Careful with that axe, Eugen

Long ago I wrote an article entitled Federation without Decentralisation which covers some important problems with the Fediverse, but there are still more I didn't touch on in the article. The main view I presented on the so-called "communities" of the fediverse is that one cannot feasibly have enough time to find one. The view I will present today is that the false equivalence between servers and communities is fundamentally absurd and renders individuals into one-dimensional abstractions.

The main reason for this hypothesis is that one can be part of many communities, yet one can only be associated with one server, so one can only be associated with one community under this equivalence. This is necessarily a reductive abstraction - it signifies nothing other than one social marker. As our favourite old dead German guy said:

If, thus, I see the humanity in you, as I see the humanity in me, and see nothing but the humanity, then I take care of you the way I take care of myself, because we both signify nothing but the mathematical proposition: A = C and B = C, therefore A = B, i.e., I am nothing but a human being and you are nothing but a human being, thus I and you are the same.

As the previous article mentioned, it only appears people use servers as a measure of anything when handling people they don't like; so it is interesting to hear people defend them now. I am guilty of this myself; the server which hosts the post that I first learnt of the news from has never produced anything worth the bill for the tiny currents which represent bits sent from computer to computer in my opinion. The only reason I can provide for this is that assholes attract other assholes (pardon my French).

From my own experience, the singular social marker that the server-community provides is not sufficient at all to find "community". Quite the contrary; when in a group of "programmers", I grew to dislike most programmers, and when in a group of "queer people", I grew to dislike most queer people. We could use this evidence to throw their notion of a community on its head: a community necessarily strays from a social marker. This antithesis suggests, of course, that the community does not exist. It is a mere attempt to stuff graphs of friends into discrete groups, and such groups are reified by pathetic domain names.

Another claim is that if a federation model is not exposed, users will act in an insecure manner. But this is quite a silly claim, as scrapers have existed for centralised media and just about all of the Web for decades; it is a very bad idea to expect to not be scraped or to expect one's files to not end up in a database anyway. There is no way to expect data to disappear in any damn model - the argument that this is unique to decentralisation is completely bogus.

Hilariously, fedi weenies are also happy to fling the same argument at replicated/distributed projects such as Secure Scuttlebutt. When we know that one's data is not going to be deleted, it is merely being polite to not have a deletion feature which may or may not work. I was very impressed by the matrix.org privacy policy when they first wrote it, as it is dead honest about deletion:

We will forget your copy of your data upon your request. We will also forward your request to be forgotten onto federated homeservers. However - these homeservers are outside our span of control, so we cannot guarantee they will forget your data.

Quite frankly, the technical aptitude of the average fediverse user is sufficient for people to complain that there are too many technical discussions, but not sufficient to actually create coherent discussion. If I ever get around to "finishing" Netfarm, it will remain without a delete function, because I would prefer to be upfront with the impossibility of making it work.

There is none. I can't tell you to do anything really, I've only made destructive statements. Okay, one thing: go complain about something else.