4 ms·

Sounds similar to what Jack Dorsey has in mind with "web5" (lol). ''' Successful social networks are usually built around a new communication primitive. Facebo

Sounds similar to what Jack Dorsey has in mind with "web5" (lol). ''' Successful social networks are usually built around a new communication primitive. Facebook had the wall, Twitter had the 140-character tweet and Snapchat had the ephemeral message. The idea maze of things you can do with decentralized identities, blockchains and zk-proofs is large, weird and interesting enough that there are probably many primitives waiting to be discovered. Decentralized social networks should explore this as a way to attract users. Offering a product experience that doesn’t yet exist is far more compelling than a clone of an existing network. ''' This is the quote. For the decentralized social network, it needs to be 10x better than existing social networks. I have a few ideas, currently playing around in this space. I think a decentralized social network can at minimum bring 10x the value to users/creators that existing social networks do. As a quick example, look up how much Tiktok/youtube make per view. Tiktok is something like 2 cents per thousand views. Per thousand views. I also think a decentralized social network can be much more profitable than a centralized network.

bckr 4y ago
<bad-idea> Maybe what will evolve the space is automatic translation from any possible format to any other format using AI. You don't need to manage a server or store anything in any particular way in any particular place. Instead, you expose whatever you want in any way you want, whether that's uploading a video to some big tech host, or exposing a bunch of XML over FTP, or a RESTful endpoint that serves binary encoding pictures of cats. My user agent scoops up everything and presents it to me in a way that I like. There would need to be some creative ideas involved for, say, "replying" (hashes? webmentions?) In other words, maybe innovation on the demand side is where the fun is. </bad-idea>
- Jonovono 4y ago
  Lets just say, I don't think it's a bad idea ;p '''My user agent scoops up everything and presents it to me in a way that I like.''' I think this is close to what the future looks like. For example, it rattles me that no matter what app I am watching 'videos' on I have different options. Some I can watch 2x speed. Some I can tap the sides to go quicker through. Some if I tap it mutes. Some if I tap it stops the video. All of this for the same type of content. Insanity. Not to mention even within certain apps like Instagram is just brutal. For the same content type - videos - they have like 3 (?) different ways to view it.
  - godshatter 4y ago
    I wish the trend wasn't for social network companies to rewrite web apps from scratch in javascript when they could simply in this case just hand me the video and my preferred local application would play it. I get that they can display suggestions for what you should watch next and keep data on how much you watch or how often you pause it or whatever. But if we're going to create a decentralized social network, then maybe we could junk all the metrics that a centralized social network keeps on their users. Let it evolve organically, rather than in a way that best suits one particular small group of people, and give people back their privacy as well as their autonomy. Edit: rewrote an awkward sentence.
    - Jonovono 4y ago
      Exactly, one way I am thinking about it right now is do away with the idea of "apps" entirely. Think about it more in the sense of "intents". If anyone is interested in building in this space, my email is in my profile! Looking for cofounders :)
- grumbel 4y ago
  > Maybe what will evolve the space is automatic translation from any possible format to any other format using AI. Easier said than done. Facebook, Google, Twitter and Co. like to gatekeep their data and don't allow third party clients access. So you don't just need an AI smart enough to capture the data and reshape it, but also one smart enough to pass whatever CAPTCHA they throw at you, which will likely get progressively more difficult the more popular your tool gets. And of course they'll lock your account when you try, so this is at best a read-only participation in the social network. The problem has shifted over the last decade, it's no longer about fixing a broken Web, but about a large part of the content no longer even being a part of the Web. Everything is either moving to Apps or getting gated behind login screens and ToS.
  - bckr 4y ago
    CAPTCHA is a big problem but not impossible to overcome. It's an arms race. That said, CAPTCHA can be defeated by changing the landscape of incentives in favor of giving the consumers what they want. The consumers will use these user agents when they get so good that they're better than the alternative and/or it's easy enough to "dual boot". Once one of the above happens, then there is a greater incentive for everyone to make their content available e.g. by uploading to public Google Drive folders. What I'm trying to get at is that the dream of decentralization can happen perhaps sooner if we focus on building the client side of it as a layer over the existing structure, instead of trying to supplant the current server side structure.
