Transcript
Well, welcome, welcome, everybody. How's it going? Let me know if you can hear me. But, yeah, great to see everybody today. My name is Sarah. I am on the developer experience team at GitHub. All of the lovely folks who normally run the show are my colleagues, and they said today I get to try it out. So, yeah, hopefully you're having a good Friday. I know it's a TGIF sort of situation here, too. But, yeah, we have a really cool project. I've learned about this project actually at PyCon US this year and was really excited to kind of see and watch how the project is growing, and they have their 1.0 release, I think, today. Double check that. But it's super exciting. So it was a great time to kind of bring them in and learn about the project. Quick thing, though, before we get started. Also, hi, everybody. Nice to meet you, too. We have Universe coming up. So if you're not familiar with what it is, it's basically GitHub's big annual developer event. It takes place in person in San Francisco, but also online, and online tickets are free. But if you are interested and or can get to the Bay Area, you may want to check out my team is running a sweepstakes where you can actually kind of enter a raffle for tickets for the actual event. So let me ‑‑ ooh, I get to figure out how to use all these fun banners. So you can see the scrolling link down below. But that basically you can join the sweepstakes, and the winners will be chosen on October 6th, because the event is at the end of October, October 28th and 29th. I should remember this. Because I will be there. So if you're there, I will see you there. Awesome. So that, I think, covers my housekeeping for now. So let's pivot to talking about DocumentDB. I'm going to add Patty to the stage. Hi, Patty. How's it going? It's nice to see you again. It's going really well. How are you? Good, good. Somebody said in chat they were deep diving into DocumentDB today and was very happy to see that we were streaming. Oh, my gosh. Thank you so much for joining. You are prescient. So, yeah, I met Patty, like I said, at PyCon US earlier this year. Tyler, I made it to an event. Let's go. Congrats. But, yeah, learned about this really, really cool open source project and kind of Patty's been around for a lot of the process, especially as it's kind of growing a lot, you know, from something that was kind of, I mean, a well-formed project, but, like, you know, kind of growing into its community presence. So, yeah, I'll stop talking and let you introduce yourself, Patty. Yeah. Well, gosh, Sarah has just been amazing and, like, honestly, running into her at so many Python events has just been great. I'm really happy that we get to stream now. But I'm Patty Chow. I am a product manager at Microsoft, and I am also a maintainer on the DocumentDB project. So DocumentDB itself is an open source MongoDB-compatible document database. So if you're familiar with MongoDB drivers, documents, and queries, I think you kind of have, like, a really familiar starting point. But I am really excited to show you what this database is and how you can use it locally on the cloud or even across multiple clouds, which is really, really cool. Sorry, I had to manage my mute because my dog decided that now was the perfect time to start scratching and digging around in the office. Yeah, no, multi-cloud solutions for things are very, very important. So hot right now. So definitely having a framework that you can kind of build your application, your systems around that gives you options is pretty great. So I have up on the screen a link to the DocumentDB.io. So that's kind of the main paid, like, website for the project, right? Yes. And, yeah, did you have a screen you want to show? Yeah, sure. Let me share my screen. So we actually donated— well, we've been working on DocumentDB for a while, but we donated it to the— oh, let me actually do this instead. Sorry, I'm having some trouble sharing my screen. But everything—so it's the magic of being live, I'll say. It always happens like this. But let's see. Oh, and now I can actually— sorry, it's okay. I see how I can share my screen now. Yeah, Patty was telling me when we met earlier, like, basically this project had been something that folks at Microsoft had been working on internally. It basically grew to the point where it kind of seemed like it was something that would be best if the community kind of drove it from there. And as I've seen a number of different companies do, basically you can take an internal project that you're working on that you want to open source, and you can open source it, and then sometimes the larger kind of like open source nonprofits like the CNCF or the Linux Foundation, sometimes they'll kind of end up being the stewards for it, especially if it's a tool or a project that, like you said, it works for different clouds. It's not like just a Microsoft service. So it kind of makes sense then for not just a Microsoft to be the open source owner for it. So, yeah. Sorry, I was just buying you some time there. No, I appreciate it. And what is it? We got to run it. We got it to run finally. But yeah, Sarah, I do agree. Ever since