Latest Announcemnent
New thread

New filesharing concept - is it possible?

Status
Not open for further replies.
T

theforumdrunk

Active Member
63
2010
2
0
So my idea is based around the p2p ideals of limewire/kazaa/grokster only instead of getting your files from other people directly you are getting them from their cloud accounts.

Let's say there is a website called "dropshare" that you sign into your premium dropshare account and search for whatever files you want and download them... right there on that website. No illegal files are stored on the dropshare servers but are streamed from and to various users cloud accounts.

Obviously cloud accounts are restricted to be viewed and used by those with the credentials, but if these credentials were shared (or the server application gained permission) and the files were searchable/downloadable by the server, you could start serving them out to another users cloud account and not to the user directly.

The dropbox API is a great way for this to be achieved without giving away your dropbox credentials. The dropbox account would authenticate the "dropshare" app which would create a folder of shared stuff in the users dropbox. Anything they put in there could be shared back to the dropshare index.

It's hard to describe but essentially you are merging all these cloud accounts into one, indexing all the files for searching and then shifting those searched files into the appropriate users cloud account.

Can anyone decode my thinking here? Maybe there's a flaw?

Thoughts?
 
7 comments
I've been thinking of ways to centralize ( while not being vulnerable ) files as well.

Anyway, this is my view on what you meant, and on how it could be achieved ( or not ):
- If I misunderstood, I'm sorry :)

You want dropbox but instead of having the files use space on your own hdd, you want it to be solely stored in the cloud.
( Currently dropbox is a mirror of your local disk, so you need the files to be on your HDD. )
Then you want those files to be searchable ( with permission ).
[ So you want RapidShare ]

It can be done if the company owning the servers is legit and has nothing to do with filesharing.
If they offer an api someone else would have to build a platform where:
- People can hook up their account.
- The account gets indexed
- The index is made public, but it's not shared who's file it is.

So far, this can be done using RapidShare.

But the problem lies when you want people to download.
You'll always need to echo the location of the file or proxy it through your own servers.
If you opt to go with your own hardware you're gonna have a bad time.
If you opt to go with echoing ( just telling people the DL-link ) you'll still get files to be removed due to DMCA reports.

In the end the only thing you end up with is a unique communitybased platform.
However, if the community is big enough it will work as there will always be duplicates available.

So, if you do this cleverly ( only showing 1downloadlink instead of everyone's ) you can make sure you always have a backup link. This will make the guys that report the links go insanse as they can only remove the next link when the previous one has been dmca'd by RapidShare.


*I'm a RapidShare lover, but I guess any hoster would do.

I don't mind developing this as I'm seriously bored for over a few months now.
If anyone else is serious about this we CAN make this happen!

Edit: Okay, it's not exactly what OP said but this works even better since we don't have to have the accounts of people. With this version they only have to post the links to stuff they have!
 
Last edited:
Its not too bad of an idea but it'd never work.

Even if you don't store the files, you are still streaming them through your server (so that introduces a single point of failure) and you are serving them, which then means you can still be taken to court over it (see the many examples from websites which embedded movies from third-party hosts).

Additionally, you'd have an incredible amount of data transfer going on (2* the size of the file per download) and you'd have increased latency due to it having to select a source then proxy that source back to the user.

Also you would be depending on storage you don't have sole control of. Meaning at any point, all your data could be lost in a worst case scenario and you would have no control over it, no ability to prevent it.

So end of the day, all laws and legalities still apply, it would be slower, it would still have a single point of failure rather than being decentralised and it would result in uncontrollable sources.

Keep thinking ;)
 
Status
Not open for further replies.
Back
Top