Hacker News Reader

Comment by _1tan | Hacker News Reader

Is the source code of the Sci-hub proxy open source (i.e. their secret sauce)?

SciHub relies on credentials donated (or sold) by people with access to journals to obtain the papers. (If you are in a university, your university likely provides access to many journals and pays for it on your behalf). So yeah, even if they did open source the code, you'll still need to obtain the credentials somehow. Which isn't very pratical.

Well, that's not necessarily true for current existing documents that are already available. I believe they're pinned on IPFS nodes, and so are available through IPFS. So if you did clone their front end and index you should be able to continue operating the system, at least with current availability.

What would be nice would be a distributed table that functions as an index specifically for these documents, that way researchers with access could pin to IPFS themselves and add the keywords and content hash to the table. This would be basically unstoppable.

Unless company behind IPFS bans it. They are a US company after all.

The fact IPFS effectively allows anyone to 'ban' a content hash makes it effectively useless.

Is that true? As in, I can ban a hash and nobody can then access it? I've never heard of that.

It's an oversimplification

If you run a node, you could block hashes of content you don't want hosting. Other nodes can still host it though.

Can other nodes find it?

If that's the case, I don't see how it is a problem. The user should be able to decide what content their node hosts.

You can do more than that. You can position yourself in the DHT in a place that even though you don't host the specific data, all users must go through you to find out who hosts it. You then falsely tell all requesters that the data doesn't exist.

Do you have any more details on this problem?

Protocol labs? How can they ban something on a distributed network?

Reply on news.ycombinator.com