3 ms·

Paying for dating makes complete sense. Women want to receive less spam. Men want there to be less competition from spammers. By adjusting the price that men ha

by mantap 6y ago

Paying for dating makes complete sense. Women want to receive less spam. Men want there to be less competition from spammers. By adjusting the price that men have to pay you can solve both of these problems. I used okcupid back in 2010 when this was written. It was a great website for taking quizzes and personality tests but far too noisy to effective for meeting people without becoming a spammer.

dane-pgp 6y ago
Spam and fake accounts are the big problems to solve for seemingly all popular sites, and unfortunately they are especially hard to solve in decentralised systems. Do people still remember Geek Code blocks?[0] In theory, all existing online dating services could be replaced by a global decentralised system, where people upload a somewhat standardised dating profile[1] and crawlers would compete to find as many of these as possible and match compatible people. Of course they would also compete in terms of how nicely they presented this data and how easy it was to create a profile. One approach for stopping spam would be to limit the number of messages that each profile (owner) can send per time period. For example, the profile could contain a "send-introductions-to" field, specifying the node that stores their inbox, and a "max-introductions-per-month" field which tells a sending node how many other introductions a user can make while still being allowed to contact them. As usual with federated systems, there is then the problem of nodes deciding which other nodes to federate with, but it should be fairly easy to test if a node is not enforcing spamminess limits. The other problem is stopping people from generating profiles on multiple nodes, which might require either a centralised service or some sort of economic system like "proof of burn"[2], unless email providers were to all start issuing time-limited anonymous tokens. [0] https://en.wikipedia.org/wiki/Geek_Code https://en.wikipedia.org/wiki/Geek_Code [1] I'm imagining using the taxonomy of Wikidata, allowing statements like { "has": "https://www.wikidata.org/wiki/Q202466 https://www.wikidata.org/wiki/Q202466" } [2] https://99bitcoins.com/what-is-proof-of-burn/ https://99bitcoins.com/what-is-proof-of-burn/

by mantap 6y ago

dane-pgp 6y ago

Spam and fake accounts are the big problems to solve for seemingly all popular sites, and unfortunately they are especially hard to solve in decentralised systems. Do people still remember Geek Code blocks?[0] In theory, all existing online dating services could be replaced by a global decentralised system, where people upload a somewhat standardised dating profile[1] and crawlers would compete to find as many of these as possible and match compatible people. Of course they would also compete in terms of how nicely they presented this data and how easy it was to create a profile. One approach for stopping spam would be to limit the number of messages that each profile (owner) can send per time period. For example, the profile could contain a "send-introductions-to" field, specifying the node that stores their inbox, and a "max-introductions-per-month" field which tells a sending node how many other introductions a user can make while still being allowed to contact them. As usual with federated systems, there is then the problem of nodes deciding which other nodes to federate with, but it should be fairly easy to test if a node is not enforcing spamminess limits. The other problem is stopping people from generating profiles on multiple nodes, which might require either a centralised service or some sort of economic system like "proof of burn"[2], unless email providers were to all start issuing time-limited anonymous tokens. [0] https://en.wikipedia.org/wiki/Geek_Code https://en.wikipedia.org/wiki/Geek_Code [1] I'm imagining using the taxonomy of Wikidata, allowing statements like { "has": "https://www.wikidata.org/wiki/Q202466 https://www.wikidata.org/wiki/Q202466" } [2] https://99bitcoins.com/what-is-proof-of-burn/ https://99bitcoins.com/what-is-proof-of-burn/