Community Guide

Hello everyone, I’m Juanita welcome to the Scientific Python interview! Today we have Melissa from NumPy. She is an applied mathematician and she is also a former university professor who turned into a software engineer. She has been involved with the Brazilian Python community for some time and she has focused on outreach and education. Right now she’s working at Quansight developing open source software and she is also the maintainer for NumPy so thanks for being with us Melissa, can you introduce yourself?

Thanks Juanita, so yes I am based in Brazil and I have my background in mathematics but right now I’m working on Numpy specifically focusing on documentation, community and a little bit of code. I actually… As I came from a mathematics background and an education background, working with documentation is something that is really close to my heart and I like doing that a lot just because I like teaching people and writing technical documentation seems like a good fit for me. Yeah it does sound like it. I really love writing documentation as well.

So so what drove you to contribute to open source and and what was your first open source contribution? So yeah, I had been around open source for about 10 years since I started my phd and started investigating different ways of doing open science and started investigating ways of not using um software you know using open source software to do research and around that time Python was exploding around scientific communities and so I ended up learning Python and understanding that the community was actually larger than the language in the sense that um I felt welcomed in the community and so I stuck around. However, I didn’t think I was fit for contributing for a long time, I think I had a lot of other responsibilities and work and family and being a mom and not having that much free time. I just felt like I didn’t have the necessary commitment to contribute to open source and so my first contributions were actually um already tied to my recent work at Quansight. So I started working at Quansight around two years ago and that’s when I actually started contributing.

How did you end up becoming a maintainer and and what do you think was the greatest challenge to to become that? I think because of the transition from academia to developer I call myself a junior senior. I had to relearn a bunch of stuff even though I had a lot of experience in academia and so figuring out the right processes and workflow for software development were sometimes a challenge but at the same time there’s also the challenge of people and the dynamics of the communities and understanding how people decide things in an environment that is sometimes chaotic like an open source community where you don’t necessarily have rigid structures and processes and just understanding every little piece of the project. And when you maintain a project obviously there are smaller pieces that you’re kind of comfortable with and others that you’re not, but sometimes you’re required to do code review or to guide people through a contribution and you have to figure things out as you go and so this has been super fun but at the same time it can be uh it can be very challenging.

Yeah it sounds like it. So so what do you think you enjoy the most about working in open source? I think one of the things that drove me to open source in the first place and that makes me stick around is the community. Like I said before and I really like working with people and so even though we do have that thought sometimes that developers like computers and they don’t necessarily like other people. I don’t think that’s true and I think there’s no way to have a successful open source project if you’re not able to communicate, if you’re not able to empathize and to be able to just reach other people. I think it’s it’s very much a people-oriented job or at least some parts of it are and so um I think working at NumPy has been super fun because the people are so nice and it’s been wonderful learning with them and so as I do documentation or you know code review or I have questions, they are always so open to answering questions, to mentoring, to teaching, and i think that’s a very valuable thing just having the experience of being near people um who can teach you stuff is wonderful and so I think that’s the part that I like the most.

That sounds great ! So now that you’ve talked about documentation and community um I would like you to maybe talk about the other contributions that people can make to open source without much coding experience, which I’m assuming it’s like a big questions for a lot of people.

Yeah so I think there’s of course there is a focus on code when you talk about open source projects because ultimately there’s code um at the core of it, but there’s just so much more that needs to happen for an open source project to to be successful. And so i think recently people have started to realize that one person cannot do everything and so in the past it was common to expect of maintainers, especially code maintainers, that they would also do all other sorts of things like documentation, community work, translations and graphic design and that’s how we end up with like really ugly logos and uh people who don’t know how to write technical documentation end up doing that because there’s no one else to do it. And sometimes we say that people don’t do that because they don’t want to, but that’s because we’re only looking at the people who write code and we’re not giving space for people who like writing documentation or doing community work or doing design to approach these open source projects. So I think one of the things that we’ve been working towards is improving uh not only diversity inclusion and equity inside these projects (and this is part of a recent czi grant that NumPy, SciPy, Matplotlib and pandas have just received to work on diversity inclusion and equity in our projects), but it’s also a part of diversifying the past for contribution and making sure that people from different backgrounds and from different uh life histories and that have different levels of knowledge can also contribute. So it doesn’t have to be just code it can be just organizing, community, translation, it can be technical writing, it can be design, it can be website development. There’s there.. I think we’re starting to realize in several different projects and communities that a lot of thing has to happen for a successful project to move forward and to be sustainable.

