exactly the activity that you want them to be doing?

Yeah, no, that, I mean, I think that's a good point. I think it looks different at different Ages. I think one of the things that like one of the motivations with school wide labs of make of bringing in this physical computing place piece is that it brings it back to the physical world and there's something very concrete and real about the physical world. So if you're writing like and this is the other approach that sometimes high school teachers take as well. It's like have the kids write in class on a piece of paper then you see what kind of. Then they could develop those writing skills. I think, I think one of the areas where like I, and this comes from my long history of me. I've. I've spent a lot of time teaching elementary science and one of the I've realized is sometimes as adults we forget how many times we had to experience something.

Yeah.

Before it sense. So you might have experienced gravity in a certain way a thousand times. And we think oh the student's just going to get this. They need to go through the experience. They might need to experience this a thousand times, they might need to write a thousand times before they're ready to have some sort of experience where this complicated thing is doing some kind of, you know, strange analysis. So I would say one just have a lot of, you know, these experiences where you know, I don't think it, you don't need to be having experiences with AI all the time. I think you. A lot of the experiences that are about inquiry and making sense of the world are things that are going to build a scaffold for yourself to approach anything, whether it be ChatGPT or some physical phenomenon that's equally complicated. So I think really giving them that foundation is a really.

Because it's essentially the paradox of education isn't it that the better a teacher understands a topic the harder it is for them to picture the experience of someone who doesn't to be able to remember what it's really like to be coming to these things blind. So I, I definitely see what you're saying in that respect. I did want to ask you and I understand what you're saying about us needing to embrace technology as it actually is, that you can't unring a bell that the world is as it is and we're not going to be able to change it in our classroom directly so we have to teach to it. But I did want to ask you, is there any technology that if you had the ability you would like to uninvent in terms of how it has changed education?

That's interesting question. I mean I don't usually. I think my perspective is always that. It's more of a inquiry stance of just trying to understand how did we get here.

Right, yeah.

Rather than. Because ultimately, unfortunately, we can't undo like we can't, we, we can't undo the dropping of a nuclear bomb and we.

Can'T even really know how the world would, would be different if we did.

I mean, some of those TV shows that try to are, are interesting where they take an alternative view of reality. But I think ultimately, I always find the most inquiry, the most interesting stance is just to try to make sense of the world as it is.

That's a hard enough task.

The world is so rich. I think I, I guess I bring a certain amount of hope to the situation. I feel like there's a lot of doom out there and there's people, especially around this AI topic of people saying, you know, that the AI is going to take over and blah, blah. Yeah, I just have this perspective of hope that I think we're going to, I'm hopeful that we can make more sense of the world and that we can find a way to make things better.

So, speaking of using technology for sense making, I wanted to get your perspective on the place of students starting to write code. Right. So one, one really exciting new thing about having generative models like ChatGPT is that it sort of lowers the barrier to entry because you can communicate more easily in, you know, just saying what you want and then, okay, maybe it doesn't get it 100% right every time based off of some specifics of what you asked for, but you, you have a concrete thing that you can have an experience with in terms of seeing does this solve my problem or not. So I wanted to get your perspective on generally just integrating technology in terms of making sense of the world. But then also this, this more specific possibility, which is now open to us of, you know, having a student be able to articulate themselves not in the language of computers and then get something in the language of computers that then they can use.

