We have mentioned that I was gonna say, and it’s important at this stage early to explain that this isn’t about exorcisms. I like to get in there early as well, cause people do often just drop off podcasts very early. If this isn’t explained so, exorcism with an e, not X or schism within out. I have stories anyway. No. No.

It’s nice to meet you today. Thank you for having me. It’s It’s also really, really hot in England. I imagine my definition of really hot is somewhat different from your definition of really hot. Yesterday was the hottest day in UK history, which meant it was about 32 degrees Celsius here. Wow. So I don’t know what that is in your American numbers. But the 90 year like that was that I think we had 38 somewhere, which made it the hottest day in UK history.

So, yeah, today is being a bit a bit cooler. But yeah, British people, this is This is heart. You know? I get a sweater looking outside. That’s that’s impressive. Sigh.

But yes, so I think that’s one of the things I enjoy about doing things as well is that sense of once and complete things and a lot of programs, I think, fall into that completion ist category. So it’s something we try to play on a little bit more.

it was actually just Katrina started it on this Katrina. Yeah, and I join slightly later is co founder. And what a slight later, like a few years later. But yes. Oh, Katrina was working it sharing chairman school. I think when she is a place that takes people who a career changing on bond, she she found that like she needed somewhere where the students she was teaching things to could actually just go and practice the things she was talking about and so built this in house just to use it.

But also just like, I think, put it on the Internet for people to use on bond. Yeah, it totally unintentionally blew up.

Vision to came out of that like last week, maybe in, like, three of the languages Java script and a few others. And there’s alcohol free version as well. 99 bottles of milk, I think, or something. Eso yeah, are plugged that as well. And it is. They’ve also added some more content. So if you already got first version, you can get like a discount on this condition

Yeah, that’s about it. I am

Yeah, yeah, that’s some hardcore things like that. And then there’s like, such a range of different languages makes you realize how many different languages that are on it makes you ask yourself why so many different languages And then, of course, and I really going about its later? That understanding why they were different programming languages is what makes each of those programming languages then fun to actually dig into to understand why somebody is designed yet another programming language.

What is what was wrong with the little ones? That’s one of the really fun things about it. Yeah, a great range going Do prologue, for example.

Yeah, on. That’s good. Because, you know, you’re you’re sort of time is your most valuable asset and, you know, you get the thing, you get something back for that time and you can choose. You know, I’m enjoying this. I’ll put more time into it. You learn more, it becomes very in your control. And then I think because of that you have a constant intrinsic motivation to carry on with it. Where is often when you pay for something, you can sort of almost feel a bit or obliged to do it because you’ve paid for it.

And then on the back of that is like your motivation for them doing it is wrong and you have all that extrinsic motivation. That doesn’t mean you then do the best possible job you can. Yeah, Time is always the most valuable asset to choose where to invest

It’s is in the natural language paradigm. It’s It’s relatively easy toe. Pick up a few, for example, German words. I was learning German recently and like pick up a few German words in a few constructs and be able to, you know, order a coffee in a coffee shop.

It’s a lot harder to come across sounding like you’re a native speaker and hold a conversation with somebody And you know, you can always google how to translate some think, you know, on your app when you in another country but to be able to really embed all of that knowledge and those, you know, lack natural language paradigms to speak fluently to sound native. Um and that’s really what we try. And we try and teach fluency Accident? That’s what it’s about. Is that fluency?

and city question? So I think like, firstly, Exercism are not for profit organization. So I don’t a tall feel like I’m in a pastor with anyone like, I never go in, look other websites and think, How do we beat these people? Because our aim is just that people get better at programming. So if somebody can go in, learn treehouse or code academy and don’t need Exercism and you know that that’s great. I’m glad that they’re learning what they need.

Where I see exes, um, fitting in is a few different ways. One, Aziz, you You know, you said everything is free. Everything is unlocked, Say, for people who can’t afford things like code Academy in Tree House, which makes up the majority of the world’s population. Um, you know, we’re a good space for that. I think if you’re learning programming right now, Exercism, um, isn’t the best place to go like if you’re starting out from scratch.