Yeah I think that that pursuit of diversity it’s really necessary in open source because like many skills are needed for for an open source project to be successful so I think it’s really great that you you’re also part of like this “chasing people” with other skills. So I also wanted to know what are the things that you find challenging about the open source community and what would you like it to see go in the future?

Yeah so I think that’s very tied to the previous question and we we want more people to be involved in open source but it’s not just all more of the same people; we also want to make sure that different groups of people especially what we call today underrepresented groups in open source or in technology or in STEM have access to these projects and can also contribute their own thoughts and knowledge and experience to these projects. And so I think there’s a real problem for example, because a lot of these open source projects are volunteer based uh it is clear that not everyone has the financial security, or the free time, or the life organization to be able to dedicate their free time to open source to volunteer and do software development for many hours a week. And so this ends up selecting a group of people that we know uh it’s it’s um a homogenous group of people, so we would like to see more people from the global south, we would like to see more people from indigenous communities, we would like to see more people of color, we would like to see more different people participating in open source; and so um I think the opportunity to invest in open source as professional development and making sure that people are actually paid for their work pay for their labor is something that has a lot of potential in terms of improving our communities and so I think the challenge of that is finding funding but also rationalizing and deciding the path forward for each community and deciding uh where people want to go. um yeah

Yeah I think diversity would be like a very good thing for for our open source community and I think those um like mentoring program programs I that you work at I think they’re going to take us there like soon so I really i’m really grateful that you’re working on that. Okay so so one last question: can you give some advice to someone that wants to contribute to open source?

I think the basic advice that everyone gives and, I don’t know if it’s correct or not, is to scratch your own itch, which means find a project that you’re interested in that will maybe help you do your work better, or that is maybe something that you’re already familiar with and that interests you that actually drives you to contribute and to offer that time to your open source project, and so find something that really sparks you know sparks joy um and this will make you feel more comfortable and this will also make you feel more motivated to contributing. On the other side i think it’s also important to find ways of communicating with people from the projects and when I was a professor I used to tell my students: “you have to have study groups, you have to get together and do things together and feel empowered as a group and this will help you move forward because if you have a question you have a colleague that might help you and you will support each other” and the same goes for open source development. If you feel like other people are in the same situation as you and they are also learning and they are also needing help, it’s usually easier not to feel that isolated. So communicating with other people who are also working in the project, both in a peer group or in a mentor relationship, I think is super important if you have that opportunity. So I know that a few projects at least have you know slack spaces, or discord spaces, or forums where you can reach out to other people and ask questions and and maybe work together and I think that’s super important; so that’s my advice.

Thank you, I think those were two great advices: so look for the project that is right for you and look for the best way of communicating with people from projects that you are working on. I think that’s all for today and thank you so much for joining us um bye bye and welcome to the Scientific Python community!

Hello everyone, I’m Juanita. Today I want to show you our experience at the SciPy conference 2022.

SciPy 2022 was the 21st annual Scientific Computing with Python conference held in Austin, Texas from July 11 to July 17, 2022. Yearly, this conference brings together attendees from industry, academia, and government to showcase their latest projects, learn from skilled users and developers, and collaborate on code development.

Let’s go!

This year the conference took place in the AT&T Hotel and Conference Center at the University of Texas campus in Austin.

The conference started with two days of tutorials that usually take place before the general conference. During these, participants get the chance to learn about a specific topic in a hands-on manner. They are usually 4 hours long and can be beginner, intermediate or advanced level. There are two sessions of simultaneous tutorials (this year 7) which means you had a lot of topics to choose from, regardless of your experience level.

The conference officially started on Tuesday with a Welcome reception hosted by Enthought, the institutional sponsor of the conference. There we got the chance to meet many of the speakers, sponsors and attendees of the conference. We had great food and made new friends!

