00:00.00 Paul Hi Welcome back to Arutek Episode 82 J D Kvelli you've been telling us some great stuff about this program that you were working on at risk called algorithm and you you hinted at it for a number of different reasons. But I've got to ask you upright. Why teach data science to kids. 00:20.52 JD Calvelli Yeah, um, very important question. Ah that we get very often unsurprisingly but you know it comes on to a couple of things for us at risk number None is that. You know the the concept of of data data privacy data security data presence you know data as an object for exploration and for analysis is super culturally salient. Um and not only is it culturally salient but it's just like fundamentally important. Um, when it comes to like navigating the world in our opinion. Um, you know there's there's a certain level of like world literacy um and information literacy that you can get ah when you can analyze a data set or even more broadly when you can look at a graph correctly. 01:51.50 Paul M. 02:03.70 JD Calvelli Because Data Science sort of expands far beyond the idea of like looking at an excel spreadsheet but it's just the whole concepts behind like looking at representations of data creating representations of data having those things say something having them represent something you know, being able to parse out exactly what they're trying to say and exactly what they're representing. 02:33.80 Paul M. 02:43.36 JD Calvelli You know this is just like not only is it culturally relevant. But it's just a direction. Um, that sort of all a lot of inquiry seems to be going now. Um, and you know, ah in our opinion. It's It's better to you know, have those situation have that conversation start as soon as possible as opposed to sort of. You know, push it off down the pipe. Um until you might take a class on it. You know in a political science. You know sociology course or or you know in College. Ah you know the thing is like it's It's an important enough element of our daily lives now that we don't feel Like. We can afford to push that off. Um on students you know Anymore. It's something that all students should have a certain Baseline knowledge of prior to entering the workforce prior to entering the world because it's just so fundamental to being able to you know, ah, learn and grow and understand and and take in information. 04:18.10 Paul Um. 04:31.42 JD Calvelli Ah, you know as as time as as we've gone forward and in time like. 02:29.12 archpodnet M. 05:00.10 Paul Yeah, no makes sense and certainly as different kinds of data are more accessible to more people through more different venues being able to assess them and interpret them is ah is is Useful. So a follow up to that question then is. Is there a sweet spot in terms of age for kids to start learning about Data. You're targeting late elementary early middle school. Um, which incidentally is what I've worked with when I was in Ed Tech Primarily. Um. 05:30.80 JD Calvelli Super cool. 06:01.12 Paul Is have you gotten into the data about when it's best to teach them about data. 05:48.30 JD Calvelli Yeah, so we've we haven't done anything specific on that front but kind of from our own anecdotal experience like data science as an idea or data science as a discipline you know, like most other disciplines build off of themselves. Right? So there. There are certain elements that you can learn earlier on that can help sort of inform and set you up to be able to learn. Ah you know, learn other things later on so one of the things you know when we were when we were talking to Steve about this project initially he was like let's make a game to try to teach. None through none graders. How to do a regression analysis and I was like Steve ah I don't know how to do a regression analysis I was like let me go learn how to do a regression analysis and then I'll tell you if I think we can teach ah you know fourth through sixth graders how to learn to do a regression analysis. Um. And you know that's that's sort of ah our facetious way of saying or you know jocular way of saying um that there there are levels to this right there there are certain levels of sort of of expectations of of what you should know at different stages. Um, and we hope that you know even at even in that age range right? we're. At the point now of combating um or rather just filling any gaps in terms of like lack of information of what data data is and and what it can be used for and so we hope sort of in the younger age groups that we can just start those fundamental conversations of like here's data. You know what is it. What can it be used for write and then maybe some rudimentary visualization stuff where it's like you know a graph is really more than just numbers on a line right? It's it's demonstrating something that can be used to demonstrate something and then you know as you go on in your career then you can start entering more of the statistics adjacent stuff. 