we donated it to the Linux Foundation last year, I've just had the opportunity to talk to so many people in and out of Microsoft about the project and really just kind of discuss about like how we can work together to, I don't know, make an open source database available for all clouds. But if you go to the site, you'll see a little bit about how to get started. So you can either run it locally with Docker, see how you can get started, you can see our docs, or you can download different packages. Again, you can open it directly via Docker, and then you can kind of pick your own package, choose your own adventure here. As we're rolling out v1.0, we're also going to be just releasing a lot of documentation with it. So that's going to be really exciting. And we're going to be publishing some blogs and samples moving forward. But if you want to stay up to date, and I'll keep talking about it, you can feel free to join our Discord, either via the documentdb.io sites that Sarah posted or through aka.ms slash documentdb discord. So if you are looking at the project this morning and want to learn more, here you go. But moving on to the repo, you can access it through github.com forward slash documentdb forward slash documentdb. So good, you had to write it twice. I do love it when the repo and org name end up being the same. Right? It's like, gosh, there's no other way. I wouldn't know how else to say it. It's just like, oh, this is the project. But I do agree. I love it when the repo and the org name are the same. And you can kind of see just our code. And then stay tuned for our upcoming release next week. So I guess our upcoming 1.0 release next week, which is really, really exciting. But with that, we've got folks joining from India, which probably set up really late or really early. All I know is it's not a, probably not a super friendly time, but great to see you all here. Yeah, I forgot to ask earlier, but yeah, please say where you're joining from and chat. We're both in actually kind of sunny Seattle. If you want to call it that, I think sunny for in general, it's enough to be a very rainy city. So every day that's not raining, it's a sunny day, I'll say. Yeah, you don't need sun. You just need not rain. That is the qualifier. Yeah. I mean, you can actually, like, I, I realized today that I may need to get curtains to block sun here, which I've never had as an issue before. So we'll get to that. That's on my weekend project list here. Also, hey, Andrea, thanks for, you can, she's one of the normal hosts for our show. So great to, great to see everybody. Yeah. So let's dive in and take a look. So we've kind of looked at the project. You mentioned, you mentioned that this is your 1.0 release. So what, there were releases before, kind of what, what makes, what made this a 1.0? Yeah. I mean, I guess we've just been making releases to, I guess as times, sorry, releases we've been working towards making document DB production ready. So you can actually just, And we wanted our goal to be so that you can just run really anything on document DB without any issues. And I feel like 1.0 has become our pinnacle of that. So you can already, well, I guess in the next release, you'll see it and compared to So you can already, well, I guess. other releases, it'll be like, you'll have better correlation and aggregation behavior. You'll see improvements to query planning and indexes and best of all, just an easier way to install everything and get started. And I guess it will work seamlessly with the Kubernetes operator to, you know, like to run it across different clouds. So it really just is the mark of, I guess we're crossing the threshold from this is just a project that we're getting started on and like, you know, we're working towards it, but you know, you can build it, but at your own risk. And we're finally releasing something that really solidifies that experience. And we really entrust the project to be just production ready or ready to run at the production level. So it's, it's really, really cool. And I'm really excited personally to build some more, you know, just some bigger applications on top of DocumentDB. Wow. Makes sense. It's always, it feels like almost like graduation or something when you get to take something, you know, like a full X dot sort of release iteration. Yeah. I'm going to, Jamie, we've got a lot of good questions coming in, so I'm going to let you, you're going to show us, what are you going to show us here? And then we'll get to some of the questions. Yeah, I'll just show you a really quick demo on how you can get started with, with installing the image, creating a container, using Docker and just installing data, running some queries. So I'm going to show you a little bit about getting started and then using the DocumentDB for VS Code extension, which is going to be really, really fun. But yeah, I guess I'd love to answer some questions before getting started. Well, so someone was asking what we were building, which I think you, this is a demo, but what are we trying to, what's our objective? Yeah, it really is to show, I really just wanted to show a little bit about the, just