Sustain

Jingna Zhang

Richard Littauer | Eriol Fox

In this episode of Sustain, Richard and Eriol talk with Jingna Zhang about Cara, AI copyright, and protecting artists’ creative ownership. The discussion also examines the intersection of art and open source, the importance of licensing, and how technology can better support creative communities without exploiting their work. Press download now to hear more!

[00:02:22] Jingna recounts her legal battle after an artist copied her photograph, won an award in Luxembourg, and initially defended his use of her work.

[00:07:01] We learn why Jingna created Cara and how it takes a practical approach to generative AI, focusing on protecting creators' rights rather than making unrealistic promises to ban AI forever.

[00:13:37] Jingna compares the financial support available to open source developers with the limited resources available to independent artists, questioning how creative work can be better sustained.

[00:15:29] The discussion explores the need for global protections.

[00:17:24] Jingna explains how Cara encourages community interaction, longer comments, and user-curated feeds rather than algorithm-driven engagement. Also, she discusses the disconnect between artists, developers, and newer AI enthusiasts.

[00:22:00] Eriol and Jingna discuss coding, design, photography, and engineering as creative disciplines.

[00:24:30] Richards asks about Lantern and Jingna explains how she helped launch Lantern after Cara experienced targeted scraping attacks.

[00:27:03] They discuss open source licensing and AI restrictions.

[00:28:21] Eriol highlights how existing open source licenses often fail to address design assets, user research, and other creative contributions beyond code.

[00:30:31] Find out where to follow Jingna and learn more about Cara and Lantern.

[00:18:17] “I would love for Cara to be a corner of a better internet.”

[00:22:52] “Good programming is art and good engineering managing is art.”

[00:24:42] “The first scraper [of Cara] actually ended up apologizing.”

CURIOSS

Special Guest: Jingna Zhang.

Support Sustain

Jiayi Sun | Shurui Zhou

Richard Littauer

In this episode of Sustain, Richard Littauer is joined by University of Toronto researchers Jiayi Sun and Shurui Zhou to discuss their research on how Codes of Conduct affect open source communities after adoption. Drawing from a dataset of more than 150,000 GitHub repositories, they explore how often Codes of Conduct are updated, whether they attract or discourage contributors, and why simply adding a policy may not be enough to create a healthy community. They also discuss using AI to analyze thousands of Codes of Conduct changes and how the rise of AI-generated contributions is creating new challenges for open source maintainers. Download and listen now!

[00:01:03] Jiayi Sun and Shurui Zhou introduce their research into collaboration, sustainability, and community dynamics within open source software.

[00:03:14] The researchers built a massive dataset covering more than 150,000 GitHub repositories with Codes of Conduct.

[00:04:04] Their paper, Beyond Adoption: Examining the Evolution and Impact of Codes of Conduct on Open Source Communities, examines what happens after a Code of Conduct is adopted and whether it changes community participation.

[00:07:24] Jiayi explains how the team identified Codes of Conduct using GitHub issues, pull requests, commit messages, repository files, and GitHub's community metadata.

[00:10:53] One of the biggest findings: more than 75% of projects never update their Code of Conduct after initially adopting it.

[00:15:10] Richard asks about the tools they used to cluster and understand the objects they were looking at. The researchers used GPT-4 Turbo to help analyze roughly 8,000 Code of Conduct-related commits and categorize the types of changes projects were making.

[00:17:41] Code of Conduct adoption was associated with a short-term increase in new contributors, along with some increase in existing contributors disengaging.

[00:20:21] Long-term effects were much smaller, suggesting that a Code of Conduct can serve as an important signal, but community health requires continued structural and cultural work beyond simply having a policy.

[00:22:01] AI-assisted and agentic software development is creating a new challenge for Codes of Conduct, as communities debate how AI-generated contributions should be handled and what behavior should be allowed.

[00:26:06] Shurui explains how the research grew out of conversations about inclusion and community management, asking a fundamental question: if so many projects already have Codes of Conduct, why do maintainers still spend so much time resolving conflicts?

[00:11:27] “Over 75% of projects never touch their code of conduct after initial adoption.”

[00:23:09] “This question of how people engage with contributors has become more important with agentic software engineering and the growing volume of AI assistant contributions.”

CURIOSS

Special Guests: Jiayi Sun and Shurui Zhou.

