One thing to do is to improve peer-2-peer technology , that it will be fit for the task.To do this we'll need a few things :
1. Torrents(as the most proven tech) should be able to handle storage of rare file (i.e. persistence) well[a].
2. A working peer to peer search mechanism - i believe tribler has that. It might need some customization for academic papers - but it's written in python , so it won't be terribly hard.
3. Tribler has mechanisms for anonymity and against censorship. One more thing that should be added is a secure pdf reader - maybe pdf.js via chrome is the most secure we have today.
4. Maybe there would be a need for some spam protection.
I don't know. like i mention in the link of the previous discussion - At least for this goal - Tribler seems to be mostly there including being part of the huge torrent community, it's written in python - so it's seem like it could easily do the job - so why complicate things at this stage which should be about and MVP and building a community ?
1. Torrents(as the most proven tech) should be able to handle storage of rare file (i.e. persistence) well[a].
2. A working peer to peer search mechanism - i believe tribler has that. It might need some customization for academic papers - but it's written in python , so it won't be terribly hard.
3. Tribler has mechanisms for anonymity and against censorship. One more thing that should be added is a secure pdf reader - maybe pdf.js via chrome is the most secure we have today.
4. Maybe there would be a need for some spam protection.
[a]search for 'bad for storing "rare" files' in https://news.ycombinator.com/item?id=10962253