 She is going to talk about access to Python education for schoolgirls using avocados, zombies and Korean. That sounds amazing. Okay. Okay, well, good morning, or at least good morning from London. I hope you're having a good year of Python. I'm sorry, we couldn't be doing this in person. I'm sure we're all sorry actually. It's been a pretty stressful customer once I'm sure for everyone, but you know, at least we're at least we're here sharing online and I'm having a one sided conversation with you, which is quite weird but same for everyone. Okay, so I'm just going to start me just adjust that I so I see less people. Okay, so accessible Python educate from schools using avocados, zombies and Korean, which in more general terms. Oops, make sure I can move my slides would be about accessible Python education using scientific computing and data science. So, I would also just want to say like, in normal presentation, I want to ask, you know, I want to understand yourself as an audience I'd like to understand, you know your background and your interests and you know are you educators are you school teachers and so forth but I can't obviously do that. So what I'm going to do. I think the way I'm going to try and use this time. I'm basically just kind of try and highlight the aspects of this story. The slides are actually meant to be. They're actually overly detailed they're supposed to be standalone. So I'm just going to set this expectation up front just so you understand how I'm going to present. The slides are meant to be standalone so if you don't listen to talk, you can understand it fully. It's also supposed to compensate for the fact that obviously we're using English. And that's obviously not the most accessible for everyone. So I've written out a lot of the text so that it's easier to follow. So hopefully that's useful. What I'm going to do is jump around in the slides a little bit and just highlight the main points I want to make. I think that we can have a conversation because I think there's a really big opportunity in this kind of space for school children, you know school girls to access scientific computing. I don't have all the answers, you'll find that out. But you know, I want to start the discussion and I'd really like to, you know, talk to yourselves and collaborate and get your feedback. So, okay, what I will go into detail is what the point of this talk is. So I'm just going to set the expectation up front. So from my observations, at least in the UK, technical education for teenagers seems to focus on these kinds of things. So, you know, this is, I'm a very visual person, I'm probably going to talk to the visuals more than the text. So if you ask a child who's doing coding, what they're doing, it's going to be very likely one of these kind of options, Raspberry Pi or Arduino, Minecraft, mods, scratch. Really interestingly, I had to make this point. There are actually a lot of competing teachers in UK schools who like VB.net. Yeah, I know it's a blast from the past for a lot of people, but that is the reality. So my point is that when it comes to learning Python, the work that my startup is doing in UK schools is demonstrating that scientific computing and data science education can actually be really effective for some teenage students. So basically this talk is going to present a case study. It's going to be of a scientific data stack education based in a London Girls' School. And I'm going to show you the kind of successful learner outcomes that can be delivered by this kind of stuff. So I hope this is familiar to you. I'm probably not great. So I'm going to have to assume that people sort of understand what Jupiter and these kind of technologies are, but we can obviously talk about the breakout if you want more detail. So basically by the end of this talk, hopefully there is more advocacy for scientific computing and data science as accessible Python education for teenagers. You know, because after all, why should us grown-ups have all the computational fun. And it is a lot of fun. I'm not going to lie. I'm biased. So I'm going to show you the data science. But I also want to make a point. Even before the pandemic, I actually found scientific computing. I actually found this science actually really helpful for my mental health. So, you know, if I felt really stressed out or really out of control as someone who is, you know, higher on the autistic spectrum. encoding often helped me, you know, feel more. It gave me a focus. It made me feel more comfortable. And actually, that's been something I've turned to as well in this pandemic. So, you know, in the UK, we've really not handled this COVID very well at all. I'm not going to get political about it. But, you know, I'm still effectively in lockdown. And, you know, although we've been very privileged in the UK in a lot of ways, it has been tough. And, you know, in the same way which Data Science has helped me cope better in this time, you know, I really want to make sure or, you know, really want to give that opportunity to other young people who maybe also might find Data Science and Science of computing, you know, something actually really positive for their well-being and not just from a kind of, you know, job tech future perspective. So, something I have to point out out front. So, obviously, we're talking about children, you know, they're safeguarding. So, the students are not going to be disclosed or identified in any way. So, they're going to be anonymous. I will have you visuals mainly from this is one of the meat communities which I founded, my main one. But if you kind of want to have a sense of the kind of audience I'll be talking about, if you are familiar with Harry Potter, basically Hermione in the first year of Hogwarts is the right kind of age and sort of gender. Sorry, that will give you a bit of a proxy. But, you know, without the magic stuff because that's fun but not real. Okay, so here am I. I am a tech entrepreneur. I've been working for really quite a while now. I think probably the