part of our, sorry. So there's a whole demo around this and you can actually find it on our GitHub. So it's a github.com forward slash DocumentDB. And in the organization, you'll find a booking agent sample. And I'll walk you through how to use DocumentDB to run sort of like a, I guess, like a home booking sites just for, for travel. So, but this is just going to be running through the initial part of it. And after that, you can run through, you can actually run through vector search. You can kind of get started with rag and you can see it all with the final example of like a, like an Airbnb or booking platform. Sweet. Okay, cool. I think we've got mostly caught up. So it looks like you've got your, your Docker pull. All of the Docker container pieces are ready. Yeah. So I pulled in, I pulled my image already, just got started with it for the sake of time and convenience. But yeah, I pulled the latest image here. And all of these are in like, I honestly just followed, all directions are in the modules themselves. So if you want to run this on your own time, like everything that you need to do is available to you. But yeah, from there I pulled the container, tagged it, ran it, I guess I like to name my containers. I ran it with the username and password and yeah, just made sure that everything was good. So here you can see it, the font might be really small, but my container and my image are both running. Cool. So now we're going to use the DocumentDB for VS Code extension. Oh, there we go. So yeah, we're going to use the DocumentDB for VS Code extension to set up the container and really have kind of like more of a visual of what we're going to do. So for me, I consider myself much more of a visual learner rather than someone who likes to type on the command line. So yeah, and the process itself is really simple. So you verify the Docker setup, download the official image, which we have already done. We've saved the connection and we can confirm that everything's okay. And I don't even have to type a connection string. It's, what is it? The extension itself finds the image and what is it? You can either choose to have some sample data or not. I'm going to check no because we're going to run our, we're going to load our own data, which is going to be great. So this visual kind of like wizard walkthrough that you're going through, this is part of the DocumentDB VS Code extension, yeah? Yes. And you can find it on the marketplace. Oh. I happen to have a link for you. Oh, no. Okay. Um, I guess it's not, oh, you know what? I think we already, we already did all of this. Oh, no. So it's okay. No, no, no. This is, this is fine. I guess I got, I got a couple of steps ahead because the Docker, the container is already created. Right. So it's starting. Yeah, I guess I just did it for the sake of, like for the sake of time, I jumped ahead a couple of steps, but I guess you don't even have to run the, what is it? You don't even have to run the commands now. This is actually like a, like a fairly new release. So what I would do usually was, I would, let's go through this clipboard. So what I would do is kind of have my connection string ready and paste. But yeah, I guess I don't even have to do that anymore, which is awesome. But yeah, using a username and password, it automatically connects to a local, what is it? Yeah, it, I guess it connects to my DocumentDB extension. So for here, we're just going to create a very simple database, call it DB. And from here, from our database, if you're familiar with MongoDB or document databases, we're going to create our collection from here. So we're looking at Airbnb kind of listings. So or I guess homeless thing, so I'm just going to call it listings. And Abdur, yes, this is the container for this. We're running a Docker container locally, right? Yes. Yes, we are. So one of the nice things I learned about the project before and kind of what got me started poking around with it is it is the installation is Docker first. So like, it makes it really easy to both try things out locally, or if you want to use things with your own like container services or code spaces on GitHub, you can basically click a single button on the DocumentDB website, which we can show you maybe in a second. But basically, you can literally get started right away and run through the samples that we're looking at right now. Yeah. One other question that looks like it came up for someone familiar with both Postgres SQL and MongoDB, what would be the biggest advantage of using DocumentDB over either of them directly? That is an awesome question. What is it? MongoDB. So I guess we can start with MongoDB. DocumentDB, again, it's a database compatible with MongoDB itself. I think the biggest difference would be the licensing. So DocumentDB itself uses the MIT license, which means you can fork the repo, do whatever you want, run any application with DocumentDB, and the program's all yours. And we're not going to, I don't know, we really are just giving it to you. So that's an alternative. It's funny that Google- It's the most open source-y of licenses, MIT. Yes. So it really is. It's super permissive. We