The appropriate metric for safety of an app store is not whether you can get attacked, but rather how many users - proportional to all users - get attacked every year.
This is the second person in this thread to think using an open source repo as an example carries weight. There are less users, therefore less malware is released here in the first place and then there’s less load for the malware filters for that that does get through.
My comment stands, why target f-droid?
> But we have historically seen issues with those repos haven’t we?
Care to elaborate?
Here’s one issue we’ve had with these repos:
https://lwn.net/Articles/295406/
Then there’s things like NPM that’s broken quite a few production deployments in spectacularly public ways.
Apple's filters are clearly garbage.
Using linux as an example here doesn’t work.
You're not perfect at anything you do. Does that make you useless? Of course not.
As a gatekeeper? Yes, yes it does; hence why there shouldn't be any gatekeepers.