Yeah, no, and I, this is an area of particular interest to me. I think I've been, I've spent a lot of time teaching computer science and maybe some of these more technical things and you know, some of this technology is really interesting and it was particularly interesting at the high school level. I was teaching Python and we were using Flask to create websites and kids wanted to do projects on things on too wide a range of things for me to have the expertise. Sure, right. And you know, in a way it's been very incremental like there's been an opportunity to copy code well before any of this generative AI. Like, of course, yeah, there was code online. The thing that the chat GPT added and I guess other, like there's other similar products out there now. Is, was really good skimming the documentation for you. So I would teach students how to read documentation, but it could, it could pinpoint the documentation you need much more efficiently. So you weren't spending hours looking through the documentation to find the right thing. But ultimately what I found, and this is true in elementary, in middle school, in high school, the fundamentals of it aren't really about this generative AI technology. It's about the depth of understanding of what's going on. So I feel like the lesson that I always end up coming back to is tracing, where you try to understand what's happening at every point in a program or a process and you're willing to go as deep or as abstract as you need to go to UN to to get a better sense of the problem. So if something's not working, you really have to step through and understand. And I think that's where students who use, you know, some kind of assistive thing to create code or they've copied code or they just make something up and get it to work on some. At some point you're not, you need to have that skill to dig in.

Right.

Understand is this working in the way I want it to. What, what actually going on? And that I feel like that is something I experienced. I studied physics as an undergraduate. That was a big thing that was exciting to me about physics. You could make sense of the world. And I think that's, that's what got me interested in computing. I was like, oh, there's another space where that kind of thinking is so important and so fundamental to success. And ultimately, like people who know who can do that well can get a lot more out of these generative AI products because they know how to use it. Where if you're, if you don't know how to get to the right level of abstraction or dig down deep into things, you're going to get stuck and you're not going to be able to get yourself unstuck.

This is exactly the message that I have been telling teachers that I've, I've spoken to. And I think also other faculty have said, you know, our students are now in a position where they can write things that if they don't have an understanding of how this relates to the real world, then how are they going to figure out exactly as you were saying, you know, is there some subtle error in what's been created here? Right. And also the questions of, you know, what are even the interesting problems to create, something to solve that also comes from an understanding of what's really out there and this sort of way of thinking about how do I frame a problem in ways that I can actually tackle it. So from there, I was curious to hear more about what is the. The overall advice that you would give for what are the key points that someone should understand if they're wanting to teach the skill of computational thinking?

Right.

Because the thing that we're alluding to about having this sort of real sense of what's out there that comes from experiences, that comes from solving problems which are relevant to you both as an individual and as part of your community. What are some of the big tips for people who are maybe less familiar with some of the literature about this, about what's really key when trying to help students develop this computational thinking skill?

Great question.

That's a great question. I think one thing that I would emphasize is creating that rich context. So maybe it's the right tools, maybe it's the right way to get kids thinking about what's possible. I mean, I think that's one of the things we're thinking about at school Wide labs is really like, how can we create this environment? Like, can this classroom display really get kids thinking about the right things and get their kind of creative juices going, but also elicit the kinds of questions that are going to move them through an interesting inquiry? So I think setting that up, and I think setting that up, it partially involves finding the right materials, but it's also like having knowing how to work with those materials and knowing how to kind of be responsive, and that kind of leads in to the other. Another piece that I think is key is really just, I think one of the most important things in all classrooms. And I don't. This isn't unique to computational thinking, but it's just as important in computational thinking is really understanding our students and understanding their questions and where they're going, because it's that deep understanding that allows you to kind of find the right pathways for them and connect with them and help them do this kind of constructing of problems and guiding their learning. So I think another piece is like really understanding students and connecting to them. And I think the third thing I would say, which I found really important, and I don't know, I think there is some value in having knowledge. I think it can, as you Said, no, it can be dangerous sometimes if you know too much, you, you have trouble explaining things. But I think more important than your knowledge is your curiosity.

Ah, yes.

And I think students really pick up on whether you're actually curious about, like, you can't take curiosity. And that curiosity is what should drive the work. And I think that's why I think this, like, learning alongside the children can be actually beneficial because it takes this curiosity stance to the work. And I think that's just so important that we're here to inspire other people to get excited about things. I think, I don't know, I read a lot of stuff in educated. It's like people are looking for shortcuts.

Right.