Support Sustain

Carol Willing | Hugo Bowne-Anderson

Richard Littauer

Richard is joined by guests Carol Willing and Hugo Bowne-Anderson for a conversation about AI, open source, and what happens to communities when generating code and content becomes nearly effortless. They discuss how AI is changing the maintainer experience, the importance of human expertise and collaboration, local and community-driven models like Satyrn, and whether open source itself may shift from sharing code toward sharing workflows, skills, and ways of solving problems. Along the way, they explore the environmental and social costs of AI, the value of local models, and why community may become even more important in an agent-driven world. Hit download now to hear more!

[00:03:58] Hugo and Carol share their paths into independent consulting and why having autonomy over their work matters.

[00:08:20] The conversation shifts to how AI is creating new challenges for open source maintainers, including automated PRs, issues, and declining use of traditional documentation.

[00:10:31] Carol describes AI as another major technological shift, useful in some areas, but disruptive to human connection and maintainer enjoyment.

[00:14:41] The group discusses how LLMs tend toward average answers unless people bring strong domain expertise and critical thinking.

[00:18:23] Carol introduces Satyrn AI, a community-focused effort exploring trusted, transparent, smaller models that can run locally.

[00:23:30] AI tools are described as amplifiers: understanding how to use them effectively may become just as important as having access to them.

[00:29:59] They question whether developers will still contribute fixes upstream or simply have agents create personalized software for them. Carol argues that human collaboration still provides something agents cannot easily replace, and open source may evolve beyond its traditional focus on code.

[00:34:52] Hugo explores how AI exposes weaknesses in incentive systems, from software development and mathematics to education.

[00:37:03] Carol suggests the future may bring renewed appreciation for design, documentation, community management, creativity, and other contributions beyond code.

[00:40:52] Hugo and Carol share advice: start small, experiment, contribute to communities you use, and keep the human side of open source at the center.

[00:10:40] “[On new tech] There’s always the storming and the norming phase.”

[00:13:15] “Who’s being helped, who’s being harmed, and who’s not even at the table with these tools?”

CURIOSS

Special Guests: Carol Willing and Hugo Bowne-Anderson.

Support Sustain

Daniel Roe | Matias Capeletto

Richard Littauer

In this episode of Sustain, Richard Littauer sits down with Daniel Roe and Matias “Patak” Capeletto to talk about npmx, the community-built alternative interface for exploring the npm registry. They dive into how a small idea exploded into hundreds of contributors and thousands of pull requests, why the project intentionally puts people and community ahead of nonstop productivity, and how they’re approaching governance, inclusion, burnout, and sustainability. They also explore how npmx could make finding the right JavaScript packages easier, the challenges of funding a fast-growing open source project, and why healthy open source communities need much more than great code. Start listening now!

[00:02:22] Daniel explains what npmx is and what started in January 2026 quickly attracted dozens of pull requests, eventually growing to hundreds of contributors and a large Discord community.

[00:05:44] They discuss modern web technology, Algolia powered search, dark mode, expanded download graphs, package comparisons, and other improvements contributed by the community, and about building around people instead of productivity.

[00:09:58] Matias shares the open source project that went on vacation.

[00:14:36] Daniel and Matias talk about managing disagreements, establishing a code of conduct, separating developer and user communities, and why shared values and inclusion are essential to the project.

[00:18:43] Matias explains npmx’s connection to the Ecosystem Performance community and how the site can surface modern or better-maintained alternatives instead of relying only on package popularity.

[00:22:52] Rather than trying to support every programming ecosystem, Daniel talks about making npmx smarter within JavaScript and TypeScript.

[00:26:54] They discuss sponsorships, grants, company partnerships, conferences, and the eventual need to fund one or two developers full-time if the project is going to thrive long term.

[00:29:58] Daniel explains why open source needs more programmers.

[00:31:54] Find out where you can go online to learn more and help with projects at npmx.

[00:08:53] “It was very special for a lot of us that were a little lost with that connection and all the noise from companies telling people they’re going to be replaced.”

[00:11:37] “You are the thief of joy.”

[00:13:03] “That was a side effect of people having fun with their friends that they were meeting for the first time online.”

[00:16:56] “Everybody that shares our values is welcome, but it’s not that everyone is welcome by itself.”

[00:30:07] “I don’t think OSS is a business.”

CURIOSS

Special Guests: Daniel Roe and Matias Capaleto.