07:02.62 archpodnet Ah. 09:26.88 JD Calvelli You know, maybe in high school definitely in college um of like you know, ah, how can we prove that certain things are true or how can we get us close to proof about certain things are true because the None thing you'll always hear is correlation is not equal causation. Um I might get that tattooed at this point I've heard it so many times. Um, but the point being like you know you you sort of graduate from an earlier space of like oh what is you know data and why is it important and then how can I use it to start making decisions and then it even pipes into you know, computer science and machine learning a lot of that is fundamentally data science. Um. And so if you ultimately get to that point you know and you're still this interested in data and and how we can use it to make better predictions about things and you can start learning about you know k-means clustering and neural nets and so on and so forth. So you know it's it's just again to the earlier point of like it's this it's this space sort of exists across many different disciplines that we feel is yet sort of underexplored um in the k through None ah space and you know again, the sort of like joking way that we described is like how many times have you ever used. You know a t I 84 calculator after you know, ah high school precalculus or houseke calculus but how many times in life. Do you need to be able to analyze a set of information and be able to make you know educated conclusions about it and so that hopefully you know can build up further and further and continue to grow. 09:50.40 archpodnet Red. 12:34.22 JD Calvelli Um, as you know with a student as their interest in the area grows if they want to take it all the way to machine learning. Great. Um, if they want to know how to you know Make Macros in an excel excel spreadsheet great if they'd like to learn how to visualize and in R Ah you know make data visualizations and R. That's great. Um, but it can grow and it can change and there's so many avenues that you can go towards and it's just a question of like getting that exposure early on because it's the exposure that seems to be missing right now. 11:16.40 archpodnet Yeah, it totally seems like looking at especially say kids that are in None to None grade right now you know just because of the advancement of technology most of your manufacturing and and other type jobs that don't really require much beyond a high school education. 14:01.26 JD Calvelli Um, the. 11:52.70 archpodnet Are slowly going away right? because we're we're We're either automating those making them more efficient or actually developing robots to do those jobs because it's cheaper. So as that happens people are going to have to be better critical thinkers and actually use this kind of thing because like you just said I mean I'm 47 and I Heard that growing up like oh you'll never use this math outside of you know, high school or something like that and that's like kind of sad. Why shouldn't I use this math outside of high school. It's because most of the people I grew up around are in construction or you know doing something where they literally don't use that kind of thing on a daily basis right? But I think some of the jobs that may be out there in the future. 14:46.60 JD Calvelli Yeah. 13:07.00 archpodnet You will actually need to think this way even though a computer might be doing it for you. It's helpful to understand the reasoning and the why in order to actually tell the computer and and what to do in ah in the right way and then understand the results that you're getting back so along those lines. No go ahead? Yeah yeah. 15:38.40 JD Calvelli Yeah, like None of sorry not to cut you off. Um you know one one of the one of the ways that like I to that point that that I kind of like conceptualize this is like at None point in time there was a brick layer. Right? Like the job was to go lay bricks somewhere my greatgra my you know my ah great grandfatherfather or rather my grandfather sorry um, ah, immigrated to the states from italy um, and he kind of worked in that in that space he did you know construction. Um, and there was a time when you know he was a big buff. 14:03.22 archpodnet Um, yeah. 16:46.40 JD Calvelli Man kind of carrying bricks around right? you know and that that was a construction job. Um and sort of like the reality is that like that kind of job has kind of disappeared because we created you know, bricklaying technology right? We created brick layying machines. Um. 15:01.46 archpodnet Move. 17:18.90 JD Calvelli But in in the absence of the bricklayer job right? There has created a necessity for the like bricklaying machine manager in sort of like this weird you know the death of 1 created the birth of another and like what is the skill set that you need to manage. 15:22.86 archpodnet Right. 