Demystifying the Rails Codebase with Ridhwana Khan - RUBY 675

also help write the Rails guides, which is awesome. That's one of the nicest things about rails is when I'm talking to people and it's like, well, how do I even get started? It's like, we've got these great guides. Just go, you know, go spend a half hour or so with them, and then you're pretty much off to the races. So why don't you go ahead and introduce yourself as far as anything else you want people to know about you, why you're popular or famous and all that stuff.

I think if I had to sum up the things that I'm passionate about, it would be making code more accessible through documentation, and that's why I write the Ruby on Rails guides, thought for engineering, and just inclusive community building, which is why I work at Circle because there are community building platform and I've built many communities over my career. I try to both communities where I see a gap. So the most obvious one is sort of the woman

in Tech gap. So over my career I've both communities like women in Tech. I've both Rails goals in South Africa, Rails Bridge, Black Gull's Code Ladies that You Ax, which is the intersection of a development and user experience. So just really passionate in that space. And on a personal note, if I'm not coding, I'm trying to keep an eye on my ever growing collection of house plants and trying to keep them alive. They're constantly dying. I also really

enjoy traveling and going to speak at conferences. It's one of my favorite things to do. And finally, just hanging out with my very opinionated feline peer programmers or per programmers, Zoe and Zeus. If you watch my talk at Rails World, you've seen some really goofy photos of them. Yeah, it's basically me.

how did you get involved there? Did you just I don't know, go tell the foundation, Hey, you need to fix the easer. I don't know what will happen there.

see how it goes. And yeah, I started writing for the Rails Foundation myself and another writer, Boomy. We started off and we worked with a couple of other peoples to people to review the guides, and that's how it all got started.

go in and we create new guides. Also, if contributors create prs for new features, sometimes that means we go through the pull requests and we update the guides with new information on new sections.

look at the existing guides. The trick will already have some notes in on what he thinks can be improved. We look at it with fresh eyes. We then decide what can be improved from our side. Sometimes that's how I started diving into the source code. I dive into the source code see how the internals work, because I feel that if you really understand how the internals of Rails works. Then you're able to explain the things better. Once that's done, we send out a PR. It goes

through internal review first within our team. Once internal review is done, we then create a PR for the open source community to be able to review the guides. Maybe they pick up some areas or gaps that we may have missed, and then finally we emerge it.

label Rails Foundation. So if you're trying to find any of the docks, you can do that, and we really encourage the Rails community to review them because it means that if you're learning Rails, or even if you've been using Rails for a long time, you basically know we some of the gaps lie what you've struggled with, and when you review them, you can bring some of their fresh perspectives to our prs.

So for example, in the talk I talk about mime types, I've traced through how RAILS handles the different mind types using math. Missing I've done. I've dug really deep into active recorde associations because that's a very big part of RAILS as well when you're writing applications, so things like the definitions of the dynamic methods and so on, I've looked at I think I mentioned in the talk, as well as validations. So RAIL uses like the dynamic callbacks

things like that. I've dove into some parts like that, and I think that the most common pattern is just the meta programming aspect of it. Right, So trying to find those patterns across the different parts of Rails or components in Rails and then trying to understand it a little bit better. The Rails codebase is really huge, and there's like always more to learn, but the magic is beginning to feel a little bit less like magic is you start understanding more of it.

in place, It does the work. And one thing I hadn't thought of with the mind type mind type is you define the method and then you just call it. I was like, oh, that that makes sense. So yeah, I really liked that, and really liked just the clean approach to explaining how all this stuff works.

how those patterns are applied across the different components. So if you understood it in one component, you can go and dive into a different component and it has very similar patterns. And once you think of it like that, it feels a lot more approachable. Because when I started off in rails, and even like up until a few years ago, I there was all this magic happening in rails. So when you reach a point of like, hey, I'm doing this thing, but it's a little bit more complex,

and now Rails isn't playing nicely with me. You don't really know where to go from there. But if you try reading the source code or you try looking into how Rails is actually working, you're able to move past

that blocking point and do more complex things. I mean, ideally you don't want to break away from the magic, but unfortunately there is real of it is when you both complex applications, you end up needing more tools under your belt, and I feel like understanding the code base is a tool under your belt.