Ways to like, oh, we're going to solve this problem by making it easier. Sometimes learning is hard work and, and it's, it's. I feel like the people who are the best learners, they lean into the challenge. They, they see the most, the steepest gradient, and that's the way they go.

Absolutely. One of the best pieces of advice I ever got was if your class is always easy, you're in the wrong class.

And to dig into this point a little bit more deeply because really key to that actually working, as you mentioned before, is that students, and if you're talking about teachers in the professional development context, feel safe to make mistakes. Right. Because if, if, if you're not comfortable making mistakes, you're going to avoid that challenge. So can you say anything about specifically what you would recommend in terms of, be it generally and also in terms of bringing technology into the classroom? What are some practices that people should keep in mind when they're creating this environment where it's okay to make mistakes with the technology?

Yeah, no, I think that's a great question. And one thing I'll say is just, I mean, there's a lot you can do around just really modeling that kind of openness to making mistakes, maybe pointing out the mistakes you make as a teacher and really making it clear that there's that safety. But I think the other thing that needs to be lifted up around the safety piece is really addressing equity.

Sure.

I think it's really important that we allow all people in the room to feel safe and we really understand the different perspectives. And it's complicated because I think science and technology has not always been and continues to remain a challenge to be safe for all people. Sure.

Sometimes deliberately so.

There's responsibility from the scientific community to push against that. But I think specifically in the classroom, you really need to, like, put effort towards Thinking about that and thinking about the ways, you know, I've been reading a lot lately about kind of language context and the way, you know, there's a lot of technical language in science and technology. Yeah. That language doesn't always work to make people feel included. And what. What is our philosophy on language is. Is the different language dialects and ways of speaking that our students bring to the classroom, is that. Is that a liability or is that an asset? And I think those are different ways of looking at the world that I need to unpack a little bit. And I think that relates back to this idea of coming together. I think we need to change some of our structures around how we get people to come together. And I know, like, in my experience, you know, studying physics as an undergraduate, there was this big kind of culture around weeding people out and making it, you know, it's hard. And there was something fun about working hard and trying to get through something. I don't think it was as inclusive as it could be. And some people were excluded not because they weren't working hard enough, but maybe because of some aspect of their identity.

And I think because the language being used to communicate with them was one that they hadn't encountered before.

For example, I can say anecdotally from my own experience, I mean, it was absolutely no secret that the purpose of large problem sets during the first year is basically that thing that you were saying of, you know, this is sort of a barrier. And, you know, in. If you're dealing in a physics to see, you know, you put up a potential barrier, does the particle get through? Okay, that's a fine experiment in a physics lab, but it's not actually what you want to do with people. Right. We want to remove those barriers to entry into.

It's one of the things that I've liked about this revolution in large language models, natural language processing, whether in our technology or not, that there is finally an idea of being able to say, hey, we are going to be able to be interacting with these powerful tools in more than one way. There's more than one correct way to get a tool to do a thing. I do love that and spinning off of that and connecting to something you said about instructors needing natural curiosity. We had a question from the audience which looked interesting to me, about how did your colleagues react to using new technology like large language models like ChatGPT in the classroom for learning? We touched on this a bit in the conversation about how technology that instructors are not familiar with, they might be hesitant to use. But I Wanted to ask, in your personal experience with these new technologies that are touted as being especially easy for new users to pick up, what's been your experience in terms of people actually using them as teaching tools?

So you're talking specifically about the large language models.

The question mentions them, but it starts with the question of what was your colleagues reaction to using new technology as in the classroom for learning, LLMs and ChatGPT being an example.

Yeah, I mean, I think the ChatGPT brought a lot of feelings with it that are different. But I think. Oh yeah, I think that tools are really powerful. And I think, you know what? One example of a really powerful tool is just written language. And we don't think about it because it's almost like the oxygen we breathe. It's all around us. But it's a really powerful tool that before humans had written language, we weren't as capable of thinking of doing things. Of all kinds of things.

Yeah, anything anyone could learn had to be something you could learn within one lifespan. And that's it.