15:33.80 archpodnet Sure. 17:53.12 JD Calvelli A set of machines right? like data science or like to a certain extent right? You need to be aware of like okay, what is the you know what's being you know I have 50 machines right? All of them are creating you know or laying this many bricks in this amount of time. Let me put that all in the spreadsheet right? I want to try to maximize the amount of time. Well maybe. 15:48.90 archpodnet Um, yeah. 18:32.10 JD Calvelli You know, let me ab test this maybe if I have 10 machines running you know in 1 place then overall as opposed to 5 and 5 then overall my amount of time you know to lay the entire bricks for the house will increase you know what I mean like I'm I'm sping here right? but but the point the point more specifically is like you know. 16:48.94 archpodnet Yeah, yeah. 19:10.16 JD Calvelli To be literate in that space. You need a certain exposure to like what is data and why is it important you know and I mean like not not to beat the dead horse on it. But it's like yeah, it's It's interesting because I think a lot of people have sort of recognized. Um this problem of like oh well, there's a lot of jobs that seem to be disappearing. Um. 17:11.52 archpodnet Um. 19:41.20 Paul Um. 19:49.72 JD Calvelli And we seem to not be preparing people in high school and beyond to like do the new things that are coming ah but the missing piece is like oh well, one of the things that we seem to need to understand or need to teach students to understand is this element of like. Managing data and understanding data and being able to make decisions about it. Um, so you know back to back to the initial point of like that's why we think it's important um to be teaching all students at every age you know? Ah what these what this is and what this what this means and how it can be utilized because you know we. 18:21.48 archpodnet Um. 21:01.60 JD Calvelli We believe that it's important in in in every function. Um, we have the benefit of going out and interviewing um sal khan of Khan Academy um where we because khan lab school sort of piloted an initial program for data science um in high school in a high school setting. 21:37.88 Paul M. 19:10.80 archpodnet On ice. 21:40.18 JD Calvelli And you know one of one of the things that he said that I think is really prescient. Um, and you know I'm not quoting him directly by any means but 1 of the things that he he said that I thought was again very prescient was like you know fifty years ago who had big data. You know, maybe major companies had big data right. 20:04.68 archpodnet No. 22:17.92 JD Calvelli Ah, you know insurance companies maybe had big data like a lot of data from which you could derive decisions. The reality is now like everyone has access to that and not only that but like there's a certain expectation of the ability to utilize that to make better decisions about whatever it is that you're doing right. 20:28.80 archpodnet Brett. 22:53.62 JD Calvelli Whether it's just searching Google right? or finding a dataset on kaggle or pulling from. Ah you know an Api an openly hosted or privately hosted Api like the access is there in a way that it's never been before and the reality is when the access is there. There's a certain expectation of ability to utilize it. 21:19.12 archpodnet Um, yeah. 23:30.80 JD Calvelli Right. 23:53.94 Paul So yes, you're you're making a strong case for why people have to be not just numerate, but but comfortable with analyzing and and applying data to their to their lives in various ways. Ah. When it comes to teaching kids I'm getting kind of ah a vibe of the same reason why you would teach them a foreign language when they're fairly young so that it becomes a natural part of how they think and how they can interact with the world. Um, so you you settled on gamification and there are many many different ways to teach people. Gamification being just one of them. So Could you just explain why you went that particular route and maybe touch on what you think the strengths and maybe even some of the limitations or weaknesses of gamification are as an educational tool. 25:14.98 JD Calvelli Yeah, for sure. Um, we yeah we we did. We did decide to approach the gamification angle I think so sort of from like ah like a metal level perspective right from from like a topdown perspective. You know we believe? um or rather. Yeah, let me back up from a top-down perspective. Um, what's interesting about gamification right? Is it kind of challenges the idea of what work is right? We have the sort of like cultural conception that like work needs to be not fun. Like the absence of fun equals work. Um, and you know we yeah all work in no play exactly or like work hard so that you can play hard. Um and and you know obviously. 