Hi All, yeah … I agree … for some time I took care of this, but I wasn’t planning on doing everything on my own … so I’m happy to provide the work on the tooling (Updating the tooling, addressing issues with it, helping with setting up things and making sure the build infrastructure works as we want it it). But I wouldn’t be willing to do the paperwork too.
Chris Von: Shane Curcuru <[email protected]> Datum: Freitag, 30. Juni 2023 um 13:12 An: [email protected] <[email protected]> Betreff: Re: [DISCUSS] Retire Apache Training? Christofer Dutz wrote on 6/30/23 6:29 AM: > Well continuing would mean that we file the incubator reports on time. > If we do that, I’m fine with continuing. If we don’t, well … guess then it’s > time to retire. Thinking forward to ASF project (i.e. after graduation) expectations, the board needs to see three PMC members who can be active enough to make reports and respond to security issues. So it sounds like you're fine to continue now, as long as Incubator reports get in. ... > Von: Mirko Kämpf <[email protected]> ... > *The Apache Training project uses automated data analysis to check (some > of) the criteria mentioned above.*Using the source code repository as a > starting point we will investigate the documentation corpus and make it > accessible for > a chat based interaction => Apply ChatGPT to the repo for contextual search. > We can generate a report from the repository which will be "reported to" or > "exposed via" Apache Training project. => We have to define in which form. > First proposal: generated ADOC files in the github repository would be fine > for an inventory. Interesting - so you're saying part of the code Training would produce is repo analysis of "what documentation/useful human content" does each ASF project have, and how can we quantify "how useful" some doc is? That could be useful, even if it's slightly different than the concept of "providing easy tools to make training courses". A few comments on this: - There are no ASF policies on specific documentation, so anything you suggest would be a best practice, not a requirement for other projects. - There are several non-ASF projects that are doing interesting work in the auto-discovery and analytics of codebases, it might be worth seeing some of their techniques (or maybe their code) in terms of how they crawl repos and the like: https://chaoss.community/ https://chaoss.github.io/grimoirelab/ https://bitergia.com/ While they focus on higher level analytics, I bet (eventually, someday) they'd also be interested in scanners for README quality and formatting and the like. Personally, my time is limited thru October, but I hope to have more volunteer time after that. I hope the podling stays around, the concept of making coder-friendly ways to easily re-use training content is important for the whole ecosystem. -- - Shane Member The Apache Software Foundation