it is that I'm trying to work on. And it it became pretty apparent once I started to learn Ruby. It was like, oh, it's doing this because Ruby works this way, and so it's just using the Ruby mechanism to do the thing. And so yeah, I love that idea of saying, Okay, look go look at the internals. It may be something you haven't done before, but it's it's not a mystery, it's not complicated or hard. It's just, you know, it's something that maybe you need to learn

a little bit more about. And then you look at it and you go, huh, Okay, this is where this is coming from. This is why I have this the way that it is.

And so by going through the Rails code base, you were able to also learn problem solving skills.

to follow. So I appreciate meta programming, but I also respect that it has a lot of complexity and it can make code harder to debug, especially if you're not familiar with the patterns, or if it's overused without like clear naming or clear direction. So I wouldn't say that I feel strongly that everybody should use meta programming. I think that meta programming needs to be used based. I think metaprogramming has a place, and depending on your use case,

meta programming may be a good fit. But if you're using metaprogramming for any for everything, you may just be over complicating things.

And that's the way I think about metaprogramming is is you know, you don't want to take more out of the wall than you have to, so to speak, you know, you don't you have to be careful not to hurt

yourself with it. Sometimes the code can be hard to debug when you're using metaprogramming, and so, yeah, you know, what I find is most of the time when I'm reaching for something like metaprogramming, I am doing something on the level of rails, right, And so what I'm doing is I'm actually using it to provide myself with the pattern I want to see in my code, and so that winds up being in like the lib folder, you know,

somewhere else where. It's it's not code that I'm going to go touch all the time, and it's not code that I'm using to solve all my major problems. What I'm doing is I'm just basically using the metaprogramming to create a better tool for me to solve the problem.

profile fields for their members. So, for example, in the talk, I use the example of a go gardening community, and for a gardening community, during onboarding, we may want a profile field that a user can enter their favorite plant, or a developer community might ask if you're available for hire. So these fields need to be highly customizable, and they

might need to be dynamically edited at run time. They also might support different data types like strings, booleans, dates, etc. So rather than hard coding a bunch of different profile attributes that are different for different communities, we use mata programming to dynamically define gatters and setters for the custom fields is the community has created them, and it meant that communities could define as many custom fields as they

want and the system would just respond. There were no migrations or extra model code or no like hardwired logic. It just made the system really flexible. And I feel like that's a great example of how meta programming is used in a real practical way to solve a problem. But like you say, I feel like they are very

like meta programming. In my career, I think that's probably like one of the very few cases I've needed to use metaprogramming, And maybe there's like one or two others, But over ten years you're mostly building standard applications that don't necessarily require that level of customization, right.

is documented. I think that's really a superpower. If you go through the Rail's code base, you'll find sort of method definitions, you'll find how the Rails API documentation is within the code base, and you can see what the different methods are doing. So I think there's a lot of sort of other practices that the Rails team uses within their sort of actual development, and I found a lot of those useful alongside the matter programming aspect and

the great architecture. I also think so may one more thing to air the like the PR. I found that when I'm writing documentation, I also sometimes tend to go through like the open issues or poor requests to see did someone touch on this. Maybe there's something I don't really understand and I want to understand it better, but maybe the rail's code base doesn't provide as much information. I find that when I go through poor request reviews or even issues, contributors and the core team are very

for both about explaining things. They do a really great job about articulating themselves and working together to form a solution. So it's really nice to go through poor requests, to go through issues, and I really encourage people to just if you're not contributing to rails, or you're not reviewing documentation prs or even just normal rails code, like just browsing the issues and the poor requests casually can provide a ton of information and useful knowledge.

one poor request in my whole life against rails. So if somebody wants to get involved and they're sitting there and they're thinking, okay, well, I just want to understand it, you know, and maybe they want to contribute to the docs, or maybe they you know, just want to understand things better and you know, maybe help with some little thing here or there. Like where do you recommend that people start with the RAILS code based to just understand what's going on in there? Yeah?