point I want to make is that I don't come from a formal education background. So, I actually I'm a social scientist. Yes, I know it's kind of weird that I've ended up in Python via R. So, this is the first community I founded was called the RLadies Cone Club which turned into RLadies London, which then became a founding member of RLadies Global. But, you know, here we are, here I am at EuroPython for the second time. And so, how did I really get into education? So, over my career, I really found that I had a, I seem to have a sort of particular ability to explain, you know, technical content to non-technical people. And then that was something I turned into community outreach. Obviously, here with R and then with the AI club. And then I managed to turn that passion into a full-time job. So, I founded an edtech startup with a social mission, which is around democratizing opportunities for young people to level up their modern computing literacy. Now, that is not a sales pitch. It actually is generally because I need to set the context for the case study, which is about a project which my startup was commissioned by a girl school. Again, I'm not going to identify the school. It may or may not be represented by this image here. I think that's for you to decide. But the project was around delivering a bootcamp for non-coders to develop Python proficiency. And I just wanted to give a shout out to the project sponsor who I can name. Just giving permission. So, this is Madeleine Copa. She's actually did math at Cambridge. And so, she is the person at the time who initiated this really quite unique pioneering project. And we'll talk about the pioneer aspects actually through this talk. Again, I'm going to go through all the details. The key things I want to pull out here is that it was compulsory. So, this is unlike the kind of meet-ups seen that we probably know and love where people opt in. Here, these girls did not opt in. They were forced to undergo this bootcamp all 120, 11 to 12-year-olds and five of the teachers. So, hopefully that sets some context. Okay, project deliverables. So, what was actually delivered? So, remember it was a bootcamp. I decided, of course, to go with Python for scientific computing as the sort of the computing application. So, I chose this data stack, scientific data stack to introduce the girls to. And, of course, the Jupyter Notebooks as a reference here. And the structure of the bootcamp. So, we said six sessions. So, that was in the last slide. So, I structured it as basically building blocks with, obviously, relevant scaffolding. Because, as I've just said, non-coders, from non-coders to this in six sessions or, frankly, six hours is obviously a stretch. And actually, because it was a stretch, I knew it was going to be challenging. It was designed to be challenging. That's the kind of person I am. I like to stretch myself and stretch other people. So, the project goals in terms of the way it was presented to the girls was basically, I was like, listen, I know, it's like you're 11 years old. It's like, you probably don't code if you do code. I mean, fantastic. But I'm going to assume that you don't have any experience. And that's actually normal. I didn't code when I was 11. I didn't code when I was 20. So, I said, in this bootcamp, anywhere you can get up on this pyramid, I called it, I said is a bonus. So, anything you can develop in terms of your computational thinking, critical thinking, creative thinking, fantastic. I guess one thing to point out here. So, unlike, I guess, other kinds of educational experiences, especially in this kind of school, which is very academic, I said, don't compare yourself. No comparison, no competition. You're on your own path and you have your own pace. So, just focus on that. Okay. I'm going to see if by the way you can still hear me because I just feel like I'm just like talking to myself, which I quite frankly could be so sorry. That's why I keep checking on this. It's perfect. It's perfect. Okay, cool. Yeah, just let me know. I don't want to be just babbling. Okay. So, project outcomes. So, there were a small minority of the cohort who had done some coding for some Python, but of course, not scientific computing. So, from a quantification, quantified, from a quantified perspective, we had 100 new teenage pythons, all girls, identified girls, and new coding champion teachers, which was super awesome. But I think I'll let the girls, they're sort of testimony, excuse me, testimonials, do the talking for me, although I guess I'll read out their testimonials, but anyway. So, I guess the things to highlight is that, from a sentiment perspective, I mean, they, on the whole, they really enjoyed it. In fact, so much sorry that one of them said that they love Python, which I'm not going to lie, was really wonderful. And, you know, it's still a really, still a really happy memory for me. But yeah, so, you know, from these kind of responses, you can see, they basically, they just wanted to keep doing it. They wanted to do it at home. They wanted to, in fact, they downloaded Anaconda at home by themselves. They wanted to keep doing it. Again, after the bootcamp was ending, they wanted to do it again next year. They didn't want to go back to normal class. You know, it was just awesome. Okay, this is a point I do want to make. So, obviously, I didn't know the students. I didn't know their backgrounds. But some of the teachers pointed out to me that there were girls who, you know, in science, for example, they were struggling. They weren't historically known for doing, particularly well on that subject. But actually, in the bootcamp, we were basically with Python, they were actually excelling. You know, they were sort of, you know, doing better than, you know, other, their sort of classmates who had done coding before. And, you know, for me, that was just really powerful. And again, that's just emphasized that, you know, I feel like scientific computing and data science could really give young people the opportunity to