And this is some ideas that I borrowed from Andrea Decessa on. I'll have to think of the name of the book that I'm. But he did some writing on this idea of like, you know, written language. It's this really powerful tool. It's a cognitive tool, but it has a huge amount of infrastructure. Like we're trained to learn how to read. There's easy implements to write and to read. There's a whole culture around these things. And I think sometimes when we're introducing new tools, we don't realize how much we have to build that culture to build that framework. I think people don't utilize spreadsheets as much as they could in schools because there's not enough. So there isn't the same social framework around the use of spreadsheets as there is around reading around the use of written text to expand.

Excellent point about the culture of Twilio speak. Important.

What would be an example of a social framework around spreadsheets? Like if it existed, what would it look like?

Well, for one, I think everyone would be. Would be. Well, one, everyone would be using them. Sure do. I mean, I guess it is a little bit of a thing where it kind of feeds on itself. Like as the teachers are more comfortable with these spreadsheets, they're, you know, in. If they're in class and something comes up, you think, oh, let's put this on a spreadsheet. We, we just collected some information. Like that's not, I don't Think that's as big a part of the culture of elementary school.

The same way that you would say, oh, that's cool, let's write that on the blackboard.

Yes. And I think if we started using these tools, like whether they're spreadsheets, whether they're, you know, mathematical tools like Mathematica, like, I think the more they become part of the lex, part of our experience, then they're accessible to us and we start to, to make sense of them and they just, they grow in their ability to accomplish things because they're part of our culture.

I have another question related to cultures and this is, I think, part of the reason why people have such strong feelings about ChatGPT.

And John, excuse me for interrupting you for a moment. Obviously I want to hear this question, but Greg, we are at the end of the scheduled hour, but we'd love to keep talking. Here's some more questions from the audience. Do you have some flexible time here?

I have a few minutes, I don't think. I have an appointment till 2:15. Okay, great.

Let me, let me then just ask you this question. So the, yeah, so the, the culture around assessment is something which is really key to a lot of students educational experience. Right. Different norms around how assessment is supposed to work and where the consequences of having assessments are these really deep cultural features of classrooms. So the question is, how would you view positive changes both in terms of what we should be aiming for and now in terms of what's possible now that we have, you know, generative models and computer systems that can, you know, very easily solve various assessment items? What do you think is most needing to be changed? And what's most possible to change around the culture of assessment that would lead to, you know, a more positive experience for students learning?

Yeah, that's a great question. And I think assessment is a particularly loaded topic. There's a lot of thinking and feelings about assessment. I think one framework that I would bring to this is just really thinking about formative versus summative assessment.

Absolutely.

What is the purpose? Like, why are we bringing in the assessment? And I think sometimes there's not enough thinking about the reasoning behind the different ways assessment is happening ultimately. And maybe this is an opportunity in AI. And I've noted, like in a classroom there is an abundance of data and I think there are ways to utilize things like language models to maybe make assessment more meaningful. But ultimately we have to spend the time to think about what are the different things we're trying to accomplish with the assessment and how much, how much Are we, are we like bringing the students into the assessment process? Because ultimately I think what we really want is we want. And this kind of connects to what I was talking about before with the micro bit. We want the students motivated to improve themselves. And I think sometimes outside assessment can run counter to cultivating that mindset because you're, you're being evaluated from the outside. But ultimately we all have to evaluate ourselves on some level. Now I don't think, I think there's value in lots of different things. I don't think we should, you know, just tell, you know, work on whatever you're working. We'll check in with you at the end of the year. Like that's not a good way to.

Sure.

Classroom.

Sure.

Ultimately, kind of maybe connected to the scaffold and the trellis piece is we want students at the end of the year feeling like they're the ones leading the charge and they're the ones who need the assessment data so that they can get better. Right. That doesn't feel like that's.

Yes.

