3.3 KiB
I'm just going to put a bunch of our loose thoughts to page right now cause I think that's a good idea
Threat modeling relies on, well, models, which are usually inaccessible or corrupted during a psychotic and/or paranoid episode
Our workaround has historically been "find folk whose threat models we can wholesale tab as Trusted", this is not necessarily feasible for most
like it only really works when you have folk around you with similar threat models in general. or folk who know you well enough to model your models and help you with that.
guide to having that: fuck if i know how this happened with us. looking back and looking at the folk we've been curating that with, i think a big bit of it is that you have to start from compatible first principles and have similar-enough lived experience and general knowledge
like a cis woman would be a trusted source for misogyny but not transmisogyny if that makes sense?
it's really kind of just similar to surrounding yourself with a diverse group these days with the caveat that you need an average high mental stability
the goal is to create the kind of relationships where you trust everyone else's models
issue is that trust is...weird and hard
Our thoughts on trust are essentially that you've gotta make threat models of everyone around you and tab the ones that are not only least risk but highest like. Like. Would this creature fight for me.
this-all runs into "how to model" in general which i'm not certain i have distilled down enough to have coherent thoughts on this morning
The way our (Morgans) trustnet works is that we can't read our models with any kind of precision during an episode, but we can still read their metadata, so to speak
It's like a tagging system for what you can trust given creatures for
cause also trust doesn't have to be all-covering, it just has to be absolute?
Yeah! I don't have to trust Luke to have a good model of when someone's in potential physical danger, but I trust him to have my best interests in mind, to recognize patterns of abuse, and to want to help, and I trust him on all three of those absolutely
The core of the concept of a trustnet links back into the whole 'know yourself and your enemy' thing. I know we're historically not the best at recognizing when we're just in a conversation to win, but we trust, like, Luke and Sophie to recognize those and say something about it. We know we have that weakness, so we can rely on others who've proven not to have it.
and by surrounding yourself with folk who are better at this aspect of modeling you get better at it yourself
Exactly!! and crucially, if I recognize I am having an Episode, I have fallbacks
thing is also like. just how a lot of things People call paranoia are reasonable concerns in Shitscape 2025
I've been called paranoid for using Signal instead of text messages
yeah lmao. a lot of basic opsec relies on a lack of global trust
and i guess that a way to address paranoia is building global trust. but building specific trust will serve you better 99 times out of 100
There are Bad Actors out there, sorta thing?
yeah! global trust kind of. like I get assuming the best of everyone as an idea is good but from an operational security standpoint doing that without a functioning threat modeling system with a quick response time will absolutely fuck you if you are in any way a vulnerable demographic in 2025