discover something that they really enjoy and are good at, you know. And I really want to make sure they have those opportunities to discover that so these awesome outcomes. So why really? Of course, as a data scientist would want to know. So I think basically, this is what I want to stress is that I'm not saying all experiences, especially not a tech education, I'm not saying they're always, you know, negative user experience. Although sometimes, you know, I'm not going to lie sometimes they are. But in this case, you know, for me, my analysis was very much that, yes, there are negatives, you know, there are things we don't enjoy. I mean, I don't know about you, but I don't get a lot of pleasure out of, you know, installation and setup. And that kind of troubleshooting is not my favorite. I also know documentation is really important. I like other people's documentation. I am not something which fills me with joy writing my own documentation. So, you know, these would be the kind of negatives. But, you know, I think in this experience, I guess, I just want to draw out that, you know, I tried really, really hard to sort of decrease the negatives, or, you know, make the things which might be bad, less bad, and the things which were good to really make them as good as possible and better. So I kind of drew out sort of six specific features. So just adjusting myself because I'm sitting on the floor. So six key features that I identified which kind of had the biggest impact on kind of creating that really good user experience. So I'll go through these in turn just to highlight the ones in yellow are the features which made the user experience less bad, the ones you mean, or the features which made the user experience better. So let's talk about that. Okay, so the most important one was really the choice of open source software. Well, the choice to use open source software as opposed to proprietary. So if you work in schools, you have experience with schools, maybe children, I'm sure you'll be aware that the school environment, the IT environment, is dominated by proprietary software. Again, you know, there are reasons, I'm sure. This is just my outsider perspective looking in. But for me, it was really important to introduce for scientific computing, to introduce Jupiter and Anaconda, because of their open source sort of nature. So for me, it really overcame a lot of potential barriers or a lot of barriers, which are often in place with other technical education experiences, such as the some of the ones which I mentioned at the beginning, but I'm not going to cast any shade. So I guess the most important thing is that there's no cost in terms of procurement cost. That's really important. So yes, this was an independent school, a private school. So this is not a state funded school. So they do have more resources than the average UK school. But we can't assume that everyone there, the students especially, so in terms of their ability to use the software outside of school. So that stack, I set up Anaconda and Jupiter or Anaconda, the open source distribution that's now installed in that school, in their labs. So I knew the students could access at school. Of course, I wanted them to also be able to install it at home if they could. And the fact that it was free was obviously a really beneficial aspect. So it was just basically whether they could or whether they had the time. Which also, thanks to the other point I wanted to make is that I really, I mean, I know those reasons why, but for me, for children and for something like computing, which is so such a critical literacy, I don't like that there was a trade off between the quality of software and basically what a child's family or legal guardian on behalf of a child can afford. That is something which I feel is really exclusive. So open source software is really positive for that. There are other things. So Anaconda doesn't need that much connectivity. If you are an adult with basic digital literacy, it's relatively straightforward to install. And at least for these girls, I can't say for all teenagers, but after I gave them initial explanation and a bit of practice, no problem, they're autonomous. They could do it all on their own, which is, again, why it was so easy for them to set up at home. So really quick tip, the Coronaze app. So a shout out to developer Nicola Holshu, I hope I pronounced your name correctly, who's created this fantastic open source app for iOS. So I do use this for another school where the girls have iPads and it works really, really well. And again, open source. Amazing. Okay, so prototyping focus is a really, for me, was really impactful in terms of creating a better user experience. So again, I just got to focus on the visual really. Basically, I basically got them to do data tricks, or at least at the beginning, because remember, first time coders, there was, you know, for some girls, a bit of trepidation, you know, it was really sort of unfamiliar. I'm sure there's probably some negative stereotypes as well. And, you know, and to be fair, I was making them do, making them learn, you know, professional level data stack. So I'm not really surprised that it was a bit intimidating, but data tricks. So especially using Unicode, that really worked really well. So obviously here you can see, you know, this is a good way of so after obviously they've got the code point for avocado, the avocado glyph. This is helpful for them learning how to use the print function, some mathematical operators. So, you know, I was like, okay, how many avocados should we make? Let's try 100. You know, you have a go. How many would you want to make? So it turns out we discovered that on a sort of standard Windows PC, you can make about a thousand, no, about a million avocados before it sort of starts to crash. But obviously no damage, you know, just restart. So from that kind of data trick, then you can start doing some more fun data tricks. So this works really well as well for young children, young or teenagers. A bit of