But if you’re trying to pick up a second language or you’ve done six months of programming and you want to get deeper into a language, I don’t think things like treehouse or code academy again, a particularly up skill. You in those ways like they might teach you how to write in a slightly different I know. So you’re doing re. But they might teach you out also writing rails, or they might teach some of the framework for some of the things, but they’re not trying to get you to go deeper and deeper and deeper.

I think where Exercism, um really thrives is in our mental ship model. So within Exercism, um you you get in, excise, you work on it, you get given a test suite, you make the test passed. So you’ve solved the exercise. But then you can give it to mental and they look at your code and they make suggestions on how how your code can be improved, how it can beam or idiomatic, which is a word that we use a lot how toe how to use that language in the way it was supposed to be written, which is what we consider fluency to bay is you’re fluent in a language if you understand the idioms of that language and can use those idioms.

And so what that helps you do is if you’re learning, say, Java script helps you write Java script like a professional JavaScript developer with white job script. Not like really be developer who’s just gone learned to be a job script. Um and so I think a really big part of space that we feel is in that going deep into a fragment language Really understanding Azi talked about earlier. Like why that programming language exists. What is it that makes JavaScript Java script one makes re be ruby on Helps you really think in that way?

Um, we’re not. You know, we’re not trying toe beast your CV by putting a course on it. And we’re not trying to Teoh get you your next job. I think exes and definitely helps with both of these things. But at the same time, you know you will enjoy Exercism in the most. If you’re passionate about the language that you’re trying to learn and passionate about learning. And I think then on the flip side there’s also the opportunity for anyone to be a mental.

And if you think you learn a lot from doing this student side of things, you’ll learn 10 times more for mental like it still blows my mind. How many different ways the bar to solve any given excites an accident. So the most basic exercise on any track is called to for Andi. In most languages, you can solve it in 123 lines of code. I mean, it’s a really, really trivial trivial exercise.

It has a bit of string interpolation where you’re just adding two or three strings together effectively on, but it has a a method called or function call that has an optional parameter Andi. So it’s a couple of relatively basic concepts. But if you look at any given language, I mean, I can speak that can speak. To say Ruby and Python is two examples I’ve looked at recently. Out of the last 500 solutions to those 350 of them will be unique, even though the solution is 1123 lines long, Yeah, so many

say I mean out of 500 or 150 of them will have already shown up. So there are three engine 50 entirely unique thinks if you normalize for white space and variable names, you get about 250 UNIX. So really? Yeah. So there were 250 unique ways that somebody has managed to solve to find rebate after the last 500 on that, that hole in any language with

it is a simple one,

Yeah, yeah, so I mean, it’s when you realize there are six or seven legitimate ways that people managed to just define the method in re be on and like, Then you know the ways that you can have the basic two functions. One is this optional parameter on you console that might have an optional parameter, or having a premature return if statement or having and if statement that returns to different strings or lows different ways like that on then just the way that people managed to add to the three strings together.

So I guess in that exercise that were probably like four yeah, probably four different like units of the exercise like option ometer string break concatenation, defining the method defining the class on if you there were maybe 10 different ways that people do, each of those say, and then the cross section of those means that you confined to brave to unique at 500. So then when you scale item is any other exercise.

So if you then scale into an exercise that has five lines of code to solve, not three lines of code, Suddenly you’re increasing the complexity by an order of magnitude or two on. So when you get to an exercise I don’t like, Bob’s a good example. So Bob is solved in 30 lines on Bob Exercises. This It’s a teenager on that. The import something to this teenager on the teenager. It’s one of my favorite exercises, and it’s again.

It’s a pretty simple excites, You know, if if, if you if you write some think in block capitals, so it So you get a phrase which is, like you would say you would say sank in block capitals, and the response would be like, Stop shouting something like that. Okay, there’s, ah, you finish it with its information. Whatever means I will chill out. Yeah, exactly on, say. And that that, you know, maybe 20 lines of code and ruby, maybe 15 lines of code and Rebic.