You don't need to be writing the guide to do that. You just need to be someone who wants to make it clear for the next person reading the guides. And especially I feel like a lot of people that are new to the industry or that feel they feel they don't have enough knowledge to contribute to RAILS or something or the other. That's how I felt for a long time.

I feel like documentation is a clear way to contribute because if you're not understanding something in the documentation, I can guarantee you there's hundreds of other people that are not understanding it as well. Another way to get yourself familiar with the rail's code based if you're looking to just dive a little bit deeper, I think read the tests. They incredibly instructive. So the Rail's test suite is massive, but it's very descriptive. It shows you how methods are

supposed to behave. It often gives you breadcrumbs to trace back to the source code. And when you're trying to understand something or if you think you know fixing a bug, test are often the best entry point, and writing or improving tests or just reading the test is a fantastic

way to start. And then if you're looking actively to contribute, you can look for a good first issue or you can look, like I mentioned, the docs, TIG the rails getthub repots very well organized, so there's very often open issues that are labeled as like a good first issue or where we need contributions on the documentations. And finally, like don't be afraid to read this was called Even if you don't understand it, that's fine. Just a lot of reading of it. Eventually your mind starts to put

the pieces together. It's like the more you read, the more somehow your mind just like connects different parts of it. You start seeing patterns emerge. There's lots of tools as well that I mentioned in my talk unlike things that will help you to navigate the rails scode better. So use some of those tools to navigate the rail source code. And finally, you don't have to do it alone. There are so many people in the rail's community who are

so kind and supportive and willing to help. So whether it's in rail slack or discord or get up discussions, you can ask questions and you can get unstuck. I think the main thing, finally, to wrap it all up, is that rail's contributing to rails isn't really about being perfect or knowing everything. It's about just being curious and wanting to make rails better, even if it's in small ways.

a really big guide. I think it's in a good start, it's in a good place, but I think they could also be a lot of improvements. We could explain things a lot better. We could maybe move out sections, make them smaller, more approachable sections. I think it's one of the biggest guides, and so I think just breaking it up into smaller, more digestible pieces will make it a little bit more approachable. There's also some other sections that

could use some improvement. I think we're already in a I think the Rails Guides in general is one of the better guides we have out there across languages. I think the Rails Foundation has been doing an amazing job, but there are just individual ones that are working progress. So like the rails initialization process that's a work in progress, as well as the active support instrumentation things like active

record encryption. So the first thing we want to do is we want to finalize those working progress guides because they may consider they may have right now not like the full picture picture of what their topic income. So we just want to fill out that information and cover the bases, and then we'll go about improving some of

the other guys and breaking them down. But we're hoping to achieve a lot this yeah, and improve it significantly, as well as create some new guys that haven't been covered with RAILS eight.

switch over to both Cape Town and to Circle. So what do you do at Circle exactly, Like, do you work on a specific part of the platform or do you just kind of hit whatever needs to be hit. And by the way, I'll just explain because I've used Circle and what I find is if you work at a place, you kind of look at it differently from the way that people use it look at it sometimes. So Circle is essentially a forum. You can think of it as kind of a focused Facebook. It kind of

feels like that Facebook group. And then on top of that you can add like courses and other things to your community. You can set it up as a paid community or a free community. And yeah, it's a really well put together system. So yeah, so what kinds of things do you work on there? And yeah, what kinds of concerns or improvements do you find at work?

the application itself. So specifically, I work on what we call the Email Hub or marketing Hub team at Circle, and we're basically responsible for things like email broadcasting and sort of that domain whereby you able to send our broadcasts to either community members or what we call contacts

within the system or leads. And there's just a lot of are things that tie into it, so for example, analytics and that ties into things like workflows where for instance, when you're onboard a community, you might want to send a broadcast and then we might want to wait for five days and send them a follow up things like that. It's a relatively new team at Circle. I think we launched the email hub slash marketing hub feature last year. I want to say around September October. It was around

