See here for a description of what I'm doing: https://poal.co/s/technology/569412
I want to make a log about what I'm doing to get community feedback while I try to raise this project from the dead.
I looked more into the dissenter code. It actually does most of the work server side. Basically when you open the extension, it sends a request to the server: ex. "https://dissenter.com/discussion/begin-extension?url=https%3A%2F%2Fnews.ycombinator.com%2F" and then displays whatever the server sends back into the contents of the extension as an iframe. Most of the code is held on the server side. If you try the example link above, you can see there is still a running server there, but it doesn't let you login or signup.
I can't find any of the server side code anywhere so it seems not to be opensource, unlike the client. I don't know how helpful it would be anyway, because I want to make something more resilient then a simple server/client model that can be easily shutdown. I might change my mind to get something running for v1 though and circle back later to make it more resilient.
I've been trying to find protocols to use that are hopefully, a) can withstand attacks, b) can handle remove spam, c) are up to date and being actively maintained.
The most promising protocol I found so far is hypercore, https://hypercore-protocol.org/ and I'm messing with it to see if it will satisfy my needs. It uses DHT and works a lot like bittorrent so there is no central server, and is being updated regularly. It also runs in javascript so I should be able to incorporate directly into the dissenter code. It basically allows for append-only data files with random r/o access. In my head, not sure if it will work or not, is that the clients will look for a log associated with the url they're visiting and read that. The log would contain any action, upvote, downvote, submit-comment, edit-comment, so all the data would be there. If there is a spam, there'd be a separate log file by made by a "validator" which contain a blacklist of spam content. Not sure of the details yet, but I'm playing around with the tutorial.
I also read the license file for the dissenter code. It's basically allows for anything, I can make changes and republish without issue. The only problem is that I'm not allowed to use their trademarks and copyrights, so I've got to rename it from "dissenter" to something else. I thought of "captaindirgo", an old star trek tng reference, he was this guy who kept pissing off Wesley Crusher and Picard and drinking moonshine while they wandered around marooned on a desert planet and it was Wesley's last regular appearance. But if anyone has some good ideas for a new name, let me know. It should have a .com and a .org name available, of course.
See here for a description of what I'm doing: https://poal.co/s/technology/569412
I want to make a log about what I'm doing to get community feedback while I try to raise this project from the dead.
I looked more into the dissenter code. It actually does most of the work server side. Basically when you open the extension, it sends a request to the server: ex. "https://dissenter.com/discussion/begin-extension?url=https%3A%2F%2Fnews.ycombinator.com%2F" and then displays whatever the server sends back into the contents of the extension as an iframe. Most of the code is held on the server side. If you try the example link above, you can see there is still a running server there, but it doesn't let you login or signup.
I can't find any of the server side code anywhere so it seems not to be opensource, unlike the client. I don't know how helpful it would be anyway, because I want to make something more resilient then a simple server/client model that can be easily shutdown. I might change my mind to get something running for v1 though and circle back later to make it more resilient.
I've been trying to find protocols to use that are hopefully, a) can withstand attacks, b) can handle remove spam, c) are up to date and being actively maintained.
The most promising protocol I found so far is hypercore, https://hypercore-protocol.org/ and I'm messing with it to see if it will satisfy my needs. It uses DHT and works a lot like bittorrent so there is no central server, and is being updated regularly. It also runs in javascript so I should be able to incorporate directly into the dissenter code. It basically allows for append-only data files with random r/o access. In my head, not sure if it will work or not, is that the clients will look for a log associated with the url they're visiting and read that. The log would contain any action, upvote, downvote, submit-comment, edit-comment, so all the data would be there. If there is a spam, there'd be a separate log file by made by a "validator" which contain a blacklist of spam content. Not sure of the details yet, but I'm playing around with the tutorial.
I also read the license file for the dissenter code. It's basically allows for anything, I can make changes and republish without issue. The only problem is that I'm not allowed to use their trademarks and copyrights, so I've got to rename it from "dissenter" to something else. I thought of "captaindirgo", an old star trek tng reference, he was this guy who kept pissing off Wesley Crusher and Picard and drinking moonshine while they wandered around marooned on a desert planet and it was Wesley's last regular appearance. But if anyone has some good ideas for a new name, let me know. It should have a .com and a .org name available, of course.
(post is archived)