seydor 4y ago
it doesnt need to be 10x linear extrapolation, it just has to have 1 extra dimension. The decentralized dimension will never work however, because it is always illegal. Otherwise, gnutella, bittorrent, bitcoin, ipfs all could have become a decentralized "social network", aka "the n-th iteration of gopher/boards/forums" > can be much more profitable than a centralized network. The money comes from advertisers, not from users or views. Advertisers perennially spend the same amount (~1% of GDP, barring periods of war) to advertising companies. The pie is not expandable and is eaten by google and FB.
AlexandrB 4y ago
I think the #1 problem that needs to be addressed is how to completely remove content from a blockchain social network. Otherwise it's going to be a permanent record of every mistake someone makes online.
- omeze 4y ago
  In Farcaster, deletes are possible. They use a hash chain (new messages refer to previous message's content hash). If a user deletes a cast, they delete the content of the message and the hash is no longer correct. This makes deletion both possible and detectable, since users can self-host their messages. Note that in Farcaster only username mappings are on-chain, the rest of the messaging protocol is off-chain.
jsemrau 4y ago
In my opinion decentralization is a non-feature. I.e., I don't believe in people waking up in the morning and thinking "gee, that app would be 10x better if decentralized". What people care about is (1) getting better connections with friends and colleagues. Here lunchclub is doing an amazing job. (2) getting better news then have a look at finclout.io. If you are a regular posted for whatever motivation, I don't think that the current state of social media works well. Decentralized Social as an implementation of a Decentralized Social Network is really terrible because it doesn't add anything to the user experience besides a pay-layer which in reality brings out the worst in people "Please buy my coin'
- Jonovono 4y ago
  This is exactly right. If the words "decentralized" are anywhere in your marketing of a decentralized social network, you have already lost ;p I didn't intend to mean that by making a social network decentralized it would be 10x better. What I mean is, that by making a decentralized social network it opens up possibilities that are just not possible in a centralized network (or at least very hard). And those possibilities are real tangible benefits to users/creators that don't directly have anything to do with 'decentralization'.
  - jsemrau 4y ago
    We are at the end of one cycle of "social" apps. If we now stop and look back over the last 15 years we see 1. Art has degraded as Artists are optimizing their work for share-ability and a quick consuming audience.[1] 2. Social networks are pulling us further apart than bringing us together.[2] 3. Don't get me started on influencer culture[3] [1]https://www.youtube.com/watch?v=BPWkHtFPHH8&t=645s https://www.youtube.com/watch?v=BPWkHtFPHH8&t=645s [2] https://www.nature.com/articles/d41586-022-00805-0 https://www.nature.com/articles/d41586-022-00805-0 [3] https://techcrunch.com/2022/01/10/hype-house-creator-economy-netflix/ https://techcrunch.com/2022/01/10/hype-house-creator-economy...
    - Jonovono 4y ago
      What do you think will arise to address this? Whats the next cycle look like?
EGreg 4y ago
Tim Berners-Lee has been working on it for a long time, I met him and his team back in 2014 when it was called SOLID. He was just very into making everything work with SPARQL. His lead developer Dmitri Zagidulin agreed to work with us for a while but ultimately went on to help create the DID standard at the W3C. And Daniel Buchner at Microsoft pushed for years to make ION (sidetree protocol) which Jack Dorsey’s Web5 is based on. If you want to see truly decentralized and working social networking, here are some examples: https://yang2020.app https://yang2020.app https://ftl.fm https://ftl.fm https://gba.community https://gba.community All built on our framework: https://qbix.com/platform https://qbix.com/platform The eventual decentralized Web2 ecosystem can look like this, and there is no blockchain: https://qbix.com/ecosystem https://qbix.com/ecosystem Why we need it: https://qbix.com/blog/2021/01/15/open-source-communities/ https://qbix.com/blog/2021/01/15/open-source-communities/ Questions, comments and threats welcome :)
- Jonovono 4y ago
  This is awesome, thanks. Taking a look now, will drop some questions if they come up! I am following TBL's SOLID work as well! Edit: I thought I had reached the bottom of all the decentralized tech out there after a few months of research. qbix hadn't come up, looks really neat!