Rails world time, so it's relatively new. We're actually hiring for the team at the moment, so I'm leading the team and we have our two engineers and we're looking to hire another two engineers for the team. So I've been doing lots of interviews and hiring at the moment. Our tax stack is Ruby on Rails and we use JavaScript as well. It's a pretty big engineering team, so each sort of sub engineering team is focused on a

particular part of the platform. So we have like a CRM team and a CMS team and you sort of focus on a specific domain.

that level of improvement. Those are the best companies to work at. So if people want to apply, how do they do it?

we can't collaborate when we need to. So maybe like with a one hour two hour overlap just in case we need some sink time, so that gives us sort of good collaboration across the board. But for the most part, we use as in tools like Slack and Notion. We document things very well.

down a little bit. I'm also not originally from Cape Town. I was based in I was living in Joburg, Johannesburg, and then moved to Cape Town only like five years ago. Johanna's Burg also has a really big Ruby community. Is to have constant meetups every couple of weeks, and yeah, it's really good. I will say though, because I work for a lot of US Over my career, I've worked for a lot of US companies, my community tends to

be more sort of global. I tend to attend conferences where I'm sort of traveling, and so it's been a little bit more difficult to keep in touch with the rails community within Cape Town. But I'd like to see Ruby Fuza hit the ground running again. That would be really cool.

gardening system. So I've been tinkering with that and it's been really fun. But between writing code the Rails Guides and then are doing some of the tinkering, and then also speaking at conferences, I've been very busy over the last few months. My next conference is I'm really excited about is Ruby kontailiand and that's next huge January. One of the keynotes speakers, So I'm really excited about that.

about a topic. That's not true. I very often I will put together ACFP about something that I know nothing about but want to learn more about, and those make for the most interesting talks. So things that you're interested in, things that you may have learned that you think others

might find useful. And especially I find that the cfps that have the most success rate, at least from my personal experience, is the cfps that talk about that have a story that is something that you've experienced and you

want to share with the world. So, for instance, my reals will talk demystify in the rails code base that came from a storyline of being a technical writer and diving into the documentation and some of the difficulties that I experienced and then sort of what I did in order to be able to learn how to demastify the rail's code base. So a lot of the time it's things that you know, sorry, things that you've experienced, and

you want to tell that story. I feel like writing an abstract first, even before you have like a full idea of the talk, that's okay. Sometimes people write their talks first, or they try to have every not necessarily to write their talks, but they try to have every aspect ined out first before you're submitting. They feel that they need to know everything before submitting. You really don't

have to. It just needs to be a pitch and then just ensure that you have enough time before the conference talk of course, and then you can learn on the way, right, Yeah, and start small and then grow. So if you've never spoken before, a good opportunity. I remember when I started talking, the first kind of places that I submitted my talks to were like local rails communities that meetups that I went to and I would

talk there. I would get a little bit more confident in front of people, I would get feedback on a talk, and then I started applying to the bigger ones. And that way, you can also put in your cfps that like, hey, I have spoken at these different communities, I have experienced speaking things like that.

picked before. But we played a new one yesterday with my So I played board games every Wednesday night with my friends, and we played a game called Colt Express and it's a train robbery game board game. Geek waits it at one point eight four, which is kind of So two is what is about where I tell people the casual gamers are kind kind of going to go, Okay, this was a little bit complicated, but not so complicated that I didn't like it, and it was fun, right.

A one is like the game should play with your kids, right, so you know, and so it's got enough to it to where there's some competition and stuff, and some of the ones are really fun, right, there's simple dynamics you just but anyway, so this one's one point eight four, so it's almost a two. It's called Cult Express and uh it plays two to six players. My friend has the the Cult Express Big Box Edition, which has two of the expansions in it, so his will play up

to nine. And apparently there's another expansion that adds another player too as well, so you can play with a lot of people. Came out twenty fourteen, and what it is is it actually has a little train cars that you put together for your game board, and so you have the locomotive and then you have one train car per player, and there's loot on the floor of the train cars, and so then what you wind up doing is you have a deck of cards, and the cards allow you to shoot at the other players. You can

