text stringlengths 0 2.68k |
|---|
**Adam Stacoviak:** Well, Elecia, thank you for your curiosity, too. I know that sometimes you're going alone, right? Sometimes you're just in the dark. But now you're not, obviously... You're book-solid with your business, you've written books, you've come on this podcast, and you're doing cool stuff, so thank you for... |
**Elecia White:** Thank you for having me. Your podcast is a lot of fun. |
**Adam Stacoviak:** Well, thank you. |
• Introduction to Ecosyste.ms and its relation to Andrew's previous projects |
• Andrew's background in open source metadata and sustainability |
• The development of Libraries.io and its focus on package manager metadata |
• The challenges of maintaining Libraries.io as a side project and the decision to sell to Tidelift |
• The issues with Tidelift's culture and the outcome of the sale to Sonar |
• The history and context of Ecosyste.ms and its goals |
• Development of Ecosyste.ms as a rewritten version of Libraries.io |
• Lessons learned from Libraries.io, including its scalability and performance issues |
• Ecosyste.ms as a modular, scalable platform for collecting and combining open source metadata |
• Reuse of code from Libraries.io, including the dependency parser and package manager mappings |
• Usage of Ecosyste.ms in research, including studying package manager behaviors and security issues |
• Potential for Ecosyste.ms to track how CLIs install themselves and their dependencies |
• Indexing of public Docker images and running SBOM scanners on them |
• The Debian popularity contest is a good proxy for download stats, but it's opt-in and not comprehensive. |
• 0.01% of packages make up 80% of usage in package ecosystems, with 10,000-15,000 packages being the most used. |
• 15,000 people are maintaining the critical level of open source usage, with 1 maintainer per package on average. |
• 25-50% of top critical packages have some way of receiving automated donations or funding. |
• Some packages use GitHub Sponsors or Open Collective as a way to sell digital goods or services, rather than donations. |
• There is a significant difference in the number of individual vs. company sponsors on GitHub Sponsors, with individuals sponsoring other individuals at a much higher rate. |
• Open Collective is used more for large company sponsorships of projects, rather than individual donations. |
• The speaker discusses the difficulty of supporting open-source projects financially |
• The impact of global economic changes, particularly the COVID-19 pandemic, on open-source funding |
• The use of Open Collective for transparent financial tracking |
• The scale of the project, with 12 million packages, 287 million repositories, and 24.5 billion dependencies |
• The technical infrastructure, including Postgres databases, dedicated machines, and Dokku as an open-source alternative to Heroku |
• The data management and caching system, including Cloudflare and aggressive caching |
• The team and contributors involved in the project, including part-time staff and volunteer contributors |
• The vision for the project, including automating analysis and publishing of open-source data to facilitate research |
• Ecosyste.ms project funding and sustainability |
• Grant from Schmidt Sciences to support initial work |
• Support from Open Collective and customers for data access and analysis |
• Recent grant from Alpha Omega to make Ecosyste.ms long-term sustainable |
• Plans to implement new features and improve onboarding for customers |
• Revenue share model with maintainers of command line tools |
• Potential income sources, including relicensing of data and customer fees |
• Balance between sharing data and paying for its maintenance |
• Efforts to standardize package metadata and support different ecosystems |
• Use of SBOMs and integration with GitHub Actions |
• Discussion of package manager ecosystems and their quirks |
• Notably hard-to-work-with ecosystems, including R, Maven, and npm |
• Legacy issues with package metadata and versioning |
• R package manager's unusual behavior and lack of API |
• The challenge of maintaining and funding open-source software |
• The potential for connecting papers and citations to software usage |
• The "unsolved social problem" of open-source contribution and maintenance |
• Software Heritage Project: a massive index of open-source files to help solve dependency management issues |
• Importance of lock files in package managers for reproducibility and maintainability |
• Using AI to scrape APIs and extract data from large datasets |
• Challenges with MCP adapters and security implications of prompt injection |
• Rate limits and polling in API usage, and strategies for managing enthusiastic users |
• Potential solutions for large-scale data queries, including a read-only column store database |
• The time requirements for research can be flexible, with some users accepting data that is a day or a week old. |
• Andrew Nesbitt's primary user is himself, allowing him to understand the needs of others. |
• The platform's APIs are open and useable, with each one having its own open API YAML spec. |
• Users are building tools on top of the platform, such as Parlay, which enriches SBOMs using Ecosyste.ms. |
• SBOM enrichment involves taking raw data and adding additional information to it. |
• The platform is being used to support multi-ecosystem use cases, where a single tool is needed to manage different types of software dependencies. |
• The substrate of the platform is being used to build various applications, including the Funds app and the Dashboards app. |
• Future development priorities include building a search engine and tools to help maintainers understand who is using their software. |
• The platform's user base includes a mix of humans and automated systems, such as Docker pulls and CI builds. |
• Building a CI that tests changes against popular downstream users to ensure compatibility |
• Providing maintainers with data-driven insights to be proactive about changes and collaborate with users |
• Using Dependabot data to track PRs and understand user behavior |
• Empowering maintainers to improve their process and make better open source software |
• Exposing data to maintainers through a user interface, such as showing top dependents and version usage |
• Using AI to help with large amounts of users and automate tasks, such as upgrading users to the latest version |
• Tracking users through natural usage patterns, rather than invasive telemetry |
• Using project-level data, rather than individual user data, to avoid complexity and maintain volunteer-driven projects |
• Avoiding automated pull requests and instead using data to inform proactive decisions |
• Information black holes for features and tracking in open source |
• Challenges of collecting and processing large amounts of data, including structured and unstructured text |
• Use of Large Language Models (LLMs) and their potential for open source discovery |
• Ad-supported tools like AMPCode for accessing LLM functionality |
• Open source taxonomy and categorization of projects based on facets like technology, user role, and domain |
• Alignment and standardization in open source discovery and community development |
• Creating a taxonomy for open-source software in research space |
• Challenges in discovering and navigating open-source projects |
• Importance of categorization and structure in open-source projects |
• Exploring new ecosystems and languages through the help of LLMs |
• Identifying gaps in specific spaces and opportunities for improvement |
• Collaboration and input from experts needed to expand and improve the taxonomy |
• Using data to have a positive impact on the open-source world |
**Jerod Santo:** Today, we're joined by, for us, an old friend, but a long time no talk... Andrew Nesbitt is here with us. And you know, Andrew, I came across Ecosystems, which is ecosyste.ms - nice domain hack; hard to say out loud, but it looks cool in the URL bar... |
**Adam Stacoviak:** It does look cool. |
**Jerod Santo:** I came across this and I thought "This is a very cool project. It seems somewhat familiar. I can't quite put my finger on what it could possibly be... And then I saw it was from you, and I'm like "Oh, it makes total sense." This is right up your alley. We've had you on the show many times back in the d... |
**Andrew Nesbitt:** Thanks for having me. Yeah, it's great to be back. |
**Jerod Santo:** Ecosyste.ms. I mean, okay, we have a lot of context that maybe our listeners don't share, but take us back to what you're interested in, which - it seems like you've been interested in similar things for a long time. And you built libraries.io around this, and Ecosyste.ms is a very similar thing... I'm... |
**Andrew Nesbitt:** Yeah, okay. So I have been swirling around the world of open source metadata for - it must be coming up to 10 years now. Starting with 24 Pull Requests. |
**Jerod Santo:** That's right, 24 Pull Requests, yeah. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.