I doubt if you looked at the last 10,000 solutions you find to that with, um me. So yeah, And this is where it gets really interesting Because actually, what you think about is, if you on a most of the sites where they give you an excise and then if you solve it, you you feel like you want. Like actually, none of the skill in Exercismn’t exercises actually solving the exercises, all of the interest, all of the skill is understanding why you’ve made the decisions to solve in the way you’ve solved it and how those compared to how somebody really experienced in that programming language would choose to solve it.

What those trade offs saw what those decisions are. And then as you start to think about why you made the decisions you made on why different people make the decisions they make. Suddenly everything becomes very interesting.

Yeah,

My okay,

on This is the interesting thing is that I mean, you don’t know what you don’t know on. It is not a way to google for that bit of information. You can’t Google for my solution. Doesn’t feel as good as it could do. So, like, no. Yeah, I’m sure at some point Google will get that get, But but yeah, this is. And this is where I think having that, you know, it is like code of you. It’s like you put it. It’s like you’re working for a company with some of the best developers in the world and you put a poor requesting with some code and they give you feedback and then you learn all of this stuff.

And that’s what you know, we sort of try and be. It is almost this, like crowd source code review facility where the aim is on. How can we How can we help everybody upscale?

I imagine the relative. So I asked him. I imagine the vast majority people don’t work at companies where there is code review on I imagine. You know, a lot of people go through their whole career with actually nobody to give them any died in sor feedback on their only option is to go on like, you know, training courses their companies put on or try and read some more blocks or whatever. And I think on So I think having the opportunity to have somebody commenting and teaching you and opening your eyes to things that you don’t know is you know, is a ridiculously special opportunity,

looking through all of your solutions and axes and working out, I found a nice nice.

So I’m It’s a good question on, so I think

Yeah, that’s a genuinely That’s the first thing I mean, I think it’s worth pointing out a couple of things on. So one Exercism, um, is primarily made by volunteers. So we have one full time employee. Then we have I’m a volunteer. Other people, like the other people who are in the leadership team of volunteers and finding time outside of the other things that we have to do. Eso all of the every language track has its own team of maintainers that, like the exercises for the languages and V three, is changing in a few ways, but that the ones that person to the question of when it launches is the work that needs to get done on DFO that there are two things.

One we’re writing a whole new set of language specific exercises. So every track is, I think you mean coming within a member probably like 30 or 40 new exercises per track that have been written to showcase specific features of that language. And so they’ll take you from how to do strings in Re Beall the way toe like How toe do really weird ruby things in rebate on on. Then along the way they will be sort of unlocking all of the exercises that are currently in V a tive veto exit.

There’s obviously a big piece of work there to write 30 or 40 exercises per language s. So we’ve got a wonderful team of 150 volunteers working on this. If you’re listening to this and you would like to help out, please de eso yet that side of things will be ready when it’s ready. We had a first hackathon day on it a couple of weeks back, which was really fun. There’s that side and then also we are We have a load of tooling. Now that’s track specific. So one of those things, for example, is a test runner.

So this specifically runs the tests of any solution that is submitted. That is a challenging thing in itself, because obviously you’re in doing that. Your learning code that somebody has arbitrarily submitted to you on the Internet, which is not something you necessarily want to do for your AWS bill. So what possibly go wrong? There are heaps of challenges in that, but that requires every track to write their own testament.

But we then also have automated mentoring for each language that is specifically written in each language as well. Um, so we’ve got trawling for that That’s being written a very long winded answer, as I have attempts to due to a simple question of When will this launch, which is my hope was two months ago. My new hope is New Year’s Eve. My a sort of end backstop Worst case date is like the end of Q one next year, but I hate that will be that late.

Yeah, So what? We have already have some of this on V to a to moment. It’s sort of semi visible to users. So we discussed earlier, like to for this first exercise on nearly every track in a good percentage of tracks. Now that’s 100% to solve problem for us. So every solution that you submit re be pretty much will either. Well, we’ll get automated feedback now at the moment that automated feedback. If it’s congratulations, this is like a perfectly acceptable solution that’s pretty idiomatic.