punch the other players. You can move up to the roof of the car or back into the main part of the car, and you can move across the car, so you can move from one car to the next. And there's a marshal and if you wind up in the car with the marshal, then then he shoots you and you move to the roof of the car. The marshal doesn't go on the roof, and so then the rest of it is when you owe, and then you can pick up loot, you can move the marshal. I

think that's it. So if you shoot another player, you put the shot card in their deck, right, and it's a no op card, so it dilutes their deck. It makes them harder, It makes it harder for them to get what they want. If you punch a player, then you make them drop a piece of loot. You know, picking up loot's worth anywhere from two hundred and fifty to five hundred dollars. There are gems that are worth

five hundred dollars. And there's a lock box or a strong box on the locomotive where the Marshal starts that's worth a thousand dollars. And so you just move around the train and try and get as much money as possible in five turns. And the way you play the turn is whoever's first, they put a card on the pile, and then you go around the circle and everyone else plays a card right, and so you can see what

they're playing. On the turns where you play face up, but there there's sometimes there are twists where you go you go around the circle the other direction right, and so you may get If you're first, you know you're it doesn't change a lot for you. But if you're set, if your last and it changes the direction, you get to play, the first player plays and then you get to play again, and so you know that that may

change the dynamics. There's others where you play and face down, so for for those cards on that on that round, you don't you don't see what the other players did, so you can't react to it. There's another where you play two cards at a time and so you play twice right, and so you can move and then pick up loot because you know that you're gonna be able to get there before anyone else. Or you move and you punch another player and make them drop loot because

you know that they're going to be there. And then if you shoot the most shots then you you get a thousand dollars. So anyway, then you tally it all up at the end, and and anyway, it was a lot of fun, a lot of fun. So and the game board is really unique because it is it's actually a train that you just set up on the table and so, you know, I mean it's real small. But anyway, it was a lot of fun, really really enjoyed that.

It took us about an hour and fifteen minutes to play, but we were learning how to play, so I think you could probably play it. We played it with five people. We probably play it in forty five minutes, I think. So anyway, very very fun. So I'm gonna pick Cult Express and then I'm going to shout out about a few others. So yesterday we had a team lunch with my team and basically we got sent gift cards for door Dash, right, so I had food show up at my house and then we played wiki race and that's

at wikidashrace dot com. And the way that wiki race works is you and you could put in your own things that that you start at and end at, but it starts you on a page on Wikipedia and then

you have to click links to get to the other page. Right, And so one of the ones we played that it was like some soccer player from like the nineteen fifties that you know, there were probably like four links that went there, right, and so, and I can't remember we started at but it was something like completely weird, right, and so you start figuring out some of the tricks right where it's okay, well, I know this guy's from England, and so I'm going to get to England, right, and

so I click some of the links that take me to the world and then countries and then England right, and then I know he played on these teams, so I go to the soccer leagues and then England right, and so anyway, it was It's very, very fun, and it's basically a race, so whoever gets their first wins, you know, and then when somebody wins, it shows their progression through Wiki Wikipedia. There was one that we got

that everybody eventually just gave up on. So it is possible to get some that just are hard to figure out. I'm sure, well I'm not sure it was possible. I just think it was really, really, really tough to figure

that one out. But anyway, it was really fun, and they picked some really obscure Wikipedia articles sometimes, and so that's fun too, because part of the game that we figured out pretty fast was Okay, the first step to figuring out how to complete this chain is to open another tab in my browser and go look up the thing that I'm trying to get to because I have no idea who this person is or what this thing is, right, and so you go look it up and it's like

it's like, oh, it's a plant South America, right, all right, so you know, then you start figuring it out. So anyway, tons of fun and so I'm gonna pick Wiki Race and then I'm just trying to think through things because there was something else I wanted to pick and I can't remember what it was, but anyway, I'll just say it for next time. Do you have some stuff you want to shout out about?

