3 ms·

I don't think that "user-agents" are the hard part either. At this point, I think any grad student would happily write a NN implementation that took various pos

I don't think that "user-agents" are the hard part either. At this point, I think any grad student would happily write a NN implementation that took various posts as input, and returned to you a sorted list based on your preferences (with input layers like: bag of words, author, links, time, etc, that the user could put more or less weight into just by simple upvote/downvote). The problem is that no one has the incentive to host such a service for free, and users wants the content to be available 24/7. So it's not as simple as just setting up a peer-to-peer network. Users who just use a phone as their primary computer will still want to be able to publish to their millions of followers, and so it wouldn't work to have those millions of people connect directly to this person's device. Maybe you can solve that with a bit-torrent like approach, but the problem gets harder when you include the ability to send messages privately.

pwdisswordfish2 6y ago
"Users who just use a phone as their primary computer will still want to able to publich to their millions of followers, and so it wouldn't work to have these millions of people connect directly to this person's device." You have shifted the discussion from small overlay network for friends and family to large overlay network for "millions of followers". Those methods of sharing content with "millions of followers" are already available and will no doubt continue to be available. A small private network is a different idea, with a different purpose. People will always have the choice of using a public network however a small overlay can avoid sending traffic through third party servers, like Facebook's. There is no requirement that a service has to be "free", or supported by ads. This is something else you injected in the discussion. I use free software to set up overlays, but I have to pay for internet service and hosting. The cost of the "service" is not the setup it is the internet access and hosting.
- ggggtez 6y ago
  Your idea doesn't sound particularly tenable. A pay social network that limits you to only your close friends and family, that will have no network effects, have less features, and be more difficult to set up... It's easy for people to point what they don't like about Facebook, but I don't think you are really comprehending why they dominated the social network space to begin with. It's not as easy as making a product that doesn't advertise, if it costs money to use instead.
  - pwdisswordfish2 6y ago
    Tenable for what? You are injecting your own ideas. Why would I care about the reasons Facebook is popular. The idea I submitted was to make the process of setting up overlay networks easier, not to try to start a web-based business. Making software easier to use is a tenable idea. The ideas you are introducing might not be tenable. However, they are not my ideas. The pattern I see is you introduce some idea, attribute it to me, then shoot it down. https://en.wikipedia.org/wiki/Straw_man https://en.wikipedia.org/wiki/Straw_man
mycall 6y ago
> The problem is that no one has the incentive to host such a service for free matrix.org is approaching this

pwdisswordfish2 6y ago

"Users who just use a phone as their primary computer will still want to able to publich to their millions of followers, and so it wouldn't work to have these millions of people connect directly to this person's device." You have shifted the discussion from small overlay network for friends and family to large overlay network for "millions of followers". Those methods of sharing content with "millions of followers" are already available and will no doubt continue to be available. A small private network is a different idea, with a different purpose. People will always have the choice of using a public network however a small overlay can avoid sending traffic through third party servers, like Facebook's. There is no requirement that a service has to be "free", or supported by ads. This is something else you injected in the discussion. I use free software to set up overlays, but I have to pay for internet service and hosting. The cost of the "service" is not the setup it is the internet access and hosting.

ggggtez 6y ago
Your idea doesn't sound particularly tenable. A pay social network that limits you to only your close friends and family, that will have no network effects, have less features, and be more difficult to set up... It's easy for people to point what they don't like about Facebook, but I don't think you are really comprehending why they dominated the social network space to begin with. It's not as easy as making a product that doesn't advertise, if it costs money to use instead.
- pwdisswordfish2 6y ago
  Tenable for what? You are injecting your own ideas. Why would I care about the reasons Facebook is popular. The idea I submitted was to make the process of setting up overlay networks easier, not to try to start a web-based business. Making software easier to use is a tenable idea. The ideas you are introducing might not be tenable. However, they are not my ideas. The pattern I see is you introduce some idea, attribute it to me, then shoot it down. https://en.wikipedia.org/wiki/Straw_man https://en.wikipedia.org/wiki/Straw_man

mycall 6y ago

> The problem is that no one has the incentive to host such a service for free matrix.org is approaching this