a thought on
#Bluesky ´s moderation problems, prompted by Turkish government requests: as I understand (?) BS, the underlying idea of both company and protocol is to build something that ‚de-bundles‘ previously unified aspects of centralised social media (I’m avoiding the ambiguous term decentralisation..). In particular, that debundling is meant to ˋsolve‘ the impossibility of content moderation by applying a bare minimum of intervention to a base corpus of posts while allowing individual users to develop tools to select subsets they wish to see. In an ideal world, that distinction corresponds to a large set of „lawful, though potentially awful“ content from which users then choose (composable moderation). Two different kinds of ˋmoderation‘ actions match the former and the latter - infrastructure takedown (purge from network) vs. labelling (hide, blur, warn). This hits the buffers of reality in two distinct ways: 1) „lawful“ varies across countries; 2) user disapproval (particularly while building out the system). Using ‚soft‘ tools for (1) leaves a US company actively hosting & facilitating content other nations view as damaging (and the „one step removed“ defense may seem disingenuous or insufficient). Conversely, user sentiment has prompted terms of service that effectively apply infrastructure takedowns (banning users) for ˋmerely‘ awful but lawful content (by US standards). Both of these undermine the position of neutrality that BS seems to want to occupy 1/
Like Us