just give it to you. But I think it's interesting that you bring up Postgres, because, what is it, DocumentDB is super, like just the architecture of DocumentDB is really interesting because it's all Postgres. The bottom layer is just Postgres. So it can really work just as a Postgres compatible extension. And if you want to use it as a Postgres database, it's a little bit more complicated, but you still can. And you can run both your Postgres queries and your MongoDB queries, just all within the same database, which is really, really cool. Yeah, I've definitely worked places where different teams couldn't agree on how we're doing databasing things, and then you have connectors to so many different architectures. It's a little annoying. Cool. I've talked to different teams where, what is it, yeah, some people are big Mongo fans, some people are big Postgres fans, and DocumentDB is cool, because it can actually cater to both audiences. So what we're going to do is just import our data now. And so you can really just import from JSON, you can import it from the Mongo show itself, but we already have our data here. And let's just do the embedded data to not have to worry about, what is it, embedding the data. But after a quick little refresh, you will see all of our data available to us. So we really are seeing, what is it, the listing itself, like an array of amenities, the bathrooms, bedrooms, beds, etc. And this is just some preloaded data here, which is awesome. You can see it as a table view, you can see it as a tree, which I don't think is super convenient, but my favorite way to see this kind of data is just through a JSON view. And if I want to play around with it, I can go to, what is it, I can go to a playground, and I can run all my MongoDB queries directly on the extension. And I can, for example, yeah, I can run it, like any query that I want, I can run all the queries at once, I can run, you know, whatever, like I can, if there's a list of queries that I have, or that I want to test, whatever, I can just list them here, and run it like on just at my own convenience, which is really, really cool. It kind of feels like, I mean, since I'm a Pythonista, like using Jupyter Notebooks, like when I want to write some Python code, I often start in a Jupyter Notebook and just kind of like run little snippets or like write a function, because somehow, like that's, it seems easier, more accessible to like kind of build it up, and then I'll like copy paste it and turn it all into a script, but, you know, kind of has that similar sort of playground sort of vibes. Kyle, yeah, the, the whole session here, as well as all open source Friday episodes are recorded, and you can watch the recordings on Twitch and YouTube. Just look for the GitHub channel on both, both of those places. Afreen says AWS handles database infrastructure and operations. Yeah, a lot of different cloud companies, and we were talking earlier that DocumentDB kind of the way, since it's containerized like this, and you have connectors to tools like Kubernetes, that allows you to kind of run it in a distributed way, right? Sorry, can you repeat the question? I was, I was trying to pull together a bunch of threads here. So DocumentDB, like, that's the main open source project we're talking about here today, but it's kind of a broader open source org, and you have connectors to a bunch of, you know, Kubernetes, kind of connector or extension for it. So that, depending on how people are, since this is a multi cloud solution, you can run DocumentDB on a bunch of different cloud providers, and if you need to also distribute it on like a Kubernetes cluster, yeah? Yes, 100%. What is it? I'm not going to run over, this is just more of like an introductory, like an introductory thing. But if you are interested in using Kubernetes to run, what is it, like DocumentDB across Azure and multiple clouds, there's actually a really cool video on YouTube. It's, I think it's on aka.ms slash doc, forward slash DocDB, forwards, or no, aka.ms forward slash DocDB dash k8. And it should be. Yeah, it should be. it should be either. You sent me this link to put in here, I think. Yes. I'll get it up and in chat for folks. Yeah, it'll be in the chat. It'll be in the description, but that's totally something that you can do. Cool, we have a couple other questions, I think. But yeah, the Kubernetes operator works directly with, or I guess our Kubernetes operator works directly with DocumentDB. And you can find the operator itself in our GitHub repo. So, it says this is amazing. So, like for folks who are joining after, like this is their 1.0 release today. There's been a lot, a lot of work that has gone into both taking this from kind of like an internal corporate product to turning it into an open source and getting a 1.0 open source release. So, it is definitely a big, big celebrate day today. I mean, I don't know, it's exciting. Like I want open, I love open source. I want open source to succeed. And, you know, new releases are always, release day is great. Someone asked, is DocumentDB a fork of HorizonDB or perhaps how do those two relate? Do you know? So, DocumentDB