26:54.18 Paul Ah, right, all work and no play. 27:09.36 Paul Me. 27:00.22 JD Calvelli There is importance to to recognizing work as as a discipline that requires effort and you know and and focus and attention and so on and so forth but to like completely divorce it from the concept of play right? or from the concept of of learning through play kind of doesn't sit right with me. Right? And I think it also doesn't sit right with a lot of of of individuals at risk at well as well. Um, you know and while we're not like you know, um, developmental psychologists again generalists No no knowledge of anything socrates all that I know is that I know nothing but you know ah that while we're not developmental psychologists by any means right? like. When you're a kid I remember or my parents told me that when I was a kid right? There's this huge emphasis on like learning through play like learn through exploration right? Like how your kid put this you know Square peg in the round hole and be like oh that doesn't work like why doesn't that work and then try something different but then after you know you're no longer a kid. 28:52.42 Paul Moving. 28:56.22 JD Calvelli Like a kid kid that seems to all disappeared ah with you know, like exactly like puppies play fighting right? like there's a space where it seems to be good to allow for a certain level of play and actually through that play you learn something. But after you're like a baby you know that that sort of understanding of play. And and it's and it's respect or and how it works in in respects with work or or rather how it can sort of be utilized to help facilitate work like disappears or seems to disappear and so again from like a higher level Perspective. We feel like gamification is really interesting because it kind of. 27:43.16 archpodnet On. 30:09.20 JD Calvelli Merges that space again where it's like oh play is good and we can learn from it and work doesn't necessarily need to be no fun right? Um, and in fact, like maybe through some fun you can actually learn more than you would have been able to Learn. Um, if you didn't have any fun while you were doing it right. Again, sort of like Anecdotally I remember ah sort of what put me back on on this path like a very very long time ago now not that long ago. But you know, but what put me back on this path of of ah of ah gamification and being interested in this space was a game that I played um in None grade that was a simulation of the black death. 31:41.18 Paul Oh. 31:23.60 JD Calvelli Um, and how it affected Medieval Europe Um, and it was a game with dice and rolling and you had to make rules and you had to defend your town and you know people could die and you know because of the plaque death if you didn't quarantine correctly and so on and so forth and you know I'll go out on the limb and say I don't remember. 29:16.84 archpodnet Yeah, yeah. 31:57.94 JD Calvelli That much about None grade right? at all. But I do remember that right and I I to this day remember that and remember that experience as being super meaningful to me. Um, so you know again, this is all sort of like higher level but we feel like. And I personally feel like games has the games have the potential to really meaningfully get people to engage with something right. 33:07.10 Paul Um, no I you're talking about None grade and and this is again another one of my asides because I got old brain. Um, when I None started working at Dalton was because of this game that was produced at the school. 33:01.68 JD Calvelli Hip. 33:41.68 Paul Based off of an archeological excavation and it was a collaborative group project and we've continued it on actually I've been helping them redo it for the None grade now. So we're talking about the same roughly the same grade range that you're looking at with ah with algorithm. Um, and. 33:28.14 JD Calvelli That's super cool. 34:17.98 Paul Using a kind of a game metaphor some of the game incentives in order to to teach kids? Um, but I I also have despite having been doing this for a long long time and understanding the value of it I also kind of have a complicated relationship with gaming. And gamification as an educational tool ah that we don't have to go in now because it's it's totally an aside but you know I think about different kinds of gamification and for me for example, Duolingo ah, which is a very famous way of gamifying language learning. 32:56.88 archpodnet Yoha. 35:32.98 Paul Ah, is actually really effective for me to brush up on languages I I get very angry when I do it I get very frustrated that keeps on giving me all these nudges and trying to keep me in its environment for you know, but it's it's extremely helpful before any time before I travel overseas. Ah. 35:22.48 JD Calvelli Um. 36:09.48 Paul But that's also even though they have these layers of um of social interaction where you can like compete against other people or you know post your progress as you go I don't want to do any of that I just want to get it to quit bugging me. 34:10.80 archpodnet Yeah, yeah. 36:42.14 Paul Ah, but but that's ah, a very solo activity and that stands in distinction from that archeotype stuff that excavation which was an organized group activity with a different sort of incentive structure and a not an in the game and so incentive structure. So yeah. I Don't even know that I've got a question here other than than a comment is that there are a lot of different kinds of incentive structures that can be built around gamification and maybe I guess if I do have a question. It's um, it's how did you decide on the particular game of and the particular incentive structure that you built into algorithm where. You know that allows the player to advance. 