@oliphant

That’s very interesting development and as an admin myself I’m also very much interested in filtering hate speech instances. A good blocklist quality test for me is the presence of 101010.pl. It’s the largest instance in Poland by number of users and has been a subject of a few fierce debates that I can only describe as tribal on the (broadly speaking) “fediblock” side. It’s classification as “extreme” has been spinned by literally a couple of Polish activists promoting their own instances and based on two posts by a marginal far-right politician who immediately faced fierce opposition from the users of the very instance and soon left by himself.[^1] They very promptly escalated this Polish-language case through their English-speaking contacts in Fediblock and this is when 101010 was originally added to the blocklist.

My recent attempts to get any rationale behind smearing the instance, which by now has already escalated to outright declaring it as “Nazi” and “endorsing skinheads”, face a wall which can be clearly seen in this[^2] thread and is increasingly Kafkaesque. Specifically, there were claims made by @Artifex and then quoted by @fediblock as “evidence”[^3]:

let me list off a few sites publicly endorsed by 101010.pl administration : skinheads.social [a long list of allegedly far-right instances]

All of the listed domains we checked are non-operational or long-abandoned one-person test-post instances, but most importantly my question how exactly the “administration” allegedly “endorsed” these websites was never answered. In my world this is classifies the claims as nothing more than a simple, unfounded smear, especially as during two years of running my own instance I have myself seen zero evidence that 101010 admin actually does endorse any far-right websites, and saw a lot of evidence to the opposite effect.

Ironically, @Artifex does realize there’s a problem with their lack of evidence for blocking that website:[^3]

an admin is complaining about #Fediblock without documentation

The above unsourced “list of endorsements” is therefore provided in lieu of evidence, without noticing the absurdity of such a circular argument.

@fediblock further seems to imply that there’s no evidence to the “endorsement” because… it has been “scrubbed” from the original website:

Most folks can check their databases and see cached things even after they’ve been scrubbed from the original site.

Which further doubles the absurdity of the argument, as the original accusation against 101010 made by the Polish activists was that they don’t moderate offensive posts sufficiently, whereas now we seem to learn that it’s the active moderation policy that prevents the ban proposers from providing any evidence.

@fediblock also claim “people were calling your instance [101010] fascist”:[^4]

You were complaining that people were calling your instance fascist, I pointed out it was not just name-calling, that’s all. Your admin can communicate out-of-band with other site admins.

Evidence of whoever is “calling it fascist” is absent, but most importantly any evidence to on what basis they would call it “fascist” is of course also absent 🤷‍♂️

My assessment therefore is that the English-speaking #Fediblock community apparently made zero effort to actually validate the original claims by the Polish activists[^1] which is easily explained by the fact that they were published as screenshots (!) in Polish (!). As result, the consensus as for the need to block 101010 is apparently based exclusively on that past decision and right now holds exclusively on their inability to admit that it was biased, lacked any due diligence and based on hearsay.

As result, what we’re dealing with right now is a textbook Kafkaesque hood court situation, where absence of evidence is presented as the evidence of guilt of its own. This situation is unacceptable for me as a supporter of transparent, rules-based and evidence-based removal of hate speech on Fediverse and I will continue to highlight how harmful these hood courts are for Fediverse as a whole.

Others may support this kind of “trust me they’re fascists” bros philosophy as it allows you to promptly ban disliked instances without much hassle, but I can assure you will quickly find yourself that the only effect it will have is turning Fediverse into yet another Facebook or Twitter where content removal decisions are taken arbitrarily and you have zero impact and zero insight into the rationale behind them.

[^1]: My summary of the original Polish-language debate with links to all original posts https://agora.echelon.pl/notice/APakwagTEPCQpkZjgu [^2]: https://agora.echelon.pl/notice/AUcCioj4nudNqVHyKG [^3]: https://artisan.chat/@Artifex/110192290187102730 [^4]: https://agora.echelon.pl/notice/AUchfV1cgyEQSGrUkC