@kick @enkiv2 @dredmorbius @freakazoid
It's more a matter of: the social problem cannot be fixed by a technical change, so we should employ a social change instead. No matter what we do on a technical level, we can't really move the needle on this.

@enkiv2 @kick @dredmorbius @freakazoid
Changing norms is harder than employing technical systems because power is not as lopsided. To change norms, you need buy-in from most participants; to change tech, you just need to be part of the small privileged group who controls commit access. This is why it's so important, though. Norms aren't set in stone but they'll only change if you can actually convince people that changing their habits is a good idea!

@enkiv2 @kick @dredmorbius @freakazoid
Most people online have had bad experiences with people weaponizing out-of-context information -- that's why technical solutions like RTBF exist. RTBF not actually working, while simultaneously pushing power into the hands of centralized corporate services, is obvious to most people too. Saying "it's impolite to dogpile on somebody without checking whether or not you've been misled first" is way less extreme.


@enkiv2 @kick @dredmorbius @freakazoid
Re: the speed at which norms can change, consider content warnings. They went from something that only a handful of folks with PhDs trying to work out experimental ways to avoid meltdowns in extreme circumstances having even heard of them to something that everybody is aware of & only jerks believe are never justified in a matter of ten years. We still argue about when they're justified but there isn't a serious contingent against using them at all.

@enkiv2 @kick @dredmorbius I'm not arguing for RTBF. I'm arguing for not making it impossible to unpublish content.

CWs are nowhere near universal and the fact that they're not proves my point quite nicely.

@enkiv2 @kick @dredmorbius There's also the fact that people deliberately exploit immutable systems to publish stuff that's damaging. For example, there's kiddie porn in the Bitcoin blockchain.

@freakazoid @enkiv2 @kick @dredmorbius
This is a fair point, though I wouldn't pick CP as a good example of infohazard. Depending on one's model, CP is contraband either because a market for it incentivizes abuse or because exposure to it incentivizes abuse. Under the former model, having it on the blockchain lowers abuse potential. Obviously a complex & emotionally charged topic (even more so than "if you burn a million dollars does the value of a dollar bill go up or down")

@enkiv2 @freakazoid @kick @dredmorbius
The risk profile of putting contraband or blackmail material on a blockchain is basically the same as the risk profile of keeping a copy on paper in a safety deposit box & periodically mailing out photocopies -- except that this latter *only* works for people with an incentive to store info indefinitely. In other words, it puts the power to select what gets remembered in the hands of whoever thinks they will want to distribute it in the far future.

@enkiv2 @freakazoid @kick @dredmorbius
Really, norm-based solutions can't work unless practically everything is immutable either. If everything is immutable then context can be retrieved in the future even if nobody thought to preserve it at the time. This functionally defangs blackmail because lies-by-omission are not backed up by layers of friction between everybody & whatever information was omitted.

@enkiv2 @kick @dredmorbius I don't see how things' not being unpublishable could defang blackmail. Blackmail will just apply to information that hasn't been published in the first place.

This goes beyond mere disagreement; this is a system I would kill to stop.

@enkiv2 @kick @dredmorbius This is the argument 4channers make against outlawing revenge porn. "Women just need to learn to stop allowing boyfriends to photograph them naked, or accept that naked pictures of them are going to be on the Internet."

No. We live in a society. You publish shit that hurts someone else, you get hurt yourself.

@freakazoid @enkiv2 @kick @dredmorbius
I'm not opposed to ramification for bad behavior. I'm trying to figure out how to encourage punishment to be equitable. Part of that is preventing motivated misrepresentation (and power asymmetry in misrepresentation). Right now would-be blackmailers choose what gets to become history, so they can spin anything as a sin.

@freakazoid @enkiv2 @kick @dredmorbius
I'm not sure, in that case, what risk profile you're talking about. Are we talking about a case where someone publishes something about themselves that they later regret? Where someone publishes something about themselves & another party takes it out of context? Or where someone publishes information about someone else without permission?

@enkiv2 @freakazoid @kick @dredmorbius
I can't think of an example of a problem that being able to unpublish only things that you yourselve have published will reliably solve, in a world where backups & blackmailers exist. (It solves the pseudo-problem of deciding that a post you've published is potentially risky and undoing it before it has actually caused a problem. I don't think that's what you're talking about, though.)

@enkiv2 @freakazoid @kick @dredmorbius
And, on the other hand, unpublishing what *other people* have published doesn't appear to be on the table. It has a lot of issues and complications, & is generally handled by lawsuits or by corporate simulations of lawsuit-style deliberation. It can be handled by admin fiat in federated systems but scaling to distributed systems means it becomes a per-post version of transitive blocking. (Cancel messages, etc.)