Support Sustain

Vlad-Stefan Harbuz

Richard Littauer | Eriol Fox | Abby Cabunoc Mayes

“Today we’re launching the Software Stewardship Lab.” - Vlad-Stefan

In this episode, Richard, Abby, and Eriol welcome Vlad-Stefan Harbuz - University of Edinburgh PhD student, Executive Director of the Software Stewardship Lab, and Director of the Open Source Pledge - to talk about the Open Source Pledge and the newly launched Software Stewardship Lab. They discuss how companies can better support the open source software they depend on, why it can be difficult to know where funding should go, and how the lab hopes to address issues like software security, hidden dependencies, and maintainer burnout. Press download now to hear more!

[00:02:03] Richard introduces Vlad and his background, and Vlad shares his path into open source research.

[00:03:30] Vlad explains how the Open Source Pledge works.

[00:06:17] The conversation turns to how the funding goes directly to maintainers.

[00:07:57] Vlad discusses why open source funding becomes a popularity contest.

[00:10:05] We hear about challenges for large companies joining the pledge.

[00:13:16] Vlad tells us about the Software Stewardship Lab.

[00:15:09] The conversation highlights the importance of design, user experience, copywriting, localization, data collection, and other essential work that is frequently overlooked or poorly funded.

[00:17:59] Vlad discusses potential partnerships with the Software Sustainability Institute and researchers studying open source, research software, and community health.

[00:20:58] We hear how the lab plans to fund its work and the hidden risk of binary dependencies.

[00:25:49] The lab’s work extends beyond dependency mapping to research on maintainer burnout, the effects of AI, mental health, and governance structures that encourage sustainable participation.

[00:28:48] Vlad urges employees and decision-makers to recognize that supporting open source is both a public good and a practical business investment that can prevent critical infrastructure from breaking.

[00:09:12] “These metrics are at best proxies currently.”

[00:17:47] “We need more educational resources to show people that this stuff is really important and it’s not just lines of code.”

[00:28:19] “It’s about creating a larger space that focused on more general questions in open source sustainability.”

[00:28:53] “I would like to tell people that work for large companies to tell their employers that their corporate commercial work relies more on open source software that people develop for free in their free time, the work is tremendously important for their business, and there are other organizations that are doing work to help create resources to make this infrastructure more sustainable.”

[00:29:31] “Try to convince your boss that this stuff is worth funding, and not just for the social good! It’s also in the company’s interest, so you have two very good reasons to do it.”

CURIOSS

Support Sustain

Chelsea Finnie

Richard Littauer

In this episode of Sustain, Richard talks with Chelsea Finnie, Chairperson of Python New Zealand and a Senior Network DevOps Engineer at REANNZ, about the serious behind-the-scenes work required to keep open source communities alive. Chelsea shares how they moved from Python conference volunteer to community leader, then walks through Python New Zealand’s financial crisis after discovering improper use of funds, the legal and governance challenges that followed, and how Linux Australia helped keep the organization and Kiwi PyCon alive. It’s a candid conversation about trust, accountability, burnout, community rescue, and why open source leadership is about much more than code. Press download now to hear more!

[00:01:47] Chelsea explains their work at REANNZ and their role as a senior network DevOps engineer including Python-based network automation and the PANDA initiative for advancing network device automation.

[00:03:08] Chelsea talks about contributing an open source “glue” service connecting Prometheus Alertmanager to Argus using Flask.

[00:04:21] Chelsea takes us through how they got involved with Python New Zealand, helping rescue AV production for a Kiwi PyCon event, and realizing they might be capable of helping run a conference themselves.

[00:06:58] They share the lighter side of joining the Python NZ committee, including a sea shanty lightning talk and an election promise to write a Python-based sea shanty.

[00:08:35] Richard shifts into the central issue: Python New Zealand leadership had been dealing with a major financial and governance crisis and Chelsea explains discovering the budget hole, seeking legal advice, reviewing the transactions, and what happened when Linux Australia stepped in.

[00:14:43] Chelsea reflects on Linux Australia’s role as a lifeline and why having financially stable open source organizations matters.

[00:15:46] Richard highlights the often-overlooked legal and financial responsibility that comes with serving on nonprofit or open source boards and Chelsea shares where Python NZ stands now.

[00:18:26] They tell us about the new safeguards and governance checks and notes that better accounting systems, such as Xero, would likely have exposed the financial problems earlier

[00:20:31] Chelsea shares how they stay motivated through heavy work and the value of open source leadership.

[00:21:57] Richard points out that Chelsea was not only helping fix financial and governance issues but also continuing to lead Python New Zealand and run events. Chelsea talks about the decision to run Kiwi PyCon on the “West Island.”

