|
 |
I'm guessing that with the move to DHT and the take-down of the Pirate Bay,
the next step is going to be to decentralize or at least plausibly deny-ify
the actual searching. I.e., you won't be able to tell what's in a torrent
until you download the torrent.
An easy way to do this would be to have data streams have descriptions of
their contents at the start, and have the searchable torrents contain just a
bloom filter full of bits.
In other words, the .torrent file would have a name that's the hash of the
.torrent file, and a section that has the bloom filter bits for the first
file in the torrent data stream. The first file in the torrent data stream
would contain the information that you'd be able to search on, like the
names of files (i.e., the manifest), the descriptions of the content, the
lyrics (for music), and so on. Then, to do a search, you plug in what you're
searching for, find the .torrent files with bloom filter bitmaps that match
your search terms, then connect to the appropriate sharers to fetch the
first few blocks of each matching torrent-data-stream to see what is in the
torrent.
This would eliminate the ability to rationally accuse some site like The
Pirate Bay of knowing what's in the torrents they're serving. The site would
have to actively go and try to download some of every data stream and then
check it to find out what's in the torrent.
Not that I'm judging the morality of such a setup. It just seems like a
logical progression to me. If you start slamming search engines for
copyright violations, make sure search engines don't know what you're
searching for.
I thought about this before back when Freenet was all in the news, as I was
trying to come up with something that actually had a pre-hoc design to it,
but it never occurred to me to use bloom filters. I was doing multiple
rounds of hashing of keywords, which obviously doesn't work nearly as well.
--
Darren New, San Diego CA, USA (PST)
There's no CD like OCD, there's no CD I knoooow!
Post a reply to this message
|
|
