3 ms·
What I'd take as lessons: 1) It's better to start with a protocol and build an application than just launch into application building. 2) Privacy is wrong mot
by joe_the_user 14y ago
What I'd take as lessons:
1) It's better to start with a protocol and build an application than just launch into application building.
2) Privacy is wrong motivation for wanting to replace Facebook. A distributed social networking application is certainly possible (one might argue the web itself an example). A secure, distributed sharing scheme makes a social networking app exponentially harder to develop (Project Xanadu tried to create a secure "transclosure" scheme. They had a ten year head-start on the web and still failed. Distributed, revocable sharing is essentially a pathological problem).
- shmerl 14y agoNo, privacy is really required for something that aims to compete with Facebook, simply because Facebook can't compete on privacy.
- 001sky 14y agoAssuming this is true, how much is control rights vs need for de-novo architechture?
- shmerl 14y agoI'm not sure if I understood the question, but if you mean how user control is related to the decentralized structure, then the answer is in the question. Decentralization essentially prevents one entity from being in control, and in edge case control is given to a user who runs a personal pod for single Diaspora account.
- 001sky 14y agoAbout the notion that the data needs to be locally stored/controlled/at the edge. What about the idea of encryption and/or persistence management? (user control) Does this provide more flexibility on Architecture. Edited: for clarity.
- shmerl 14y agoI think encryption is good to have and D* devs had it in some further plans. But it's probably not the first priority, until federation protocol is rock solid.