then put it together. So I've been looking for a good tool that will be able to help me a little bit with no taking, especially with AI and its advancements. So recently I've been exploring a few of these. So the one that I've been using has been CRISP dot AI. I don't know if you've part of it, but it's

basically an AI assistant for collaborative meetings. So some of the team meetings are basically enable CRISP and it will It does recording if you want it to, or it can do transcribing, so it will give you the entire sort of transcribed version as well as the audio recording of the meeting. But that's not the coolest part. The coolest part is that it will give you a summary, It will give you the action items from the meeting. It will give you a detailed outline if you ask

it to. But then also it has this AI chat that is attached to sort of every meeting once it's been uploaded, and with the AI chat you can get pretty specific. So if there's a specific part of the meeting that you want to dive into, you can just ask the AI chat to summarize or detail certain parts

of the meeting, which is really nice. Sometimes the output of the meeting is I maybe need to I have specific questions and I just want to summarize these, summarize the meeting, and answer these by answering these specific questions. And so I just go in and I put in the questions in the AI chat and it will basically spit out the solutions that we came up with in the meeting. So really really cool. I've been exploring that against a different AI too, which is Granola AI, and

that's like an AI note piered people in meetings. So the difference being is that CRIS will transcribe the entire meeting and potentially if you wanted to record as well, whereas Granola will simply just provide you with sort of notes for the meetings, so it will take your raw meeting notes and make them into something more legible. So both those tools have been pretty interesting, and I found that it's made meetings a lot easier, especially like team meetings.

We'll be making big decisions and I just want to summarize it. I also find myself sometimes so focused on taking notes that I'm not processing what someone's saying in real time. So the fact that I can just give my notes over to a tool and let the tool handle the note take in and I can be present and process. What people are saying makes it a lot easier and a lot more efficient for me personally.

their thing and you know, chasing seals and whatever. And then you know, sometimes they do show part of the town or at least the coast and it just looks amazing. It just looks like, oh wow, this would be a really cool place to visit. And I don't know that I would necessarily want to go out on a boat looking for sharks, but you know, I think it would

be a blast. And I have to say that my wife, after watching some of these shows, she might be the one going, I want to go out on the boat looking for sharks.

There's a lot of if you're an outdoor person, there's a lot of things to do, lots of like diving, swimming, lots of long drives, lots of good sunset views, and I think people here are just very relaxed and laid back. I will say that took me a long time to get used to. I'm still I'm five years in Cape Town and I'm still not used to that because from Johannesburg, we're very like hustle and very serious about things hand.

Cape Town is just a vibe that's a real's that's the way I can describe it, a very child vibe. So yeah, lots of fun things to do in Cape Town. People are very friendly as well. I find South Africa to be one of like the friendliest people around. Like you'll walk in the street and you'll say hello to people and you'll smile at them. Not every city is like that in the world, but South Africans are really friendly people.

So Cape Town is I don't know if you know this, but Cape Town has a part of Cape Town and we call it Cape of Good Hope or Cape Point. It's where the Atlantic and the Indian Ocean meet, and so you can actually see a some point of the year where you can see like the different colors of the ocean and they sort of meeting at that point. It's really beautiful. But unfortunately you're not really able or

I personally don't swim in the in the ocean. Here, the Atlantic Ocean is really cold and just not for me. But if you travel to Dlbin, that's the Indian Ocean, and the Indian Ocean is just beautifully warm and really nice to swim in. So very often we'd cop over to Durbin and go swimming at the beach and things like that. And then in Johannesburg you'd mostly like do

a lot of shopping there. There's a lot of big buildings, a lot of corporates there as well, so higher earning potential, a lot of people that are just very busy doing things, a lot of good takeouts, yeah, and just a lot of shopping malls. And then we have the Kruger National Park as well. So if you want to see what we call the Big five in South Africa, which is the lion, leopard, elephant, rhino in Buffalo, you'd basically go to Kuruga National Park and you could do a safari

and see them. No, we do not have animals walking around the streets of South Africa. I know, sometimes I can ask the question, now, really that's not the case. I mean sometimes we have had a tiger on the loose once or twice, but not on a daily basis. So yeah, really good safaris as well. Lots of very cool things to.