Thanks for confirming the situation @rbirkner. This is very useful context.
I really value the work that has gone into this proposal, and I’m also very much looking forward to decentralisation of boundary node control (awesome work!). However, I can’t adopt a proposal that unnecessarily removes the ability for voters to review and vote on what takes places within a system subnet. Given that the trust requirement is not essential, I see it as unjustified.
There’s no way for me to validate what a principal will do once they have privileges to deploy to a system subnet (there are certainly malicious things that could be done). This is problematic (after all, verifiability and absence/minimisation of trust is what this whole enterprise is about).
I’ve rejected Subnet Management proposals in the past for this sort of reason, after agreeing steps forward to improve verifiability. Here’s an example for reference.
Can I ask why this hasn’t been seen as a priority?
If I do reject this proposal, my main intention will be to escalate the need for a more robust and suitable deployment process (not necessarily to try and block this specific proposal from executing).