The features are theoretical in the sense that there is no real guarantee they’d be possible after BSky corp changes their behavior and that they are in use only in the least significant way possible, for tiny and irrelevant numbers of users. But of course this is just restating the obvious again. For a network truly to be shielded against this sort of thing it should be decentralized already before.
See this for how constellation makes no difference.
In terms of how prepared atproto users are to use adversarial migration, I did a point-in-time survey a couple months ago: https://rob.leaflet.pub/3m7isflo7ls23
I suspect the situation has slightly improved since then, as some or all of the migration tools offer to set up rotation keys for you, but I haven’t had time time to do a followup.
The features are theoretical in the sense that there is no real guarantee they’d be possible after BSky corp changes their behavior and that they are in use only in the least significant way possible, for tiny and irrelevant numbers of users. But of course this is just restating the obvious again. For a network truly to be shielded against this sort of thing it should be decentralized already before.
See this for how constellation makes no difference.
@73ms @irelephant
I also wrote a more complete thing about constellation here: https://discuss.systems/@ricci/116132447206279469
@73ms @irelephant
In terms of how prepared atproto users are to use adversarial migration, I did a point-in-time survey a couple months ago: https://rob.leaflet.pub/3m7isflo7ls23
I suspect the situation has slightly improved since then, as some or all of the migration tools offer to set up rotation keys for you, but I haven’t had time time to do a followup.
@ricci@discuss.systems Thanks for the link to the explanation, that thread in general and the survey! All were quite insightful.