and HorizonDB are completely different. I guess they're two different services. So, for, what is it? Just as a, for people who are not familiar with HorizonDB, it is part of Azure's PostgreSQL offerings. I think it's the newest product or release. I don't want to get it wrong, but it is like, it is one of the newest releases. But yeah, DocumentDB is, it's a completely separate product. So, what is it? It works mostly with document databases and, what is it? And MongoDB queries. But, what is, it's just really interesting that it's just Postgres in the backend. Hopefully that answers your question. Yeah, I think so. Cool. Let's see. Is there any other parts of the demo that you want to show? Otherwise, I've got some more questions for you. Okay. Well, I guess just finishing up on the demo. This is a fairly simple, I guess, just a fairly simple, what is it? JSON file that we've uploaded here. But once you start running just different workloads, different, what is it? In different collections and even different databases, you can see, what is it? You can see, monitor your indexes, create your own indexes. And once you start just running queries automatically, you can actually see how your queries are doing. And get suggestions on making them run faster. So you don't have to face that certain lag that comes with running super complicated queries. Yeah. Ooh, I like performance. I've been needing to get a lot of performance insights on my own queries, especially with Copilot. Just to, it's always nice to have that sort of assisted, like you want to see the logs, you want to kind of understand what's going on. That's kind of a necessary for debugging, but like also having something help make suggestions about how you might be able to do it better is definitely helpful. Because I use databases, I don't generally create them. So knowing how my queries can be, or however I'm interacting with them, how I'm interacting with it can be improved is pretty great. Yeah. Someone asked, oh, sorry, go ahead. I'll let you finish. No, I was going to say, I, gosh, I, even though I work in databases, whenever I run my applications, I am focused on the application level. And I just want my database to run completely, or I guess just seamlessly. But if I ever want to check in on my database, see how it's going, especially while running queries that I have used Copilot or any AI to use. And like, it's just like a really long-winded query. Like I just, it's nice to see how my database is doing, just check on it. And if there's any way to improve it, there's different solutions. So I get AI to just offer ways to simplify my queries and get the same results, but just a little faster. Hey, I think pretty relevant question here. So if someone was building AI applications, where would be the place that maybe you would suggest to kind of fit DocumentDB into that as compared to the traditional Postgres, PG vector sort of scenario? Yeah, so I guess there's, when it comes to the way that you build AI applications, there's certain situations where, you know, maybe a Postgres or a PG vector setup could work best. Let's say, I guess the way that I see is if you work primarily with a Mern stack, for example. If you're comfortable with Postgres and PG vector, then by all means, that is still a great setup. But if you use MongoDB in the backend, or if you want to go more into the NoSQL route, this is just, this is an option for you. But yeah, I think it's a simple plugin for like the Mern stack, which is MongoDB, Express, was it React and Node.js. I love when they kind of like make stacks into stackernames themselves. Yeah. But yeah, I think it's a good point, like part of the point of open sources that you have options, like just because there exists a way that you could do something doesn't mean that's the only, like nobody should ever do anything else. The number of times I have been like, hmm, I wish this, you know, thing to customize my terminal existed. I looked around, didn't find, found like 10 things that did kind of what I wanted. But then of course you go make another one, a la XKCD, make a new standard if none of the standards work for you. Cool. Yeah, Dheeraj, I put a link in chat, but we are celebrating DocumentDB's 1.0 release, which is a really cool open source database project. And yeah, I think in the interest of time, I'll ask you some more, unless there's anything else burning. No, I think, yeah, I think this was great, but yeah, so to summarize, we were looking at basically there's, on the DocumentDB org page, there's a repo that's basically like a Airbnb data sample, where you can kind of walk through and look at the same things that we were looking at just now, which we were mainly controlling from the DocumentDB extension for VS Code. So if you like VS Code or any of the other tools, VS Codium, anything that kind of can pull extensions from the VS Code marketplace, you can check it out. Cool. So let's see, I think we've kind of, folks in chat have gotten, you know, looked at my notes and cheated ahead and they asked a lot of the questions I was gonna ask you. So let's, we'll start with an important one. How can folks