what I'd like to call emoji math. So you can sort of see basically how that kind of prototyping as opposed to, you know, kind of long tedious sort of set exercise, you can see how this, hopefully you'll see this, this is quite fun for children. Scope originality. This is the one I'm just going to also talk about more detailed, oops, this chord, but that's not a message for me. Scope originality. So this is quite related to the last point about prototyping. So basically, so, you know, we said it was a bootcamp, you know, they had to learn a specific syllabus, you know, which obviously I agreed with the school beforehand. So, you know, in terms of these building blocks, that was set. But what I, the way I designed the experience was that, you know, we'd learn a building block, we'd learn some sort of computing information. But then I really encouraged them to think about a kind of a context to link it to. So in the kind of cognitive psychology domain, this is what's called knowledge transfer. So that's obviously a term for some of you, which has got picked up by deep learning. But the idea is that you basically fuse old knowledge with new knowledge. So this is your old knowledge, or your sort of existing memories. And this was the new knowledge to computing. And basically, when you relate the two, it becomes, well, you learn it more effectively, and it also becomes more fun. So let me give you an example. Oh, these are things basically the girls, some of the kind of things that the girls came up with. So, you know, you can see why that was quite fun in a Python class. So these are two examples. Here we go. We talked about avocados here, the zombies and the Korean language. So the building block here is obviously conditional statements. And then the context of the prior knowledge is technical term, was about zombie apocalypse. We did these. These were two things we did together, like as a class. So, you know, I basically set up the conditional statements and then said, you know, ask them to input, you know, what, what kind of a status is that they think, you know, we should have in this program. And then, you know, what kind of actions should we have. So that was really fun for them. And then also another example. So this was obviously around introducing dictionaries or, you know, generates the concept of data structures. And then we used Korean, or I chose Korean as a topic because I assumed correctly that girls that age would at least be familiar with K-pop, if not fans of K-pop. In fact, they were fans of K-pop. So they were really delighted to learn how to say hello in Korean, which is I hope we pronounce that correctly. So I said, in case you ever meet BTS, which is a really famous K-pop band. So they loved that. Okay, I'm just going to whack through these, the last points quite quickly, because I want to kind of finish in the next four minutes so that we can have some sort of questions. So low latency. So the point I want to make here is that there are times in technical education, definitely at these at the adult level, where, you know, you're trying a tutorial, you're trying something out. It's like it's a lot of effort. Okay, well, that's okay, whether regardless of whether it's effort, you know, you do something and you want to find out whether you're right or wrong. Okay, now you don't really mind too much if it's wrong. But you kind of want to know, right, you don't want to be waiting like an hour and then go, oh, oh, rubbish, you know, there was whatever was just literally this one character, you want to find out fast, so you want to fail fast. Okay, great if you succeed fast, but the worst is basically in terms of speed is if things are slow. So something which I think really helps in this bootcamp was that, so obviously drawing back to this visual, was that you could debug faster or sooner because you could find out your error quicker. So this is related to, so you know, that sequence of avocados and tomatoes, which is supposed to be a sequence. So you can see if they basically, they missed out the nested parentheses, then you get one avocado and a thousand tomatoes, which is not, of course, the intention, but you can find that out really quickly. And actually, it's not that bad, you know, you don't mind getting it wrong because, you know, it's kind of fun to try and solve, at least in this context, you know, they seem to find it, you know, a kind of, you know, just something to try and figure out what was happening and not they didn't feel bad about themselves or frustrated. So hence, less bad user experience. Okay. Okay, really quickly, this slide is about the fact that between myself and then my colleague, so Dario, so dogs remember well. So we were the Python instructors for this course. It, you know, it was noticed by the teachers that the fact that, you know, Dario and I are not, should we say, the sort of stereotypical homogenous tech identities and all background, you know, was noted as relevant by the teachers. So, you know, this is a very academic school, but it's also an ethnically and culturally diverse school. And so, you know, the teachers noticed that this, the fact that these were the people in front of them, you know, doing coding, you know, okay, yes, Dario has a very technical background that, you know, he's won awards from the Royal Society for technical things, computational food dynamics, which I obviously don't understand because I have an economics background. But, you know, again, that sort of challenges the stereotype that, you know, someone who comes from a social science degree, you know, can end up doing, you know, quite, quite cool stuff, which I won't talk about because this is not really the talk for it. But anyway, so that was a positive or less bad user experience that could have been. And then there is the fact that everything was sort of set in as much real world context as possible. So, you know, we talked about the pop culture and the fun sort of aspects and the originality. Because this school is very academic, I also knew