[00:25:08] Chelsea says one of their goals as chairperson is to identify and support the next people who will lead Python New Zealand and future events.

[00:26:14] Richard closes by encouraging listeners to get involved in their local open source communities, whether Python New Zealand or any other regional group.

[00:26:51] Richard’s spotlight is, The Book of Knights.

CURIOSS

Special Guest: Chelsea Finnie.

Support Sustain

At UN Open Source Week in New York, Richard hosts a short, on-the-fly Sustain mini-episode from IBM’s offices with Andrew Nesbitt of Ecosyste.ms, a registry-of-registries built from dependency data across 300 million repositories. Nesbitt explains how that data is being used to prioritize security scanning with new AI models, revealing that many heavily used but dormant transitive dependencies—often unseen and unaudited—are high risk; in his testing of about 3,000 critical, abandoned projects across multiple ecosystems, roughly 50% had serious vulnerabilities, including logic flaws that can chain across dependency trees. He warns that LLMs accelerate both discovery and exploitation, shrinking traditional 90-day disclosure windows to effectively zero. They urge developers to prune and update dependencies and devices promptly, and to support maintainers respectfully—financially or with broader help—amid vulnerability floods and initiatives like the “Summer of Bliss.”

00:00 UN Open Source Week

Special Guest: Andrew Nesbitt.

Support Sustain

Courtney Miller

Richard Littauer

In this episode of Sustain, Richard welcomes back Courtney Miller to unpack her PhD research on one of open source’s most overlooked problems: what happens when widely used software is abandoned. Courtney explains why abandonment is not always simple, or even always bad, but can create real risks for the developers and projects that depend on it. From npm package research and downstream impact to Abandabot, AI-assisted tooling, maintainer burnout, and responsible sunsetting, this conversation explores how the open source ecosystem can better understand, detect, and respond when the software we rely on stops being maintained. Press download now!

[00:01:28] Courtney explains the focus of her dissertation.

[00:02:34] Courtney defines abandonment.

[00:03:44] Her ecosystem-wide analysis focused on the npm JavaScript ecosystem, looking specifically at widely used packages.

[00:05:23] The first part of the dissertation involved interviews with maintainers who rely on abandoned packages and often lack tools or clear processes for responding.

[00:06:31] Courtney describes two types of abandonment: Explicit Notice Abandonment and Activity Based Abandonment.

[00:09:27] Courtney explains the third and final chapter called, Designing Abandabot.

[00:11:10] Richard raises the point that some software can be “done” and still function fine. Courtney agrees, noting that not all abandonment matters and beyond alerts remediation matters.

[00:13:22] The conversation expands into under-resourced and under-maintained projects, which can also become supply chain risks before they are fully abandoned.

[00:14:53] Richard brings up the “Whale Fall” idea and Courtney agrees and points to responsible sunsetting as an important research area.

[00:17:39] We learn about Courtney’s experience bringing AI into the dissertation, especially for building Abandabot’s prediction system.

[00:20:54] Richard asks whether AI is already making abandonment more common.

[00:24:52] Courtney talks about staying grounded in real practitioner problems as the open source and AI landscape changes quickly.

[00:26:30] Final Takeaways: Courtney argues that abandonment needs to be addressed now, especially through software composition analysis tools that can help developers understand and respond to real dependency risk.

[00:01:35] “The title of my dissertation is: “Supporting the Sustainable Use of Open Source Software.”

[00:07:10] “There is no right answer how to define abandonment.”

[00:07:26] “Explicit Notice Abandonment”- where the maintainers of a package publicly express their intent to no longer do so.”

[00:07:42] “The other type of abandonment was called “Activity Based Abandonment” -commonly used as a way of identifying abandonment in open source sustainability literature.”

[00:08:26] “Out of the widely used packages, around 15% had abandonment issues.”

[00:11:38] “Not all abandonment matters. If left pad is abandoned, who cares?”

[00:21:35] “Maybe projects never have to die. You can create a fork and maintain it on your own.”

CURIOSS

Special Guest: Courtney Miller.

Support Sustain

Jeff Young | Daniel Shown

Richard Littauer | Eriol Fox

In this episode of Sustain, Richard Littauer and Eriol Fox talk with Jeff Young from Georgia Tech and Daniel Shown from Saint Louis University (SLU) about how academic OSPOs are bringing students into open source in meaningful, sustainable ways. They discuss experiential learning, research software engineering, near-peer mentorship, student motivation, maintainer burnout, and how universities can help students build real-world skills while strengthening open source communities. Press download now!