contribute to this project? This is a great open source project. You know, it's really got a lot of momentum with this 1.0 release. You know, what kind of skills might somebody need to contribute? Like, where are you looking for contributors? All that good stuff. I think with the 1.0 release, we really are looking for people to just try it out. Really go through just the Download DocumentDB and let's say you feel more comfortable with Postgres and PgVector. Why not try adding just the, what is it, replacing it with DocumentDB or trying out DocumentDB with that sort of application? We really are looking for people to just going in, trying it, and if they ever run into any issues or wanna ask any questions, just go to the Discord. If you are looking to contribute though, there is a great way to contribute using DevContainers. So all of your dependencies are readily available for you right on VS Code or wherever you like to go. So yeah, I think more than anything, we really are looking for people to just try it out and asking questions. So I'm really excited to answer them on Discord or if you ever wanna file an issue, you can do it right in our, what is it, in our GitHub project. Yeah, I will take any excuse and we totally didn't pre-plan plugging Codespaces because that's my favorite thing on GitHub. No. Totally not. But, which is basically just you can, if your project has kind of a Docker-like configuration already, when you go to the page on GitHub, you can basically have github.com spin up a Docker container in the background and you can just interact with it right in your browser. So great for demos. Fun fact. So I have the link up for the Discord on the screen. I'm actually in the Discord. That's low-key how I found out about the 1.0 release. So it's a great place. They do office hours. It's, I think, just a really kind of, if you're even just kind of wanna get a sense of what does it look like to participate or what does a pretty good example of an open-source community look like, it's a great place to go check out. Open-source everything. Yeah, that is this kind of stream. This is Open Source Fridays. We're here, I think, most every Friday. But yeah, cool. What would be, obviously, the projects often have roadmaps and stuff. What is something that is coming up that you're super excited? I mean, obviously, the 1.0 release is a lot. That already, let's just assume that's the answer. But also, is there anything kind of coming up that you're excited about or you think is interesting for the kind of project? Yeah, I guess with DocumentDB, one thing that I'm really excited about, actually, is the Kubernetes operator. That's something that I actually haven't, what is it, I haven't built with before, but now that it is production-ready, I'm super excited to get it to run across multiple clouds. I know we've already, I mean, as a team, we've already done some testing with AWS and GCP. We've got some awesome AWS folks, what is it, also working on it. So I'm really excited to see how we can cross-collaborate, which is sort of the beauty in open-source to make this just compatible across different clouds. Yeah, no, that's, I mean, DocumentDB is MIT-licensed. It's hosted by the Linux Foundation. Multiple companies and folks are contributing. You know, it's neat to be able to see, and I've seen a number of projects go from, oh, this was a useful tool internally, to now, actually, everyone gets to be the steward for it. So, yeah, anyway, let's see. Are there any other questions from the audience for folks that you want to ask Patti or about either DocumentDB or maybe some open-source sorts of questions more generally? What got you into open-source, actually, in the beginning, Patti? For me, what is it? I used to, what is it? I am a couple of years out of college, and back in college, my friends and I worked on some awesome projects using just, I guess it was in early 2020, like in the early 2020, so we were just learning about AI, and we used AI to grab a video and, like, I guess clip it, or, like, get the most impactful clips. Got it. Yeah, called it Clips AI, and it was, and at some point, we wanted to, what is it? We went our different ways, but, or we started working on different things, but we still wanted to make an impact in the community. So that's when I learned about open-source and really just putting it out there for people to use. So it was a really great project. I miss it a lot, and I'm really happy that it is available for everybody. And going into corporate life, And, you know, I really wanted to focus into something, into working on creating tools that were free and accessible to everybody. So whenever I saw the opportunity with Document TV, I just knew I had to jump right into it. And there's so much power in open source. So I'm just super happy to be really deep into it. And as Ryan said, just be part of open source, everything. Yeah, yeah, yeah. No, it's funny. I asked that and then Ryan literally put in chat basically the same question. So jinx indeed. One other question for you from, trying to make it show up on the screen. Where do you see DocumentDB going in the next few years? I