that the girls would care about, you know, certain universities, certain kinds of disciplines, you know, certain kinds of organizations. And so I basically made as many links as possible with the Python that they were learning and these kinds of real world things, which definitely helps them or would have made a better user experience. Okay, so to summarize the user experience, those were sort of six key features. Obviously, there are more. But that sort of drove a better user experience, which drove the awesome outcomes that we saw. Of course, that's quite a mouthful really to go through all those different different aspects. I think in the most simple terms, the way I describe it is that the project was successful, because the girls got to discover how to do new relevant things they couldn't do before, and also upgrade things they were currently doing, which I like to call unlock and power up. And not just students, but also the teachers. So this isn't going to make a lot of sense if someone looks at these slides standalone. But this is a, this is a gift, just a little thank you from one of the teachers who had really appreciated the bootcamp experience, for the students, but also for herself. So the other thing I want to mention is that this case study, I've said scientific and computing education is effective. Of course, there are different approaches. I like to divide it into direct and indirect. So this case study would have been direct because it was very Python first. So it's a Python bootcamp, you've got a Python syllabus, yes, then we tie in with your sort of existing, familiar knowledge. But you know, the Python is always the kind of driver. However, there's an indirect approach, which I use with other schools, which can be really effective. So the way that works is that, you know, you've got their interests. You can sort of pick one, which is sort of more common, should we say, or work more students share that. So let's say art and design, ads in Python, computational eyes. And in this instance, for example, you get code art. So like I said, you know, the girls discover or the students would discover how to do new things. So maybe they've never done fractals before, fractal art like this, or maybe they already do fractal art. But actually maybe doing in Python, it gives them, you know, more power, more functionality. And basically, it sort of makes something they're doing already better. So I guess is the point really about meeting kids where they are, not where you are, or not necessarily where you want them to be, but where they are today. So in conclusion, I hope I've convinced you to at least consider that Python for Scientific Computing Day and Science isn't just for grown-ups. You know, it can really spark joy for teenagers and adults. So why not let teenagers have a go. And that was all I had planned. So thank you so much for listening to my talk. And please get in touch or, you know, hopefully we'll have questions. Send you a lot about your message to Cheney in our Discord channels. So I have a few questions for you. Awesome. Hannah is asking, do you have any advice for how to get involved with volunteering to teach coding skills to girls? Well, I mean, of course, you can always, I mean, we would always have volunteers, so my ed tech is, effectively is run as a social enterprise. So, you know, we do charge for projects where sort of possible, but, you know, we do do pro bono work as well. So if you're a volunteer, we would definitely like to find a way of working with you. But in terms of other ways, schoolgirls, I think, you know, there are, there are initiatives to outreach Python, obviously, to, you know, to everyone. I think for me, the scientific computing is really the aspect where, you know, there isn't necessarily that kind of opportunity or access and availability. So I would have to say, I mean, I also, I couldn't say for sure, but I, I fairly certain I haven't seen really anyone else doing what we're doing, or at least not with schools, or kind of on a more successful basis. I know there, there is a, there is a market leader in the UK who, I believe they do do something similar, but they charge a lot. Let's just put that way. And it is outside of school and in certain areas. So I guess, yeah, sorry, the quick answer really is, please go as much, you know, in the breakout room, maybe. Yeah, I think, I think you can. So there is a, you, you can continue this question in the, in the discord channel for this track, but there is also a channel in particular for this talk, the name of the channel is talk school, school gears. Sorry. I think it was okay. I think I saw earlier. I think it's quite into school girls. I have it here is Python for school girls. Yes. Sorry. So it's talk Python for school girls. And then one more question. Lil is saying great talk. How did you find the structure say in terms of technicians supports and you co workers that were involved in the project? What was their, their prayer background? There, sorry, was it? So what was the background of your co workers? Oh, find infrastructure, right? Like, if you have the support technical support, that kind of things. Oh, my co workers, we are a small team. It is, it was a how do I find my co workers or their background? Well, my background, the micro workers tend to have a kind of more general data science background. And then I kind of tend to use my experience, my experience in industry, basically hacking, you know, it solutions or, you know, just basically making things work. So it tends to be me. So in fact, it is me coordinating with the school ID departments, which is painful. I'm sure people are aware of that. But I hope that answered the question. But yeah, please, we'll talk about the breakout room if that didn't provide the information. Okay, so we have time for one more question. So if anyone wants to ask the question live, you can click the button to raise your hand and I can enable the microphone. So let's go. Okay, so I think we're done. Thank you very much. Cheers.