37:35.64 JD Calvelli Yeah, no that I mean I was actually going to follow up. Um your point by saying are common rather by saying that. Yeah I mean Duolingo is a very classic example of like gamification done right? Um, or what a lot of people perceive to be gamification done right?? um. But in my opinion Duolingo is like gamification done the way that people expect um which is not necessarily a bad thing right? but it it is sort of like the expectation of like you know how do you make something into a game. Oh well, you make it a competition. 38:48.42 Paul M. 36:19.86 archpodnet M. 38:51.14 JD Calvelli Ah, you give points. Ah you know and that's a game right? Um, and I think you know for for in a lot of cases. Yeah, like that that can be a game right? Um, and you know, no no knock on Duolingo at all I think it you know works very well definitely.. It seems um for foreign language Acquisition. Um. But I think and you know back back to something I mentioned much earlier like gamification and how to make a game so that it teaches people things right is still very much in its infancy right of like how do you do that right? because when you ask people when you ask a lot of people. Um at least Anecdotally in my experience right? when you ask people like oh. What is gamification right? or like what do we want gamification or how how would you expect gamification to work. They're like oh put points on it right? and and make it a competition and that's how you gamify something? Um, and so we with algorithm really wanted to try to challenge that. 38:22.40 archpodnet Yeah, yeah. 40:42.58 JD Calvelli A little bit and so our idea then was to develop an incentive structure where the only way that you could meaningfully engage with the mechanic of the game is by understanding the information. The subject matter. Um, that we expect you to understand that we hope that you can learn by virtue of playing the game. So. It's a little bit backwards but the but the way that I would explain the the simplest way I think I I can explain that is like in the last stage of the game. You know the mechanic of the game is dragging and dropping songs. 39:16.62 archpodnet Um. 41:50.22 JD Calvelli On to requests so that you can fulfill them to get a high score but sort of the the conflict is that you have a time limit. Ah you you don't have all the time in the world to be able to make this decision and so when you have a list of None songs to go through and only 30 seconds to make a choice about what fulfills this role right? or what what fulfills this request you have to use the data in order to progress in the game right in order to get that point you have to understand that oh I need data. 40:34.80 archpodnet Yeah, yeah. 42:57.84 JD Calvelli Ah, because with using data I can make this decision faster and more accurately than I could have if I didn't use the data and so it's it's not quite putting points on things and making the competition. It's a little different. It's saying. 41:01.12 archpodnet Um. 43:29.60 JD Calvelli No in order to advance. You need to fundamentally understand the thing that we're doing because if you don't then you'll fail the game right? Not sure if that that makes sense but I think that's kind of where we were going with it to to answer that question. 41:32.88 archpodnet Right. 44:19.90 Paul No thank you that that makes sense. 41:51.92 archpodnet Okay, yeah, um, all right? Well this has been really fantastic and. So Paul and I have a lot more questions to ask j d and he's willing to stick around for a bonus segment. So if you're a member of the archeology podcast network head over to http://arcpodnet.com and check out the bonus segment area in your members area. So if you just go to the top of the page or click on the hamburger icon. You can see your members area and then go to bonus and or actually just go to hq downloads or free downloads. Whatever we call it I don't even know I made the whole thing. But anyway, go there. And you'll find the ad free version of the episode and then the bonus segment sitting right next to it and and it's right. There. So if you're not a member head over to http://arcpodnet.comforward/memberssupport us get more stuff like the bonus segments and and other things and with that I think we will say goodbye to Jd for this part but again. Members check out the bonus segment that we're going to record to continue having this conversation with jd all right. We'll see you guys next time. 46:02.98 JD Calvelli Um, thanks everyone. 46:26.60 Paul Um, thank you.