@enkiv2 @kick @dredmorbius My goal is to make it easy to indicate to people who don't want to publish stuff against the will of folks who are impacted by it that you'd like them to take it down.

@enkiv2 @kick @dredmorbius The archive.org situation is one: even though they will take stuff down on request you have to separately ask them and everyone else.

Yes, there will be attempts to abuse such a system, which is why it should not be legislated into place by government but built by people who want to have a robust publication system that at least makes an attempt to minimize harm.

@enkiv2 @kick @dredmorbius I think the big issue here is reachability vs discoverability. This was an issue Mark Zuckerberg did not understand when designing graph search, until Facebook employees practically revolted and told him that it was a bad idea to let people bypass permissions like friends list visibility just because it was possible to construct someone's friends list by scraping others' pages. It's also encountered when public records go online.

@freakazoid @enkiv2 @kick @dredmorbius
Absolutely! I've sort of been arguing for this. When I pushed transitive blocking over unpublishing, it's because I think the biggest issue is the flatness of addresses/access: folks outside your group, who do not share your norms, can read your messages and force replies on you.

@freakazoid @enkiv2 @kick @dredmorbius
OK. I'm fine with that, and most mature systems for static content have facilities for that (ex., IPFS has a hash blacklist for both fetching & forwarding that's basically the same as a killfile, along with mechanisms for folks to share these blacklists with each other).

@enkiv2 @kick @dredmorbius It's not a question of unpublishing what others have solved. It's about supporting the ability to ask that others unpublish things they have published. It need neither be reliable nor perfect in order to reduce harm. But it needs to exist.

@enkiv2 @kick @dredmorbius At any rate I feel that I've given conclusive proof that this needs to exist. If you remain unconvinced then there seems to be little point in my expending additional effort trying to convince you.

@freakazoid @enkiv2 @kick @dredmorbius
OK, yeah, I'm perfectly fine with this as harm reduction. I wouldn't call it 'unpublishing' because on a technical level, on a service that otherwise supported static content, it would be implemented as a blacklist of addresses (which eventually would become un-hosted as the number of nodes with a copy approached zero).

@freakazoid @enkiv2 @kick @dredmorbius
CWs are not universal, but they are near-universal in communities in which they exist at all. The risk of immutability of public material is mostly around blackmail (which only works within one's in-group or in groups where one is forced to operate) & centralized enforcement (which can't be performed against sufficiently large groups). When immutability & good norms around context coincide, outsiders are largely irrelevant -- unless it ultimately loses.

@freakazoid @enkiv2 @kick @dredmorbius i feel like making it possible to unpublish information requires criminalizing private information sharing and archival and anonymous communication; is there a less severe way?

@kragen @freakazoid @enkiv2 @kick @dredmorbius
This is sort of my point -- either we have an 'ask nicely' or we have a state-enforced 'ask nicely', & 'ask nicely' without state enforcement is a social norm thing.

@enkiv2 @kragen @freakazoid @dredmorbius For certain classes of published stuff, in Europe we already have legal ways to demand unpublication and the sky has not fallen as a result. And I think it's forcing techies in EU to (usually grudgingly) accept that technical choices are rarely free of social consequences.

@cathal @enkiv2 @freakazoid @dredmorbius I am skeptical; we will see how much longer the sky of open societies remains standing. Brexit, BoJo, the yellow jackets, Erdogan, Orban, and the Ukraine crisis do not seem like promising developments.

@cathal @enkiv2 @kragen @freakazoid @dredmorbius
RTBF has some issues, mostly because it's a mechanism where the EU deals directly with Google. It's hard to see how it would apply to somebody running a site off their home internet connection, let alone a p2p system. It's not like it doesn't do some good, but because of its structure it's limited & increases the de-facto power of the stacks.

@enkiv2 @kragen @freakazoid @dredmorbius Well, the RTBF got codified and generalised significantly by the GDPR - the right to demand the amendment of false information, to require delisting or deletion of personally identifying data that is not in the public interest, all that looks like RTBF to me.

@cathal @enkiv2 @kragen @freakazoid @dredmorbius
Fair enough. I always considered the primary result of the GDPR to be more reasonable defaults about data collection -- nobody will actually *agree* to all their traffic being vacuumed up the way it has been, given the choice, & it can't be justified as necessary, so if you want to do business in the EU you just delete logs more often and ditch the tracking pixels.

Sign in to participate in the conversation
Eldritch Café

Une instance se voulant accueillante pour les personnes queers, féministes et anarchistes ainsi que pour leurs sympathisant·e·s. Nous sommes principalement francophones, mais vous êtes les bienvenu·e·s quelle que soit votre langue.

A welcoming instance for queer, feminist and anarchist people as well as their sympathizers. We are mainly French-speaking people, but you are welcome whatever your language might be.