On the legal front it's like Dropbox -- storing your personal media collection online and letting us transcode and organize it is fine, especially because the content is private (like Dropbox) and not publicly listed (we're not like Grooveshark). Of course, we're DMCA compliant and take down files if they're deemed illegally obtained.
So does that also mean you have to store every file? or more like iTunes Match where if the file exists on the server, everyone gets served the master, instead of storing duplicates from millions of users.
You don't really have to answer that. it's not an end-user concern.
We're actually storing a single copy of every unique file. i.e. if you upload a completely identical, MD5 hash duplicate copy of something we already have stored, we'll just give you a pointer to that file. This way, we don't have to store every file, and we're not like iTunes Match (which actually requires licensing) because we're not giving you a master file (iTunes Match gives you a master file that plays at 256kbps, even if you upload the same song at 128Kbps -- they can do this through licensing; we can't give you a master 720p video file even if you upload a 360p version because we don't have those licenses). Legally, doing the MD5 hash duplicate technique was deemed legal after MP3Tunes' case (not referring to their entire case, just the part about the MD5 hash duplication) because it's just like any other method of compression.
Comments
On the legal front it's like Dropbox -- storing your personal media collection online and letting us transcode and organize it is fine, especially because the content is private (like Dropbox) and not publicly listed (we're not like Grooveshark). Of course, we're DMCA compliant and take down files if they're deemed illegally obtained.
I see, thanks for that.
So does that also mean you have to store every file? or more like iTunes Match where if the file exists on the server, everyone gets served the master, instead of storing duplicates from millions of users.
You don't really have to answer that. it's not an end-user concern.
We're actually storing a single copy of every unique file. i.e. if you upload a completely identical, MD5 hash duplicate copy of something we already have stored, we'll just give you a pointer to that file. This way, we don't have to store every file, and we're not like iTunes Match (which actually requires licensing) because we're not giving you a master file (iTunes Match gives you a master file that plays at 256kbps, even if you upload the same song at 128Kbps -- they can do this through licensing; we can't give you a master 720p video file even if you upload a 360p version because we don't have those licenses). Legally, doing the MD5 hash duplicate technique was deemed legal after MP3Tunes' case (not referring to their entire case, just the part about the MD5 hash duplication) because it's just like any other method of compression.
So if I know the MD5 digest of some content I would like, all I have to do is generate a collision and then "upload" it?