Sounds like fun. I often encounter similar problems: 1. As a new contributor, I would like to know the history of the project, related discussions, or the RoadMap; 2. As a maintainer, I would like to have a way to quickly and easily provide contributors with information about the project.
But it seems to be just an idea at the moment? It needs to be further developed into a project that can fulfill the related needs :) Best, Shuxin Pan Yongjun Hong <yongj...@apache.org> 于2025年9月18日周四 23:31写道： > > Hello everyone, > > My name is Yongjun Hong, and I’m a software developer based in Korea. > I’m also a committer on the Seata (incubating) project. Above all, I’m a > developer who loves open source and is deeply interested in the health and > growth of the open source ecosystem. > > Lately, I’ve been thinking a lot about a problem I’m sure many of us have > felt: the slow drain of knowledge and context from our projects. > I have an idea to tackle this, tentatively called Apache Compass, and I’m > looking for the first few people to help me shape it. > > Once we have a core team and a clearer vision, the ultimate goal is to > propose this formally to the Apache Incubator and grow it with the > community's guidance 🚀 > > Here’s the problem I’m hoping to solve : So many of our projects run on the > passion of just a handful of key maintainers, right? The real problem is > that so much of the *"why" *the history behind a big decision, the design > philosophy lives only in their heads. When they (inevitably) have less time > or step back, that crucial knowledge is often gone for good. It creates an > information imbalance, makes it incredibly hard for new people to step up, > and puts the long-term health of our projects at risk. > > The core idea: The heart of my idea is an AI assistant I'm calling the > "Maintainer's Co-Pilot" (MCP). But instead of trying to automatically > scrape decades of data (which sounds impossibly hard), my proposed approach > is to start by *teaching* it. The idea is that we, as maintainers, would > feed it the important stuff: PMC meeting minutes, design documents, and > architectural decision records (ADRs). (This is getting easier now that > tools like Google Meet can automatically generate meeting transcripts). The > process would be like curating the project's 'brain'. > > With that foundation, imagine if we could: > - Have the MCP draft a new ADR after being fed the notes from a few > different design meetings. > - Let a new contributor ask, "Hey, what's the history behind the new auth > module?" and get a real, sourced answer like, "Based on the PMC minutes > from last September, the team chose option A because..." > - Connect the dots between scattered discussions on a single topic that > happened months apart. > > Why start this at Apache? I honestly think the ASF is the only place this > could work. This isn't just another tool; it's a strategic asset to help us > preserve our own collective intelligence. The wealth of knowledge in our > project archives is the perfect ground to build and prove this model. > > And way down the road, I dream of the framework we build here being adopted > by companies to manage their own critical internal knowledge, all built on > a foundation proven at the ASF. > > This is just the start of a conversation. I truly believe an idea like this > could help reduce maintainer burnout and make our entire ecosystem > stronger. Long term, I also believe that Apache Compass could be valuable > even in companies that have been running for many years, helping them > preserve and manage critical knowledge over time. 👍🏻 > > Just to be clear, the features I mentioned above are really just examples > to get the ball rolling. The very first step for anyone who joins would be > to figure out what we should actually build together. So at this stage, I’m > much more interested in feedback on the general vision. Is this a problem > worth solving? than on the nitty-gritty details of any single feature. > > I’d love to hear what you all think, the good, the bad, and the crazy > ideas. 😀😀 Most of all, I'm hoping to find a few mentors and collaborators > who feel this is a problem worth solving with me. > If you’re interested, you can reply to the mailing list or reach out to me > directly via email. > > Thanks for your time, > Yongjun Hong / YongGoose <https://github.com/YongGoose> --------------------------------------------------------------------- To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org For additional commands, e-mail: general-h...@incubator.apache.org