There might be one or two tiny things. We’ll tell the use of that. So you’re receive automated feedback from Exercism, um, if it’s negative feedback. So I this there are lots of things that you could improve or whatever that gets sent directly to the mental on the men talk and then pretty much copy and paste that to the student with a bit of their own sort of personality attached to it. So we’ve automated a year’s worth of mentors time so far.

So what I mean by that is the amount of time it would have taken mentors to give feedback on some of these early exercises that we’ve saved has been a year in the last sort of 6 to 9 months. So quite pleased that which has obviously meant the response time on all of the other stuff that mentors are doing instead has heart. I never know where the response times after double it. It’s got twice as fast. So yeah, so but for V three, the aim is that these new concept excited that I’m talking about the 40 exercises per track that attract specific.

We’re aiming to have, like 99% automated coverage on those say, for any solution you submit, you get some level of automated feedback from us, and this is a really interesting technical problem because, as I said at the beginning, every solution pretty much is unique even to the most basic of exercises. So there’s there’s two interesting things that we do here. One is that every time a mental gifts feedback on an X on a submission. If somebody else ever submits the same submission, we can automatically re use that feedback.

So if two people So we said, like normalizing for white space and variable names and that sort stuff say we’re rolling, we’ve rolled that functionality and that’s one of the bits of hauling the oil, the tractor building. But they normalize every solution for those things. But it means that if two people submit the same ruby solution slightly from whites based, slightly different variable names, we can ignore that and still give the same feedback and just sort of change the variable names when we refer to them in that feedback.

So that means that love them, waiting for mental in a mental, having to give the same sort of feedback hoovering over again. If it’s something we’ve seen before, then the mental never has to give the same piece feedback twice on. Then over time, we’ll just take all of those submissions that people are doing and can just automate rules about

Yes. So all of the’s new exercises are focused on learning on bats. Sort of the difference. So every exercise is being designed to teach a program a language concept so that there might be an excise that said earlier on, like strings on how you use strings in this language and that will be focusing on how strings are different in Java. Script to re B. So you’ll be learning. Not just this is strings, because we’re sort of targeted out programmers.

But this is how strings work in this language all the way down to more complex things like this is what a block is in Ruby in this how to use yield, and this is how you can do meta programming in BB And this is except except so each of these new languages is being designed to teach one, maybe two, very rarely three of these individual language concepts. Whereas at the moment we’ve got these props sex sizes, which is sort of cute little toy problems in which you might sort of serendipitously use a few different programming ideas. The mentoring mode is gonna change as well.

So at the moment you get mental ing on the sort of core exercises in V two were flipping that entirely. So there’s gonna be no mentoring on that core on those core learning exercises in V three. All of the mentoring envy threes concept exercises will be automated or not at all on bats because that exercises are designed to teach that concept.

So just by getting the tests passing, you will have had to have use that concept, and you’ll have to have thought about that concept and then so there’s no sort of slow down waiting for a mental, which is one of the big problems and excessive at the moment is sometimes you can get stuck waiting a few days for mental, which could be super frustrating. Um, I know Yeah, exactly. Everyone everyone does on. There’s no one more frustrated than the mentors who turned up to see, like 500 new exercises that need mental mental.

So So, yeah, you’ll have the learning process that you can work through entirely at your own pace. You’ll be able to then go in Seoul, perhaps excises on. Then you’ll be able to request mental ing on those traps. Excises and we will be will be proactively asking students to say why they want mentoring or what they want mental and gone. So you can say, Hey, this just feels a bit wrong To me.

This feels like there’s a better way of doing this a little bit a little bit like you said earlier with the Matrix example, you could also say, like this particular line like I know I could do it this way, but I don’t understand why this is better than that or whatever is on street trying to get the student to proactively be engaging with the mental about what they want to learn from this exercise, where they feel a learning gap might be, or just saying like is there anything here that could be better.

