I’m not going to make a forecast of how successful this social network will be but I will note that rather than a screenshot on the landing page there is a JSON blob.
I also think that public/private key pairs are too tough of a concept to drop on non-technical users. It took me a while to understand them, I still don’t really grok them, and I’ve been programming for over a decade.
I think they can be a great building block, but they need a better face. Like HTTP is good for what it does, but we still slap a browser on top of it, for it to actually be used. No one (except Stallman, maybe) goes around browsing the web with curl.
a truly censorship-resistant & global social network.
While the idea of preventing censorship sounds laudable at first glance, I don’t think it’s really a net benefit to most users of a social network. The biggest differentiator between social networks to most users is the user base, who is on it. A major issue I see with these anti-censorship social networks is they’re full of people who don’t want to be censored for much more nefarious reasons than myself. You end up with a user base similar to 8chan.
I also wonder what they do with spam. Nobody thinks spammers should have an undeniable right to post uncensorable spam and could not be deplatformed for spamming. But on a technical level it’s pretty much the same thing.
I am not a fan of Nostr’s design for other reasons, but I am a fan of decentralization and censorship-resistance. I’ll try to make an approachable pitch for it. (Perhaps we need words with less baggage to describe these things… Deplatformed? Silenced? Shadow-banned? Selective restricting? Algorithmic timelines?)
Third-party developers got deplatformed off of Twitter, many times over
Non-paying users’ tweets are being silenced relative to paying users’ tweets.
Tweets that promote outrage were said to be shadow-banned (not sure if this is still true).
Not long ago there was a lobsters article of a Mastodon instance changing their policy and deplatforming at least one of their long-time users. I know of at least a few instances personally of this happening.
When Facebook disproportonally surfaces propaganda to your aging relatives, it’s relatively and selectively restricting other kinds of content
When Google shows inferior ads (often outright malware and scams!) above superior results
When Apple duplicates an app’s functionality and kicks them off the App Store
When a country filters or shuts down their internet to squash a civil rights protest
When a sex worker can’t get a bank account
I can go on and on, there is a near-infinite list of examples in modern society of people in positions of power disadvantaging people who are not, to various degrees of atrocity.
These are all things that are addressed by a protocol that is censorship-resistant and decentralized – it removes a single point of control for what the consumers are allowed to interact with.
It’s important to note that moderation is separate from a protocol being censorship-resistant: Moderation is a layer that can be built above it, which can be implemented as an opt-in (like block lists) or as part of a third-party client.
You know a social network (this is a stretch, but bear with me) that has none of these problems? Lobsters.
And yet, it’s fully centralized, has no technical capabilities to prevent users from being banned, shadow banned, censored, deplatformed, etc.
The reason it works is because content quality in a social network is not a technical problem, it’s a social one, and there’s no amount of purely technical solutions that you can throw at it that will solve it.
Lobsters works because it’s small, the community trust itself and the mods, and we create and police our own rules and social protocols.
The more I think about it, the more I feel like maybe extremely massive social networks are a bit of an unsolvable problem. There might be heuristics and technical tools to make it work in limited contexts, but on the general case, it’s just impossible.
Lobsters works because it’s small, the community trust itself and the
mods, and we create and police our own rules and social protocols.
It is also gate-kept [1] by being invite-only. Effectively, to be a member
of lobsters, you knew someone from an existing social network who was
willing to vouch for you by sending an invite. Admittedly, “knew
someone from an existing social network” is very broad, and it includes
such things as spending 5 minutes proving to someone in an IRC channel
that you have something to contribute.
The more I think about it, the more I feel like maybe extremely massive
social networks are a bit of an unsolvable problem.
I’m inclined to agree with that. I also tend to flinch when I hear the
phrase “at scale”, because humans and human communities just aren’t built
to operate that way. Very very few things should truly operate “at
scale”. At-scale is fundamentally inhumane, and if you take it to its
logical conclusion, you end up with something like the Borg from Star Trek.
