Last night, I hung around at the #Catalyst beer o'clock until very late, and introduced a lot of people to the idea of #scuttlebutt and #ssb (with the former being this social network, and the latter being the protocol, am I right?)

One thing I realised is that the non-repudiation of this platform, while fantastic in many ways, can also be dangerous in others. And I am wondering in what ways you people have thought about this.

If I publish a message on Twitter, I can delete it. I think people understand the implications of that: some may have seen it, there may have been screenshots even, and Twitter keeps a copy anyway, but it's unlikely it'll appear as a genuine tweet again in the public.

This is not the case on Scuttlebutt. Once you publish something here, there is no way to ever remove it again from the public. The protocol may grow the ability for messages to surmount previous messages (editing), but there is never a guarantee that the edit will have been reached as far as the original message.

In what ways do you think are users new to this platform aware of this, and what could we do to better communicate and mediate this, such that this doesn't become a show-stopper in the adoption to many (or should become a show-stopper, but people don't realise…)?

rather than focus on the negative outcomes (which is sometimes only discomfort with a new system that doesn't match our current expectations), we can also look at the positive outcomes. the hypothesis here is that for standard communication: not being able to secretly hide or modify your previous messages is a good thing! (see also %KC9l6H8...)

as with human communication, there's no takesie-backsies here. if i say or do something i regret, i can't remove it from the memories of the other humans who experienced it, i can only apologize (publish a new message).

@martin ➬ do you have examples of communication you want removed?

from a game theory perspective, having secure communication with non-repudiation means we all have verifiable knowledge of the other players' past actions, so there's a strong incentive not to "defect" (as then everyone will have verifiable knowledge that you defected). as an example, compare the culture of Scuttlebutt with the culture of 4chan, where on 4chan there is little to no knowledge of the other players' past actions, the 4chan culture is toxic af (i spent my childhood there, seriously messed me up).

my understanding is that our append-only logs will from the Scuttlebutt communication backbone, then on the side we can have alternate communication channels with different properties, like for off-the-record chats.

@dinosaur I don't think @martin ➬ was asking for delete, he was asking "how do we communicate there is no delete?"

In what ways do you think are users new to this platform aware of this, and what could we do to better communicate and mediate this, such that this doesn't become a show-stopper in the adoption to many (or should become a show-stopper, but people don't realise…)?

Often new users express surprise at this, or even find it shocking, but usually when someone asks about that, they are met with a bunch of replies from people who have not only gotten used to it, but convinced themselves it's a good thing. that is, adapted to it.

It's both a strength and a weakness. maybe the publish button should be labled "publish forever"?

Indeed, I know one cannot delete. That's the point.

@dinosaur, when you say it's like real life where you cannot take things back, but only apologise for them, then that's of course true. However, even collective memory of the most embarrassing events fades, unlike the permanent timelines we append to here. And real-life hardly has the same level of non-repudiation as is available here.

So yes, my question was about how much this feature of Scuttlebutt could impede uptake, and how we could address this, e.g. through crystal-clear communication.

@martin ➬ - I was surprised by this property at first, but now actually think it's a good thing, both from a technical perspective in the simplification in the protocol and from a social media perspective.

The only issue I have is that users need to also be able to delete content they've replicated from others (.eg. if it's illegal), which is a practical matter that will likely be added in a future release. Blocking more or less gets you there, and then the disk needs scrubbing of that feed.

Not being a FB user, I can only guess at the garbage that is there to wade through. I think if we can communicate the value of non-repudiation we can foster a more civilized and thoughtful culture.

I'm sure there are some real world examples that might help understanding this. For example, here in the US, in New England, land deeds are never deleted, they are corrected, and sometimes quit claims are used to repudiate earlier conveyances.

In a peer to peer system, where all nodes have copies of the data, it's a reality that it can't truly be deleted, regardless of what your client tells you. Having monotonic timestamps helps in that one can review a feed and see that Alice said such and such and then later retracted that statement.

Everyone owns their feed, the signing of the messages ensures the integrity, but once the feed is replicated others also own it. Can we somehow make this clear in a way that links it to reputation? After all reputation is how bad actors are managed in P2P systems. One's ability to become a highly connected influencer in the network is linked to reputation.

I've just had a discussion tonight about an overlay of Karma. Just some ideas. Obviously, the karma I attribute to you or anyone is my thing, just like spam is really a subjective term, but I do wonder if there are ways in which karma could be shared more programmatically, so that it'd be easier for me to incorporate things like "oh, four of my friends think this message is nonsense" in helping me make up my own mind about credibility.