You know, I feel like this is good, but I also feel like I don’t know why. I don’t know. And so that sort of learning and practice division is going to become a lot harder.

it’s a great question. So I think it’s why, like one of the things that especially some of the existing mentors worry about is that by introducing more off the automation and by taking the mental in away from the learning things that we sort of are making mental inm or that a secondary element to V three on That’s not the case it’ll like. What we’re trying to do is move the mentoring to the place where the students and most receptive to it on where they can learn the most.

And one of the big frustrations that occur with sexism is if you’re a student and you’ve been waiting a month for, like, our moments, maybe to people giving up. But I’d like five days for a mentor to come along. Um, when you then get that mental often, you just want the mental to freeze your exercise so you can move on to the next. One is at the moment you have to wait from and talk to approve it, and that’s like a real frustration pain point, which means, rather than the students being there to learn absolutely as much as they can from exercise and, like drip every single sort of drip of learning they can get out of it.

They just want to move on as quickly as possible, which is them really frustrating for mental who’s given their time. And actually, often the mentors are just acting is like the gate keepers between on a student being out to carry on rather than like the person training Givins and give tips. And so what we’re trying to get to is a point where the mental in the students, um or lined in their aims so that students there in that mentoring session, because they really want to learn something because we’re trying to make their questions more proactive at front as well.

That means that often they’ll be telling a mentor what they want to learn, which is then more helpful for a mentor, Bob than just being given a wall of code, not knowing what particular supposed to be helping with. Also, things like more of this automation to give them give them hints on where they might want to point a student, too. Um, so there’s a lot of things like that.

We’re also building in a reputation system into V three, which is similar to sort of stack overflows, reputation system. So the mawr that you do toe help sexism and to contribute to excess, and whether that’s building exercises where it’s been tall ing, or whether it’s Mental Inc will be giving you reputation awarding your reputation as you do that. That, then, is something that is like your badge of honor.

If you like on the site appears next to your name, it means that you, if you have high reputation, will will move higher up the mental and Q’s so you’re more likely to get mentoring faster yourself. So the more you give, the more you get. Um, we’re also gonna be introducing annexes and shop. We have swag that you can get on. A load of stuff becomes unlocked the more you do so, the more you mentor, the more things you get locked.

We’re also trying to work out if we how we price that under the things with that, it might be that you get things about cheaper as well. If you’ve got more reputation, Children work all these things out, but yes, so there’s sort of some more injured extrinsic Gamification. Stiffer mental is that we’re adding in there just to make it feel a bit more fun and a little less all about a mentor just choosing to volunteer their time to effectively work through a que

Um so I mean, like, for good memory be professionally nowadays? Same. Yeah. I mean, I mean, I being using re be since, Like, veils came out in, I guess, like 2004 Ieave was some surveys time along a long time ago now. So yeah, uh and I love it. So I guess that that is my favorite on I also my degree, my undergrad degree was in a I artificial intelligence. Andi Andi, Back in the day, there was things like prologue were explored heavily for that. I really loved programming given in prologue.

It’s just fun and different and interesting. I’ve also recently discovered slightly more of a love for JavaScript. So after many years of of not liking JavaScript because of I guess how it had evolved without necessarily design or purpose. It seems to me that people behind JavaScript are have now really added a lot of, like, a lot of structure and a lot of thought into how to, like, make it into, you know, really modern, really modern programming language that that has it feels more like it’s got a clear paradigm and clear things in there.

And so I sort of I had to really unlearn everything kind of unknown about JavaScript in order to be able to go through and and feel like I’ve actually program probably in Java script. And I feel like I’m still early on that journey, but for the first time in, you know, being pregnant for a long time, like 28 years or something. But it feels like the first time I don’t just feel like I’m flailing around with my arms randomly, hoping to hit something. I’m actually like designing something front, and it’s working.

exercise tracks. Yeah, definitely.

You know, it’s amazing. It’s such a beautiful place. Why would I know? Good. And

Orson, thank you. Thank you for having me. Thank you, listeners for listening on. So, yeah, my two worlds right now, our exes, which is what I’ve been talking about. The thing I always say is we love people to come and help. Sexism is built by the community for the community, aim is help each other be the best possible program This way. Focus hard on inclusivity and welcoming people on way. We try and have the May sort of diverse group of people building exes.