Yes. And can you real briefly, for audience members who might not be familiar with the language of summative and formative assessment, can you brief. Elaborate on how do you, how do you go from formative assessments to these summative assessments where the students are saying, yes, I need this summative assessment because it's important to my success.

So for the way I think of it is that formative assessment, often for the context of the teacher, can play an important role in figuring out how you're going to make teaching decisions from moment to moment or from class to class.

Because that's what formative assessment is. It's assessment of the student as they're progressing usually without a really high stakes sort of like impact on, on their final grade or whatever.

Yeah. And also it can, it can be kind of connected to this idea of students like, understand like getting information about how they're doing in the moment. Where, where summative assessment is to kind of is more about this piece of evaluating where a student is at a given point in time.

Right.

What's the total sum of their learning.

Resulted in A given instrument can, can feed into both purposes. But I think it's really the reason I bring up that frame is just to. I think we should always be thinking about the why the things that we're doing. If we're collecting, like why do we need to know where they are? How is that going to help help them in this moment? And there are, there are useful things like maybe they need to do something a little differently. Maybe I need to do something differently as a teacher to help them in the next phase. There's all kinds of actors and decision making, but I think sometimes we're not, we're like implementing assessment, but we're not talking a lot about the why and how to fine tune it and make it appropriate for everyone in the room. I mean, I think that's the other thing that maybe some of this technology can do is make it so that we're not using these kind of broad, very kind of fixed and static devices that aren't really meeting all the needs in the moment. And maybe some of this computational technology can make some of these things even more effectual, but only if we really know what we want to do, what's our purpose here? What are we trying to accomplish? And are we bringing the students into this decision making process as well? I think they have a role to play, of course, like, you know, there. We don't want anyone to be the only role.

In terms of educators and students being able to articulate, you know, what they think. It's important as the assessment items. You know, you work with a lot of educators and students, and from my experience as an educator, a lot of the decisions about assessment are sort of made from larger than the classroom context.

Oh my.

So if you could summarize some of the things that you have heard both students and teachers say, these are the meaningful assessments to us. Is there a brief short list of things of examples that are of the meaningful type? You've mentioned several as we've been talking. But are there any other ones that come to mind?

I mean, I guess there's a lot of different things, like examples I could give, but I feel like the things that resonate the strongest with me are just experiences from my own classroom. When I've seen a student struggling with a problem and they figure something out and they're able to accomplish something that they weren't able to accomplish before. It could be, you know, a high school programming class. It could be like, I'm thinking of an example in, in a fifth grade science class, I once created sort of this electrical circuit thing inside of a container and they had to like, figure out what was inside of it, running electricity or various things. And there's this moment when a student knows that they accomplished something that they, they were struggling with and they accomplished something they could, couldn't. Like it took the process to accomplish that. Like they knew it came and, and when, and they have that ownership of it that's I think some of the most meaningful assessment and the most meaningful experiences that I've seen on students, and you can see it on their face. It's not. It's not something I gave them. It's not. It's almost. It's. They. They own it, and they.

Absolutely.

And I just. I just haven't seen a lot of those experiences and standardized tests, and I'm against all standardized tests, but I think it's those moments that drive lifelong learners. We need to be making that available. And I think there's a role for benchmark and, like, figuring out, like, what people need. I don't think we. It's like, there's one answ to this that you never have that experience on a standardized test. And those are the things that you remember and carry with you through your whole life.

Excellent point.

I love this. There's a lot of attention, rightly so, on the eureka sort of experience. The I have found it sort of experience. But I've heard a lot less about this sort of thing of I did it of this was hard to do. I couldn't have done this before. And now because of these things I've learned, I. I did it. And I do love that sort of experience.

I wanted to ask one other thing.

Yeah.

Feel like. Like I've been reading books of different. Like, you know, I gave the example of Seymour Paper. Like, a lot of these authors, they'll talk about some of their own formative experiences. And I've noticed a number of them are experiences that happened outside of school. And I think one of our goals should be to try to bring more of these things into school.