bckr 4y ago

<bad-idea> Maybe what will evolve the space is automatic translation from any possible format to any other format using AI. You don't need to manage a server or store anything in any particular way in any particular place. Instead, you expose whatever you want in any way you want, whether that's uploading a video to some big tech host, or exposing a bunch of XML over FTP, or a RESTful endpoint that serves binary encoding pictures of cats. My user agent scoops up everything and presents it to me in a way that I like. There would need to be some creative ideas involved for, say, "replying" (hashes? webmentions?) In other words, maybe innovation on the demand side is where the fun is. </bad-idea>

Jonovono 4y ago
Lets just say, I don't think it's a bad idea ;p '''My user agent scoops up everything and presents it to me in a way that I like.''' I think this is close to what the future looks like. For example, it rattles me that no matter what app I am watching 'videos' on I have different options. Some I can watch 2x speed. Some I can tap the sides to go quicker through. Some if I tap it mutes. Some if I tap it stops the video. All of this for the same type of content. Insanity. Not to mention even within certain apps like Instagram is just brutal. For the same content type - videos - they have like 3 (?) different ways to view it.
- godshatter 4y ago
  I wish the trend wasn't for social network companies to rewrite web apps from scratch in javascript when they could simply in this case just hand me the video and my preferred local application would play it. I get that they can display suggestions for what you should watch next and keep data on how much you watch or how often you pause it or whatever. But if we're going to create a decentralized social network, then maybe we could junk all the metrics that a centralized social network keeps on their users. Let it evolve organically, rather than in a way that best suits one particular small group of people, and give people back their privacy as well as their autonomy. Edit: rewrote an awkward sentence.
  - Jonovono 4y ago
    Exactly, one way I am thinking about it right now is do away with the idea of "apps" entirely. Think about it more in the sense of "intents". If anyone is interested in building in this space, my email is in my profile! Looking for cofounders :)
grumbel 4y ago
> Maybe what will evolve the space is automatic translation from any possible format to any other format using AI. Easier said than done. Facebook, Google, Twitter and Co. like to gatekeep their data and don't allow third party clients access. So you don't just need an AI smart enough to capture the data and reshape it, but also one smart enough to pass whatever CAPTCHA they throw at you, which will likely get progressively more difficult the more popular your tool gets. And of course they'll lock your account when you try, so this is at best a read-only participation in the social network. The problem has shifted over the last decade, it's no longer about fixing a broken Web, but about a large part of the content no longer even being a part of the Web. Everything is either moving to Apps or getting gated behind login screens and ToS.
- bckr 4y ago
  CAPTCHA is a big problem but not impossible to overcome. It's an arms race. That said, CAPTCHA can be defeated by changing the landscape of incentives in favor of giving the consumers what they want. The consumers will use these user agents when they get so good that they're better than the alternative and/or it's easy enough to "dual boot". Once one of the above happens, then there is a greater incentive for everyone to make their content available e.g. by uploading to public Google Drive folders. What I'm trying to get at is that the dream of decentralization can happen perhaps sooner if we focus on building the client side of it as a layer over the existing structure, instead of trying to supplant the current server side structure.

seydor 4y ago

it doesnt need to be 10x linear extrapolation, it just has to have 1 extra dimension. The decentralized dimension will never work however, because it is always illegal. Otherwise, gnutella, bittorrent, bitcoin, ipfs all could have become a decentralized "social network", aka "the n-th iteration of gopher/boards/forums" > can be much more profitable than a centralized network. The money comes from advertisers, not from users or views. Advertisers perennially spend the same amount (~1% of GDP, barring periods of war) to advertising companies. The pie is not expandable and is eaten by google and FB.

