I’ve wondered if the risk is about the same as a jailbroken iPhone , which with that considering you stick to legit sources and such i have never had a issue there surely it’s more important to just not install random software more than anything could say the same about installing random stuff on a Mac without oclp and installing homebrew.
Take your jailbroken iPhone and replace the network layer with uncertified 3rd-party code. That's a better analogy. Even if the 3rd-party code is from a trusted/secure source, you need to decide for yourself if the risk is worth it.
Jailbreaking (rooting) has one set of exposed vulnerabilities. Replacing the network layer after you've rooted the device? At least worth some consideration.
EDIT: Then, after you've rooted the device and replaced the network layer, make sure that all installed apps and their future upgrades don't exploit the vulnerabilites or allow other apps to exploit the vulnerabilities. If you know enough, the security concerns grow exponentially and will hurt your head to think about.
EDIT: I'm thoroughly enjoying the dialog with what are clearly competent team mates. I mean that sincerely, both 'competent' and 'team.' So far, the questions and 'challenges' have only solidified my belief in my position about security vulnerabilities in OCLP-patched Sonoma.
I am totally open to a Dev visiting and opining on why I am full of $#!+ and we have nothing to worry about. Finding out I am wrong would be the best news possible.
EDIT: My wife says no matter what the Devs say, I'm still full of $#!+... and today's a good day
===============================
EDIT: For the interested reader ... this will bore most. In a previous life, I developed NDIS drivers for Windows (up to Windows 7). My NDIS drivers were designed to operate at a "low" network layer that gave them complete access to the IP stack. Each NDIS driver had a debug switch that I could turn on and off to enable "transparent" mode where I could see all the packets traversing the stack. Since my drivers operated in "promiscuous" mode, I could see the headers of all the packets that the PC observed on the network (like a network "sniffer"). Initially, when my NDIS drivers were installed, Windows would flash a prompt asking for permission to install the driver. My team became so good with the driver development, that we found ways to install the drivers such that the Windows "malicious software" detection wasn't alerted and no messages were displayed. Our software could be installed without the user's knowledge. Fortunately, we were the good guys and we never used the capability with malicious intent. I'd like to think that everyone is one of the good guys.