[1] Gatekeeping isn’t necessarily a bad thing. It becomes a bad thing
when it leads to accumulation and siloing of wealth, power, and
knowledge, or when it is based on factors that are entirely out of one’s
control, like genetics. Lobsters is gatekept, but it isn’t a knowledge
silo. All of the content here, save private messages, is publicly
readable. It’s all very transparent and open.
Not long ago there was a lobsters article of a Mastodon instance changing their policy and deplatforming at least one of their long-time users. I know of at least a few instances personally of this happening.
I can’t link to private instances of this, but here’s a couple of publicly discussed ones:
https://twitter.com/LefterisJP/status/1593934653114785793 is the big one that IIRC was featured on lobste.rs at some point but I can’t find it now (maybe removed? or I can’t find the URL that was used) – note that he had an account on that instance for several years before the policy change (regardless of how legitimate the police is in the first place).
The reality is that the mastodon instance operator owns your identity (in the FQN sense) and your connections (and can prevent you from migrating them), and your post metadata (can’t reproduce your posts elsewhere without losing metadata). This is what makes Mastodon federated rather than decentralized.
There’s no technical reason why this needs to be true (aside from it being easier to implement centralized versions of things), we have proper decentralized solutions for all of these things (nostr only solves 1/3 of these, we can do better), but Mastodon is not one of them (yet).
I agree. For me, a good selling point would be “control your own feed using an API” or “sophisticated search capabilities”. I would like less censorship, as a concept, but in practice it doesn’t really add a lot of value to my daily experience.
Social networks succeed by membership and use, and they catch on by being discussed. How the heck do you even pronounce this word? How does a listener know how to spell that sound, in order to find it and join? These are aspects of a product, especially of a social product, that ought to be foolproof.
“tumbler” is an existing English word, and the “blr” part of Tumblr can’t really be pronounced in any other way in English. “Tumblra” doesn’t work, for example.
Nostr isn’t close to any existing word, and could as easily be pronounced “noster”, “nostra”, “nose-ter”, ““no-sitter” etc.
Tumble is a pretty normal English word, to this non-native speaker.
Nostr makes me think of Latin (Nostradamus, Nostrum), and how the (probable?) English pronunciation would be different than many other parts of the world.
If you want to create an account and you don’t have an iOS device, they recommend a web app that requires a chrome extension called Alby. Putting aside the fact that this really limits the usefulness of a web app, this chrome extension appears to be for bitcoin payments…? (it does mention Nostr further down, but what?)
The title is technically misleading. Nostr is not a social network.
Like ActivityPub, it is a protocol: a foundation upon which social
networks can be built.
I’m not going to make a forecast of how successful this social network will be but I will note that rather than a screenshot on the landing page there is a JSON blob.
I also think that public/private key pairs are too tough of a concept to drop on non-technical users. It took me a while to understand them, I still don’t really grok them, and I’ve been programming for over a decade.
I think they can be a great building block, but they need a better face. Like HTTP is good for what it does, but we still slap a browser on top of it, for it to actually be used. No one (except Stallman, maybe) goes around browsing the web with curl.
While the idea of preventing censorship sounds laudable at first glance, I don’t think it’s really a net benefit to most users of a social network. The biggest differentiator between social networks to most users is the user base, who is on it. A major issue I see with these anti-censorship social networks is they’re full of people who don’t want to be censored for much more nefarious reasons than myself. You end up with a user base similar to 8chan.
I also wonder what they do with spam. Nobody thinks spammers should have an undeniable right to post uncensorable spam and could not be deplatformed for spamming. But on a technical level it’s pretty much the same thing.
I am not a fan of Nostr’s design for other reasons, but I am a fan of decentralization and censorship-resistance. I’ll try to make an approachable pitch for it. (Perhaps we need words with less baggage to describe these things… Deplatformed? Silenced? Shadow-banned? Selective restricting? Algorithmic timelines?)
I can go on and on, there is a near-infinite list of examples in modern society of people in positions of power disadvantaging people who are not, to various degrees of atrocity.
These are all things that are addressed by a protocol that is censorship-resistant and decentralized – it removes a single point of control for what the consumers are allowed to interact with.