Oh, yeah.

So that people are having these great experiences. And, you know, these are just a few examples. But I think when we see great examples, we should be bringing them in and allow. Because the more that's happening in there, the more we can kind of grow on those experiences.

So since we're. During the time you said that you might need to have another commitment, I wanted to ask, you know, are there any other questions, big takeaways that you'd like for the audience to have? I just want to make sure that you get a chance to elaborate on any takeaways, since we want to be respectful of your time as well.

Yeah, no, that makes sense. I think I've just really enjoyed our conversation.

So have we.

Likewise talking about these ideas. I'm always excited to talk about them, but I feel like you guys have created a really rich space. Maybe. Maybe this is some sort of, like, maybe you've done some of what I was talking about of creating the right environment so that right. The right thinking and the right ideas can come across. Because I've, I've really enjoyed just thinking about all these things.

Likewise. And you know, we'd love to chat either. Yeah, we'd love to chat either online or offline about, you know, you know, rubber meets road. Right. Where you know, bringing some of these, these experiences for students would be amazing.

In, in the little time that we had left. I did want to ask, you mentioned questioning why we wanted to use certain assessments, what they're for. And I remember in conversations we've had outside of this particular interview, you mentioned that a lot of things in education need questioning. And I wanted to ask you do any particularly spring to mind that you'd especially like to question? Not necessarily to say that people are doing these things wrong, but just that not enough people are questioning it and that they should and that that could be valuable?

I mean, I don't think I necessarily have, like, I think I'm not the only person kind of pushing some of these points.

Sure.

But I do think this, this agency piece and another thing that has come out of our research and a number of, and we've gotten from other researchers as well, is this idea of coherence. And one of the distinctions that our group makes is often school experiences are. Can be constructed coherently for the teacher. So you set, you know, this is the next kind of learning progression for this concept.

You know where you're going, you know how it all connects.

But ultimately the students need to have that coherence. And maybe the students need to be driving more of. There needs to be more questions. Well, what do we need to focus on next? What question on our question board should we tackle first? Those are the kinds of things that I think can really be transformative because ultimately it's not about disrupting. It's not about this disruptive framework. We're going to find the new answer that's going to make it better. The answer is in the students. Right. And we just need to understand and give them some power, but also give them the right context so that they don't have the wrong power. And really almost, it's almost like we're, we're, we're finding what's. It's already there. We just need to, we need to have the curiosity.

Yes.

Bring it out. And so that's, that's my wonderful perspective.

I'm seeing, I'm seeing someone in the, the chat say communication is key and It's. It's so true. Just, it sounds like the Montessori method writ large, that what we're talking about is not just questioning what's in the curriculum or how we're teaching the curriculum. We're questioning the concept of curriculum. We are questioning the concept of the instructors, the administrators, the government deciding what is taught, when and in what order and in what structure, and saying maybe the kids have something to say about that. I love that idea.

And, and something that I've seen you've written about as schools go for these more student focused directing of the curriculum, is that to make this successful, it actually requires more logistics on the part of the people who are helping the students and supporting them. Right. And so I'm personally very thankful that there are people like you who are putting in a lot of work to help develop these logistical supports that can allow such things to succeed.

Yeah. And I think, like, as much as I want, I want to emphasize the students, but I think the teachers are just as important and we have to, like, I think they are. They are the ones who know the most about the students. And we need to understand the teachers as well and their perspective. And that's something in a lot of education reform, that there's not enough emphasis on what the teachers bring to the table. And so that's something I want to emphasize as well.

Well, Greg, thank you so much for joining us today. Thank you to everyone else who has been watching us and commenting and asking questions or just listening. This has been incredibly gratifying for me personally and look forward to being able to communicate with both you, Greg, and with all of our viewers more in the future.

Thank you so much.

Yeah, Greg, let's definitely chat again.

Wonderful. Thank you.