The amount of duplication of effort also leads to many packages being lower quality than they could be. I remember when I did package maintenance for the AUR, I didn’t have much motivation to maintain a package that usually had a handful of users. But make that available on flatpak and I can justify spending more time because it benefits far more people.
I use AUR very rarely as it precisely does not have the trusted distribution maintainer part that the above article talks about (anyone can upload scripts to AUR), but isn’t the usual way to make a AUR script and then propose that to the Arch maintainers for inclusion in the main repositories at some point? That would surely result in much more potential users then a flatpack ever could (not that I consider number of users a very useful metric).
The amount of duplication of effort also leads to many packages being lower quality than they could be. I remember when I did package maintenance for the AUR, I didn’t have much motivation to maintain a package that usually had a handful of users. But make that available on flatpak and I can justify spending more time because it benefits far more people.
I use AUR very rarely as it precisely does not have the trusted distribution maintainer part that the above article talks about (anyone can upload scripts to AUR), but isn’t the usual way to make a AUR script and then propose that to the Arch maintainers for inclusion in the main repositories at some point? That would surely result in much more potential users then a flatpack ever could (not that I consider number of users a very useful metric).