[00:01:46] Jeff describes Georgia Tech’s OSPO focus.

[00:03:49] Daniel explains SLU’s experiential learning model.

[00:05:29] Daniel and Jeff share how many students they’re working with in their programs.

[00:06:13] Jeff talks about how students engage with open source and Daniel describes meeting students and the wide range of student experience levels.

[00:09:33] They discuss what the students bring to the sustainability of open source software and the value of seeing open source as a community effort.

[00:13:05] Richard asks how academic programs connect students with maintainers when many open source maintainers are already overwhelmed. Daniel explains how SLU uses internal tech leads as near-peer mentors and describes training student maintainers.

[00:15:26] Daniel describes training student maintainers.

[00:16:45] Jeff discusses Georgia Tech’s growing mentor model.

[00:18:55] Eriol asks the guests what their “dream programs” would be if they had more resources.

[00:23:44] Richard asks how to excite universities, funders, administrators, teachers, students, and other stakeholders about open source education. They discuss open source, AI, student excitement, grounded storytelling, and real impact.

[00:28:24] Find out where you can learn more about Georgia Tech and SLU’s OSPOs programs online.

[00:05:24] “I call the students developers. They’re not students, they’re actual developers.”

[00:10:50] “Those students may not be the greatest developers, but they still have a great perspective and insight that helps grow and diversify some of these open source projects.”

[00:13:20] “Our project is fun in that I don’t have to put pressure on upstream projects to find maintainers to support the contributions from students.”

[00:13:57] “It helps create a sense of empathy for maintainers of bigger projects that are out there.”

[00:26:39] “I had a student tell me at the end of last semester: 'More grounded, fewer aspirations.'"

[00:27:18] “You wouldn’t have stoplights without OSS.”

CURIOSS

Special Guests: Daniel Shown and Jeff Young.

Support Sustain

Alan Rubin

Richard Littauer

On this episode of Sustain, Richard Littauer sits down with computational biologist Alan Rubin to explore how open source software supports scientific research, clinical genetics, and cancer-related data infrastructure. Their conversation centers on MaveDB, a project that began as a way to organize hard-to-find variant data from research papers and has since evolved into a valuable resource for both scientists and clinicians. Along the way, they discuss infrastructure funding, research software sustainability, and why open source communities and academic researchers have a lot to learn from each other. Press download now to hear more!

[00:01:24] Alan explains his role leading a research group focused on genomics, cancer medicine, and improving patient care through genetics.

[00:02:46] We learn more about what MaveDB does.

[00:06:52] Alan details why a database was needed.

[00:08:26] Alan shares how the project grew out of collaboration, PyCon AU inspiration, Django, and Python tooling that let a small team build a practical research database.

[00:11:54] There’s a discussion on the infrastructure funding problem and Alan explains a major theme is how hard it is to fund scientific infrastructure, since most grants favor new discoveries rather than maintaining shared tools and databases.

[00:17:55] The project took a major turn when clinical geneticists began using the data to interpret patient variants, pushing the team to rethink the interface and user needs.

[00:21:13] Alan describes the new clinical-facing interface, Mave for Medicine (MaveMD), designed to help doctors evaluate specific variants for diagnosis and treatment decisions.

[00:22:02] Alan talks about managing the project through a distributed team, shared responsibilities, and a role that now centers more on direction, priorities, and community than day-to-day coding.

[00:23:36] They discuss why research software rarely attracts hobbyist contributors, even when the mission is compelling, and how scientific projects often function more like small product teams.

[00:27:44] Alan makes the case that scientists often learn more about improving their software craft at events like PyCon than at discipline-specific conferences.

[00:30:38] Alan highlights how academic software depends heavily on mature, well-documented open source tools and encourages more connection between technical communities and scientific work.

[00:34:15] Find out where you can learn more about MaveDB and Alan’s work.

[00:10:04] “We quite literally followed the Django Girls tutorial, but instead of a building a blog, we built a database for research scientists.”

[00:12:35] “Infrastructure is something everybody wants to have it exist and nobody wants to pay for.”

[00:26:08] “I have never been successful in engaging the broader open source community, despite having tried many times to contribute to this or any other scientific project.”

[00:31:01] “I think people who work in OSS should be excited about the kind of stuff that their work is enabling, even if they don’t really hear about it.”

CURIOSS

Special Guest: Alan Rubin.

Support Sustain