Or is that a slippery slope?

In the age of unlimited surveillance and unlimited storage it's reasonable to anything you publish online is permanent. In this sense the ssb protocol is a clearly communicated version of this.

For those for whom anonymity is very important I would not suggest ssb. I think there is also a tension with what we mean by secure-(scuttle-butt).

RE: the analogy that this is just like not being able to "un-say" something in conversation - I am not sure I will ever use this analogy.

@martin ➬ "So yes, my question was about how much this feature of Scuttlebutt could impede uptake, and how we could address this, e.g. through crystal-clear communication."

From my own experience of utelising #ssb more it was that ssb could do something which no other tool could do. It was able to operate even when I was #offline , which was important when I moved to #fiji and had a really shakey connection for the first months. During this period I was also trying to work on an open source project ( #mmt ) with folx who were also open to using ssb as our comms platform (it didn't hurt we had a member, mix, who was super invested/embedded in ssb).

This for me has formed one route through which there will be adoption of p2p alternatives such as ssb. Do something people need to do that other things can't do.

There is a time and place for forgetting as well as remembering the OTR designers argue that even business negioations are frequently undertaken informally. Indeed, often intoxicating liquids are consumed, which serve to improve denyability.

But, you should not sign the contract while drunk! SSB is designed for sober usage. However, another protocol (that even utilized ssb identities) could be bootstrapped on the side, and optimized for drunk usage, we just havn't gotten to that part yet.

Reportedly, in ancient persia, when making a decision, they would discuss the matter while drunk, come to a decision, but tomorrow when they were sober again, consider the matter again, and if it still seemed like a good idea, the decision was finalized.

source: https://skepticalphilosopher.blogspot.co.nz/2009/09/persian-strategy-deliberating-while.html

Haha, thanks @Dominic

SSB is designed for sober usage

… means we're splicing our target market!

I think I have a 9:1 drunk:sober ssb usage. That's why I stay posting to IRC

I think we have 2 education problems @martin ➬

1. helping avoid new people's surprise and the inability to delete
2. helping all people to understand that in distributed systems the best you can ever get is a polite request to change / delete something, and that there is no guarantee they will honor it.

I think 2 is going to be a major issue when delete is added. People will think "oh i'm safe no-one will see [insert something stupid they did]" meanwhile everyone has a copy of it and many / all will probably have easy access to all versions. That's going to be a hard think to educate folks to... maybe if the major clients go out of their way to make it obvious when a message has been edited (or deleted) and make it easy to see all versions folks will see that and be prepared for it before they eventually use the functionality themselves.

as with human communication, there's no takesie-backsies here. if i say or do something i regret, i can't remove it from the memories of the other humans who experienced it, i can only apologize (publish a new message). - @dinosaur

I think that's the key. I think emphasizing "hey, you're already familiar with this behavior. it's the way the real world works" is absolutely the best way to handle this. It reframes the "problem" perfectly and puts it in a "oh, i know how that works" perspective.

--

The protocol may grow the ability for messages to surmount previous messages (editing), but there is never a guarantee that the edit will have been reached as far as the original message. - @martin ➬

just to be extra clear... even if it did reach, there's no guaranteed that every client will honor the edit / delete request.

well, interesting but I see karma as something very personal (from your persona). But let me see if I understood well...
You have an idea and before publish that you decide to review that ideia between your peers.. sounds okay for me, based on the opnion of your peer you decide or not to publish that idea.. If you publish, it will generate your karma (it can be good or bad, depends on acceptance). You can grow or not your credibility based on the results for sure.
Nothing is too SSA, I guess you have to try at least!

I think that's the key. I think emphasizing "hey, you're already familiar with this behavior. it's the way the real world works" is absolutely the best way to handle this. It reframes the "problem" perfectly and puts it in a "oh, i know how that works" perspective.

The problem I have with this, @masukomi, is that we create plenty of situations in the real world, where we do not keep a meticulous record of what people say, and when. Statements get lost in social contexts, and memories fade over time (until your parents bring out the photos on your wedding).

Not so much with #ssb…

Huh, it just occurred to me. If I take the public key my Scuttlebutt identity is based on, and publish its private key publically, then I can't delete my old posts, but I can technically claim that everyone simply decided to conspire against me and fake them, framing me for something I didn't write.

Not exactly an argument that'd hold up in court, unless I can claim my private key was leaked before everyone received the post I'm trying to "delete."

From a discussion offline I have suddenly used the term 'eternal messages' for describing ssb posts. I think that's very fitting...