Um, that we can try and then help the most amount of people. So whatever your background, whoever you are, you’ll find a welcome home taxes and we’re always looking for mentors. You could just sign up on get started on way encourage people to do at least a few exercises first, so you get a field howitzers and works, but you don’t need to complete everything we will say right now on this big push for V three.

If you’d like to help out with that, especially in writing some new excises if you’ve got any background and education, especially even if you’re you, don’t consider yourself to be in the best program in the world. But maybe you’re good explaining things or get a teaching things then we could really do with your expertise. You can email me at Jeremy a Texas and I A or pretty much anything exes, and I choose your favorite at We’ll get to me. We have a big get have organization.

We have 250 different repositories that make up sexism, so feel free to jump on to get a dot com slash sexism and just have a nose around. And my other world is kinda kaido dog, which is a start up that I am building with some lovely people, which is tryingto help companies take better care of their employees. Way are a relatively successful with big companies way have some lots of the company’s bugle at the moment. I hope by the time this comes out, we’ll have started doing some stuff with us to look after their employees as well.

But we’re really trying to push to small businesses right now to understand the needs and the the challenge is that small business. So if you work for a business any size, but especially small businesses, I would love to hear from you about how you would like your employees to look after you. It’s really helpful for us to understand that. And if you’re an employer being just a manager in a company or you run a company, I would also love to chat to you about that and to understand how you might want to take better care of the physical and mental well being of your team and build their health and, well being into company culture.

You can email me again at Jeremy Exes and die, Or maybe more pertinently, at Jeremy Kaido torque on. I’d love to chat to you about these things as well. On you can find me on Twitter. I’m I heard everywhere. That’s a I H I D. Everyone light on you can find me at Twitter com slash i hit, Although I want to Sort of permanence about sexual media was about myself. But I’m also the personally and also the person that tends to be the one tweeting out that Exercism on Twitter as well.

Which is, I think, sexism. Ioo. Exe him and destroy. You have to find underscore undescribed. Thank you. Twitter will try and say I think you meant ec sources. And and you have to explain. Um but yeah. I mean, the website itself access and that is free. Come on, join in. Just try it. See what you like. Try the command line and keep your keep your ears open for when we launch the three. What? We’re making a song and dance about it.

We don’t advertise anywhere. We’ve grown to 1/2 1,000,000 people just by word of mouth on that seems to be working quite well for us. So the other thing I’d say is, if you do like sexism and you enjoy it and find it useful, please tell your friends and your colleagues and encouraged we rely on you to do that

this tool. But thank you for being the final thing I would say as well is that we’ve been working with some researchers in Chicago on, and we think some research into how people have a programming language is that people think about use affect the way that they program way. They learn new languages. It’s an interesting project. I’m hoping that we can publish some interesting stuff on it and help push the science of programming education forward on.

There were so helping us funding employees and on us to help us with stuff. So we’ve got a website research taxes and where you can go in and you can just do a few programming exercises. They’re all sort of 15 minutes to 30 minutes. You can try them in a few different languages, and we’re gonna be then getting some people to face, um, data science and artificial intelligence and just general hashtag science that your code to understand a little bit more about how you program.

And we have a survey there for you to fill in this world. So if you want to help us out, that’s a really hopefully fun way for you to give us some interesting dates to work with us. Well, that’s research dot Exe ism data, but that’s it. I hope you find excessive use. Thank you. If you use it. Thank you. If you contribute to you mental as well. Your wonderful people. Thank you for having me on the podcast. Thank you.

um, like profiles. But let us see. Like what you guys gotta be. Yeah. Yeah, you can.

a picture, maybe don’t share it with its can connect with and then come tell us

slash discord and come tell us about your exorcism experiences. I don’t like the way also love to learn. So you’ve got bad experiences are frustrating experiences. You’ll always have a sympathetic ear here as well. So feel free to tell me your experiences so we can learn from them. Well, I do never really good things that are wrong, but which one? But

directly. Exactly. 100 set