It’s important to note that moderation is separate from a protocol being censorship-resistant: Moderation is a layer that can be built above it, which can be implemented as an opt-in (like block lists) or as part of a third-party client.
You know a social network (this is a stretch, but bear with me) that has none of these problems? Lobsters.
And yet, it’s fully centralized, has no technical capabilities to prevent users from being banned, shadow banned, censored, deplatformed, etc.
The reason it works is because content quality in a social network is not a technical problem, it’s a social one, and there’s no amount of purely technical solutions that you can throw at it that will solve it.
Lobsters works because it’s small, the community trust itself and the mods, and we create and police our own rules and social protocols.
The more I think about it, the more I feel like maybe extremely massive social networks are a bit of an unsolvable problem. There might be heuristics and technical tools to make it work in limited contexts, but on the general case, it’s just impossible.
It is also gate-kept [1] by being invite-only. Effectively, to be a member of lobsters, you knew someone from an existing social network who was willing to vouch for you by sending an invite. Admittedly, “knew someone from an existing social network” is very broad, and it includes such things as spending 5 minutes proving to someone in an IRC channel that you have something to contribute.
I’m inclined to agree with that. I also tend to flinch when I hear the phrase “at scale”, because humans and human communities just aren’t built to operate that way. Very very few things should truly operate “at scale”. At-scale is fundamentally inhumane, and if you take it to its logical conclusion, you end up with something like the Borg from Star Trek.
[1] Gatekeeping isn’t necessarily a bad thing. It becomes a bad thing when it leads to accumulation and siloing of wealth, power, and knowledge, or when it is based on factors that are entirely out of one’s control, like genetics. Lobsters is gatekept, but it isn’t a knowledge silo. All of the content here, save private messages, is publicly readable. It’s all very transparent and open.
I wrote about a similar observation here: https://srid.ca/niche
Interesting. Could you link to them?
I can’t link to private instances of this, but here’s a couple of publicly discussed ones:
The reality is that the mastodon instance operator owns your identity (in the FQN sense) and your connections (and can prevent you from migrating them), and your post metadata (can’t reproduce your posts elsewhere without losing metadata). This is what makes Mastodon federated rather than decentralized.
There’s no technical reason why this needs to be true (aside from it being easier to implement centralized versions of things), we have proper decentralized solutions for all of these things (nostr only solves 1/3 of these, we can do better), but Mastodon is not one of them (yet).
I agree. For me, a good selling point would be “control your own feed using an API” or “sophisticated search capabilities”. I would like less censorship, as a concept, but in practice it doesn’t really add a lot of value to my daily experience.
Social networks succeed by membership and use, and they catch on by being discussed. How the heck do you even pronounce this word? How does a listener know how to spell that sound, in order to find it and join? These are aspects of a product, especially of a social product, that ought to be foolproof.
“Nostr” doesn’t strike me as more difficult to pronounce than “Tumblr”, for example.
“tumbler” is an existing English word, and the “blr” part of Tumblr can’t really be pronounced in any other way in English. “Tumblra” doesn’t work, for example.
Nostr isn’t close to any existing word, and could as easily be pronounced “noster”, “nostra”, “nose-ter”, ““no-sitter” etc.
Yep.
Tumble is a pretty normal English word, to this non-native speaker.
Nostr makes me think of Latin (Nostradamus, Nostrum), and how the (probable?) English pronunciation would be different than many other parts of the world.
I’m going for ‘nostril, without the il’.
If you want to create an account and you don’t have an iOS device, they recommend a web app that requires a chrome extension called Alby. Putting aside the fact that this really limits the usefulness of a web app, this chrome extension appears to be for bitcoin payments…? (it does mention Nostr further down, but what?)
The developer is a really really really big Bitcoin enthusiast, and his personal page is just a Bitcoin company.
The title is technically misleading. Nostr is not a social network. Like ActivityPub, it is a protocol: a foundation upon which social networks can be built.
So it’s making the same protocol-over-product mistake, like so many other open source protocols of before.