On Mon, Sep 27, 2021 at 09:35:23AM -0400, Stephen John Smoogen wrote: > In the end, we have never been able to keep a pool of people > interested in making Java work. We aren't the only ones as this > problem occurs in Debian also (and it occurs in other languages like > Ruby, JS, Python also). The people who want to volunteer time on the > language are more likely to work in an area where others interested in > the language are already there. Dealing with the vaguries of 40 > different slightly different disk layouts, file permissions, and other > tools from every distribution are grit in the shoe when you already > have a 'working layout' and have people who speak your language and > problems elsewhere. Yes, these observations are really key to sustainability problems with language packaging. For anything that is written in older vintage languages like C/C++ the value of the distro packaging is fairly unequivocal. There is little-to-no support for package / dependancy management that comes with the language toolchain as standard. Without the distro packages, you are left doing all the hard work yourself to handle building and install of native libraries you depend on. Both app developers and users have self-interest in volunteering time as a distro maintainers which benefits all parties. For Java / Python / Go / Rust / etc, there are increasingly decent package / dependancy management facilities that come as standard with the language toolchain. Users turn to these standard tools as their first choice, rather than the distros tools. So from the POV of both application developers & users, distros are no longer a critically important cog for helping users get the software built + installed. All they really need from the distro is the basic language toolchain. In fact distros often just get in the way by sucking in bug reports that then don't get seen by upstreams, or by letting users run with un-tested combinations of dependancies. It is easier to tell users "run go build and let it download the dependancies" (replace with pip install, or whatever other language tools fit), than to explain to people how to install the distro packaged go code for dependancies, across all the different Linux distros and non-Linux OS plaforms. Thus we don't attract as many people into the packaging world and have to rely on a smaller groups of heroic people to keep language packages actively maintined. We're also getting squeezed by the increasing number of upstreams directly building & shipping their deliverables to users, as containers (whether docker/podman containers, or flatpaks), built and tested in a controlled environment with CI systems integrated into the git forges they use. There are still valuable things done by distros of course, but it is getting harder to demonstrate this value, in order to attract a sufficient level of interest in package maint work :-( Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