think something that I see just in the landscape of data is, I think given the landscape and especially just, I guess, recent loss, let's say in the EU, like EU data loss are very, are a lot more restrictive just as a, just FYI. But we, well, I am really seeing this push into multi-cloud. So what is it? Making this, so I really see DocumentDB kind of being at the, at the forefront of that multi-cloud push, but also when it comes to, what is it? Standards. There is no standard for NoSQL, like there is in like ANSI SQL, for example. So if we can make that push for a NoSQL standard, it's a big and ambitious goal, but it is something that I really want to pursue in the next couple of years with the success of this project. And this v1.0 release is just, I think it's a huge step into that. Yeah, no, I am a weirdo who likes standards. Maybe it's just because I've done PM work in the past, but yeah, I think it's, it actually is something that helps the open source ecosystem grow because then it gives you kind of an interface where projects can actually then interoperate easily with, as opposed to like, you know, when I, the last time I remember building, like kind of bespoke database sorts of things for like laboratory data equipment stuff, you just had to have point-to-point connectors for every system. And it's like, if there would have been a nice standard, then we would have been able to, you know, I didn't need a bunch of different converters to go different directions to connect things. Cool. Let's see. How about, we're getting pretty close to time here. So definitely just checking to see if there's any other good questions here in chat. Let's see. Well, here's one I can answer. Someone who is a second year computer science student. What is the best way to, summarizing, what's the best way to learn GitHub and why might you want to? The best way probably is if you go to learn.github.com, I'll put a link in chat, but we have all kinds of really nice self-directed tutorials, learning paths. I'll, Patty, why might someone want to learn GitHub? Yeah, well, I mean, gosh, I feel like GitHub is the home for all things code. And if I feel like an ad right now, but what is it? I didn't set you up and we are, I work at GitHub, but that's why I'm letting you answer the question. For me, I just, GitHub is the home for all things code. To be honest, if I want to open source a project and make it publicly available for everybody or collaborating on something like public or private with friends, coworkers, collaborators, you really can use GitHub. I guess it like, I work at, I guess working at Microsoft GitHub is super easy to use. I love using like just different models in GitHub Copilot and VS Code to, I don't know. And I like, I think I can just kind of experiment with different kind of models using GitHub Copilot. And it just, it works so well. I guess it's a, gosh, I got into so many tangents, but I love the integration of the platform itself, the way that I can collaborate with friends, GitHub Copilot and how AI can be really well integrated into, I guess, the storage and sharing of repositories. And I guess how it works really seamlessly with, well, I use VS Code as my main coding platform, but how it can work really well with other platforms. Yeah. One of the reasons, like I've wanted to work at GitHub for pretty much my entire career. So I'm very excited that I now have that job. But yeah, I mean, GitHub was where I learned what open source was. It's kind of been the home of open source since the beginning, because what is, you have no open source if it's not open. And there's a way for other people to access and contribute to that code. And like I'm on the developer experience team. So my job is to literally make sure that both GitHub and all of the tools that GitHub makes are actually the same. Good and useful for developers. So I love making tools. I love finding new tools like DocumentDB to help kind of with different, you know, whatever I'm trying to build. So yeah. Thank you for joining. It's fun. And like open source is great. Like I met Patty doing open source sprints at the end of a conference. It's a great way. Need more coffee. But no, sprints, joining online. I'm gonna fix this. It's a great way to both build your portfolio for your career. Like honestly, when people ask me like, what should I learn? Or what should I get into? I just say open source. Like find a cool project. It doesn't have to be, you know, like you can contribute to VS Code. It's one of the biggest open source projects there is. That might be a little hard. So, you know, my first open source contribution was to a small Python project that allowed you to connect different laboratory equipment to like to collect data from them. Cause each different device, like your signal generator and your oscilloscope, they all had very different data models. So it was something that could basically unify those data models. And if I had a little bit more knowledge later, or more knowledge at the time, I would have been like, ah, the next best place for this data is a database. Cause I basically already have my schema set up from this. But I