Actually, thanks for the reminder about history -- I should check that hit understands my directory restructuring as moves rather than delete-and-recreate. (It *should* be smart enough to recognize that automagically, but doesn't always do so.)
-- /_ Joe Kesselman (he/him/his) -/ _) My Alexa skill for New Music/New Sounds fans: / https://www.amazon.com/dp/B09WJ3H657/ () Plaintext Ribbon Campaign /\ Stamp out HTML mail! ________________________________ From: Joseph Kesselman <kesh...@alum.mit.edu> Sent: Wednesday, September 27, 2023 1:59:28 AM To: dev@xalan.apache.org <dev@xalan.apache.org> Subject: Re: Maven build approaching usable... Thanks for flagging that, I'll try to stay in thread. Re reviewing/merging: "Please be patient, we are making mistakes as efficiently as possible." There's a lot of backlog, not well prioritized, plus demand that we expedite the Maven cutover. I'm just getting back into Xalan after being away from it for about 15 years. I'm frankly surprised I remember it as well as I do. It's going to take some time to find the best points of attack, and unlike Xalan's early days I'm not being paid to do this, not do I have the team I was part of. I'm mostly going to have to work on one thing at a time. I'm open to input about priorities, but it can't all be done first. Re pull request: My practice has been not to issue PRs until I believe the code is, in fact, a candidate for merge. My fork should be publicly visible, and you can look at the changes reported there about as easily as you could a PR. I'd certainly welcome comments and suggestions. Continuous integration testing is a *wonderful* thing and we ought to have it. Again, priorities and available cycles.... Other open PRs: again, priorities. I agree they should be looked at as soon as possible, but not sooner. Re the 3.0 stuff: Volunteers work on what volunteers want to work on. The partial 3.0 support isn't where I would have suggested someone start, but it isn't doing any harm and might eventually be useful. Again, priorities and cycles .. and we don't always have consensus on priorities. I'm not in a position to drive that yet, if I ever will be. Basically, if the 3.0 stuff isn't useful to you, if ore it and help us understand what the worst blockers are. Latest release on Central... if you mean Maven Central, I agree that we need to keep that up to date. We've already agreed that delete-and-recreate was a mistake by someone new to the project. I believe there was a plan to fix that, but I'm still relearning some of the tools and I don't think I'm the best person to apply that fix. I hope someone is actually working on it but am not certain. Maintainers/Committers: We have one more than a few months ago. Development: Is happening, it just isn't the development you (or I).would have focused on first. Adding a person should help, but ... cycles and priorities. And cycles. And priorities. There's only so much a part-time volunteer who is still swapping this code back into wetware can do per day. -- /_ Joe Kesselman (he/him/his) -/ _) My Alexa skill for New Music/New Sounds fans: / https://www.amazon.com/dp/B09WJ3H657/ () Plaintext Ribbon Campaign /\ Stamp out HTML mail! ________________________________ From: Vladimir Sitnikov <sitnikov.vladi...@gmail.com> Sent: Wednesday, September 27, 2023 12:55:22 AM To: dev@xalan.apache.org <dev@xalan.apache.org> Subject: Re: Maven build approaching usable... Joseph, thank you for working on xalan. Would you please reply to the same thead to keep related messages together? Currently you send every update as a brand new message which makes it hard to follow. --- Have you considered reviewing and merging https://github.com/apache/xalan-java/pull/9 so CI can prove that code builds on more than one machine? All these "Tried a build under windows" would much better be a CI running with different platforms and tests running with different Java versions. Have you considered creating a PR so everybody could see the suggested changes? Have you considered reviewing and merging the other open PRs? Have you considered inviting more committers? --- I maintain Apache JMeter, and it depends on Xalan. However, the more I follow the recent activity in Xalan, the more I reassure JMeter should stop using Xalan. There is some activity, however the team does not listen users, the team is not accepting contributions, and the development is literally halted. Who cares of a 3.0 function added on a 3.0 branch when the latest release on Central is broken and it breaks builds for everybody worldwide? That is a sign of non-maintenace to me. The main branch deletes all the code, and replaces it. It should not be like that. The goal of Apache Software Foundation is to produce **source** code, and such "delete all files, replace all files" commits make it hard to follow the **source** history and analyze issues. No maintenance && no development && no new committers == project is dead. Vladimir