On Wednesday we went through the registration early in the morning and received our SciPy swag!

The conference kicked off with some Welcome announcements, the keynote speaker and the SciPy Tools plenary Session, where developers from each of the tools usually give short updates on their packages. These three sessions were held in the main room every day of the conference (from Wednesday to Friday).

Most of the other talks and sessions were held simustaneously with other two sessions in two other rooms so you only could attend one of them! The best play was to keep track of the SciPy schedule in order to figure out where to go next.

This year, the talks in sessions were divided according to their topic. Some of them included Earth, Ocean, Geo, Materials & Chemistry, Physics & Astronomy, Machine Learning and Data Life Cycle. There was also a mantainers track where maintainers of Scientific Python packages presented cool stuff about their tools or updates on new releases.

During the breaks, the conference center always offered amazing food, coffee and QUESO! We got the chance to visit the sponsor tables and learn about some of the most relevant companies supporting open source. Some of them had very cool swag to add to our SciPy bag. And of course this was the time to make new friends and meet amazing people from all around the world.

There was also a Job Fair and Poster Session during Wednesday night where companies from around the world were recruiting for new positions and developers were presenting their work in posters! This was a great way to know about people’s works and learn about what the community is up to with Scientific Python. And if you were also interested, a great way to end up with some job options and very promising contacts.

The conference also had awesome Birds of a Feather sessions this year, meant for people to gather and discuss around shared interests. This is a really cool space to have conversations with other people that have similar thoughts, in order to build new ideas and create plans together around a specific topic. There were four of these sessions in total, split into two days, Thursday and Friday with at least four BOF’s happening on different topics, so lots of options to get involved in interesting conversations!

Another great space of the conference this year was the Maintainers panel where some maintainers of the Scientific Python ecosystem gathered to asnwer questions from attendees of the conference regarding different topics. This was the perfect space to directly interact with the developers of some of the Scientific Python packages and learn everything about the ecoystem. Not every day you get the chance to talk to a developer of your favorite Python library.

Evenings at SciPy were filled with all the cool stuff. Lighning talks this year were on fire! These are 5 minute talks that people register to give during the conference about any topic. Literally any topic: from funny packages like catplotlib meant to plot cat images, to talks about planning SciPy on Mars using an astrodynamics Python package called Poliastro or even a SciPy song. People don’t usually come prepared to give them before the conference, they just get ideas during it and plan these short talks on the fly and results can be really amazing and almost always very funny. This is the perfect opportunity to lose the stage fright and try something new!

Almost every day after sessions at the conference were over, there was some other activity happening that included food for sure. This year we had the Annual Reception sponsored by IBM, Data Drinks and Bites with Deepnote, Happy Hour by Mozilla, Taco Truck night and many other activities where we got to meet new friends and talk to people we have only ever interacted before on GitHub. After all, the best part of the conference is the community around it. Getting to meet the developers you have been working with for years or the maintainer of your favorite Scientific Python package is an opportunity that you can’t miss.

The last two days of the conference Saturday and Sunday (not everyone stays for these), were the sprints where people work together with project maintainers on open-source projects. This is a great chance to submit your first PR in a Scientific Python project, you will get the help of an amazing group of people who are willing to teach and share their knowledge with the community. Sprints are an informal part of the conference, so if you are not interested in coding, you are welcome to just exchange ideas or get involved in conversations. A great output of a sprint can be a lasting connections with a developer you want to start working with or a PR on your favorite package.

The conference this year had 992 attendees from 32 different countries. There were 27 tutorials, 61 talks, 12 BoFs, 4 keynotes and 5 networking sessions. SciPy 2022 was certainly a blast! At the end we got to go home with new knowledge, new ideas and new friends. Thank you so much to Enthought, the conference organizers, and the speakers for putting together this extraordinary conference. The open source Scientific Python community is amazing and the SciPy conference makes it possible to share ideas and build knowledge together.

I hope to you see you at SciPy 2023 and we are very happy for you to join the Scientific Python community.

Welcome to Scientific Python!