didn't know that at the time. What was your first open source contribution, Patty? To be honest, it was honestly just open sourcing Clips.ai. And then what is it? I worked on another project in between just before joining Microsoft that I also open sourced. But I guess when it comes to contributing to projects themselves, my first real contribution has been to DocumentDB, which I'm now a maintainer of. So I think I just dove in head first into the world of open source. And I think that's a great thing to say and for people to hear, because I know like myself, I was too scared to contribute to open source. For like, I was using it for years. I'm like, I'm not good enough. Like I have a PhD in quantum physics, but I'm not good enough to contribute to it. So I was like, I'm not gonna contribute to open source. Eventually I learned that that was wrong. But yeah, you don't have to be this good or have this many contributions or these kinds of contributions, not only to, well, obviously to contribute, but to be a maintainer. I can promise you that if there's a project you're interested in helping maintain, the maintainers are probably going to be very happy to hear that. And we're grateful more than welcome or help. Like obviously look through their contributing guide. Most repos, it's contributing.md, look through the process, kind of see if they have anything, kind of documented about how they would onboard or handle adding new maintainers. But definitely you don't have to wait before trying to kind of level up and help maintain a project. And I will add documentdb-has-a-contributing.md guide. With the one over release, we're gonna have some good first issues, but, or I guess just issues with the tag, good first issues. So if you want to get started with contributing, you can totally, what is it, take an issue, but even just trying it out. And what is it? Sharing your feedback, filing an issue, or just connecting with us on Discord. It's a contribution in and of itself. I think as a growing project, especially with such a big milestone like this, we really are looking for people to just try it out. And you don't need a PhD in quantum physics. Don't get one. You really just need to, what is it? Ask questions. Like, what is it? How does documentdb compare to MongoDB and Postgres? And just, or like the, what is it? The Postgres and pgVector question. That was awesome. And really just, I guess, trying it out, sharing your feedback, because that's super useful to make, like that could be, that could change your application and your feedback could change the way that we run this project, which would be awesome. Yeah, no, like I wish more people would emphasize a contribution can literally just be a question. Like whether that's filing an issue or sometimes projects use discussion pages on GitHub as well. Like if you feel like if filing an issue feels too formal or sometimes I also feel like that feels very accusatory, like an issue is, I am taking an issue with you. No, it's just a way for you to post something about what you're working on. Ask a question, you know, does this work on your machine? Cause it doesn't work on my machine, which is almost always what the issue I'm posting. So anyway, well, I think we're getting pretty close to time, but one last question for you. What emoji best describes you? I was thinking about this a lot. There's a lot. I really like the, like the hand emoji. That's like this. Oh, like the hand, okay, yeah. Party hands? I don't actually know what you call that one. I don't know, but it's like, it's like this. And what is it? Cause I feel like it's a- Hands in air is the tag for it. Hands in air. Yeah. But honestly, I feel like there's a, there's a lot to celebrate right now in and out of DocumentDB. So what is it? And I feel like, especially with like, I'm like TGIF? Hands in air? Yeah, it is a TGIF emoji. Or even like, I guess the, it's called bang bang, but it's the two red exclamation points. Yes. Yeah, yes. Not sure why it's called that. Not sure, but- I think it's because the exclamation point can also be called a bang operator in some programming languages. Oh. So that was probably the people who were writing the emoji standards, Hansi or whoever, that was probably what it was. And it's not called a bang, but the two exclamation points are called a bang bang. Yes, because there's two of them. Yeah. So I don't know. Somebody, somebody answer this. I will seek answers. Well, I'll look through chat afterwards for the emoji experts. Awesome. Well, thank you again for taking time out of your release day to come chat with us and share all the cool stuff you've been working on. What's coming up and yeah, just chat about how much we all like open source. Thank you so much for joining everybody and thanks for having me, Sarah. me, Sarah. Absolutely. And yeah, we'll see y'all. We've got Open Source Fridays as one of our regular shows, and you can find recordings for other episodes on YouTube and Twitch, as well as this one, probably very shortly. So we'll see y'all soon, and have a great weekend. Bye, everybody. Bye. Transcribed by https://otter.ai Transcribed by https://otter.ai
