I want to emphasize a few things:
- (as far as I am aware) the addons are not the primary focus of dae when maintaining AnkiWeb
- “user fragmentation” is a non-issue, as there are already established AddOn-Creators sharing their work on alternative platforms (i.e. Patreon)
- malicious code is a fundamental problem with the way that addons work (when installing an add-on, is downloading code that is executed unimpeded with the same privileges as the user — the cardinal sin of cybersecurity)
- there is a disclaimer “As add-ons are programs downloaded from the internet, they are potentially malicious. You should only download add-ons you trust.“ on every addon page
- currently, there are also no checks for malicious add-on code anywhere within AnkiWeb, so trust is either a given or earned through open-sourcing the add-on code on GitHub (although one could also check the downloaded code in the add-on folder if one wanted to)
- as seen with other platforms (e.g. AnkiHub), with a large enough user base even fully private “addons” will be integrated into the main Anki/AnkiWeb ecosystem
- when linking to GitHub downloads, server costs for this project would be minimal, as would be the load on AnkiWeb (exactly 0)
@Unlucky-life and @peacemonk basically already hit all of the points I was going to make.
Nonetheless, I wouldn’t think about this as a competing alternative, but rather as an improving alternative to the current state of things.
Personal note: Being responsible for the add-on of a large community project, there is a need for better download statistics and version information to keep track of what version people are on (are they updating or not?) and how large the user base is.