AlexandrB 4y ago

I think the #1 problem that needs to be addressed is how to completely remove content from a blockchain social network. Otherwise it's going to be a permanent record of every mistake someone makes online.

omeze 4y ago
In Farcaster, deletes are possible. They use a hash chain (new messages refer to previous message's content hash). If a user deletes a cast, they delete the content of the message and the hash is no longer correct. This makes deletion both possible and detectable, since users can self-host their messages. Note that in Farcaster only username mappings are on-chain, the rest of the messaging protocol is off-chain.

jsemrau 4y ago

In my opinion decentralization is a non-feature. I.e., I don't believe in people waking up in the morning and thinking "gee, that app would be 10x better if decentralized". What people care about is (1) getting better connections with friends and colleagues. Here lunchclub is doing an amazing job. (2) getting better news then have a look at finclout.io. If you are a regular posted for whatever motivation, I don't think that the current state of social media works well. Decentralized Social as an implementation of a Decentralized Social Network is really terrible because it doesn't add anything to the user experience besides a pay-layer which in reality brings out the worst in people "Please buy my coin'

Jonovono 4y ago
This is exactly right. If the words "decentralized" are anywhere in your marketing of a decentralized social network, you have already lost ;p I didn't intend to mean that by making a social network decentralized it would be 10x better. What I mean is, that by making a decentralized social network it opens up possibilities that are just not possible in a centralized network (or at least very hard). And those possibilities are real tangible benefits to users/creators that don't directly have anything to do with 'decentralization'.
- jsemrau 4y ago
  We are at the end of one cycle of "social" apps. If we now stop and look back over the last 15 years we see 1. Art has degraded as Artists are optimizing their work for share-ability and a quick consuming audience.[1] 2. Social networks are pulling us further apart than bringing us together.[2] 3. Don't get me started on influencer culture[3] [1]https://www.youtube.com/watch?v=BPWkHtFPHH8&t=645s https://www.youtube.com/watch?v=BPWkHtFPHH8&t=645s [2] https://www.nature.com/articles/d41586-022-00805-0 https://www.nature.com/articles/d41586-022-00805-0 [3] https://techcrunch.com/2022/01/10/hype-house-creator-economy-netflix/ https://techcrunch.com/2022/01/10/hype-house-creator-economy...
  - Jonovono 4y ago
    What do you think will arise to address this? Whats the next cycle look like?

EGreg 4y ago

Tim Berners-Lee has been working on it for a long time, I met him and his team back in 2014 when it was called SOLID. He was just very into making everything work with SPARQL. His lead developer Dmitri Zagidulin agreed to work with us for a while but ultimately went on to help create the DID standard at the W3C. And Daniel Buchner at Microsoft pushed for years to make ION (sidetree protocol) which Jack Dorsey’s Web5 is based on. If you want to see truly decentralized and working social networking, here are some examples: https://yang2020.app https://yang2020.app https://ftl.fm https://ftl.fm https://gba.community https://gba.community All built on our framework: https://qbix.com/platform https://qbix.com/platform The eventual decentralized Web2 ecosystem can look like this, and there is no blockchain: https://qbix.com/ecosystem https://qbix.com/ecosystem Why we need it: https://qbix.com/blog/2021/01/15/open-source-communities/ https://qbix.com/blog/2021/01/15/open-source-communities/ Questions, comments and threats welcome :)

Jonovono 4y ago
This is awesome, thanks. Taking a look now, will drop some questions if they come up! I am following TBL's SOLID work as well! Edit: I thought I had reached the bottom of all the decentralized tech out there after a few months of research. qbix hadn't come up, looks really neat!