–m

At any rate anonymity will severely reduced download speeds and that’s just not a good start, people will stick with their old stuff if the new stuff ‘doesn’t work’.

Although starting from scratch is of course a bit exaggerated every new protocol benefits from its predecessors but again just adding new features doesn’t cut it imo.

Concerning privacy there is nothing to work on unless you make MAJOR changes to the protocol and these changes would be so big you can better create a new protocol from scratch.

No need to start from scratch. Anonymous P2P and semi-anonymous (e.g. friend-to-friend) networks have existed for years and are continually be worked on. If you want to use them all they need is a userbase.

Which would probably be a good thing but since uT is owned by BitTorrent, Inc. the days of developing (more or less effective) encryption stuff and such with Azu are over… so the guys from TPB will probably not get much love from the biggest client distributor on the net. What ever they do.

No changes to the client means no changes to the server. Unless we all change clients *cough* and all the sites change along with that. New tracker new everything… never gonna happen.

So… gz we are stuck with what we got for now hence new tracker even though new a system would be better.

Use what talent you possess: the woods would be very silent if no birds sang except those that sang best.
~ Henry Van Dyke

More than that you can poison the stats pretty easy if you relay on client information and there is not much else to relay on.

Last but not least this would create more stress for the tracker (bandwidth, load and memory -wise) with virtually no improvements to the protocol.

BT is not an anonymous protocol. Sorry but that idea is just bs.
Splitting the swarm… em no thank you! It would improve privacy though.

the evil DHT is not the answer because the evil DHT permits peers blocked by a safe tracker to join a swarm and log downloads

Haha. How exactly do you think that a tracker based system will prevent monitoring? IP addresses are easy to get (Mediadefender was using their employees home connections). Retrieving peer IP addresses is normal behaviour for a Bittorrent client. How will they detect people that are not actually spoofing files? Even if you could detect most of them it still wouldn’t be good enough: In contrast to spoofing, very few ‘rogue’ peers are needed to monitor successfully.

The only way to prevent monitoring are onion networks or Friend-to-Friend networks, both of which necessarily suffer from far slower speeds than non-anonymous networks like Bittorrent.

The thing is it’ll probably be difficult to get it right for everyone. Since it’s being developed by tpb they’ll focus on public tracker stuff (bandwidth, load balancing, cache, open fds and whatnot). They’ll have to limit the dev team which leaves not much space for interoperability.
I sure hope this doesn’t become a fork fest.

Am I the only person whose never heard of EZ2net?

Todos sabemos que thepiratebay es un sitio muy conocido de torrents. Desgraciadamente, no estaba rindiendo muy bien últimamente. Ahora el equipo de thepiratebay está trabajando en un tracker open source más eficiente que, entre otras cosas, proteger…

Another possible idea I’d like to see is the OPTION of choosing a “secure tracker”. By that I mean a tracker that :

1. blocks out bad IPs
2. supports multiple swarm cells — The tracker gives out the fewest number of peers needed for a reasonably fast download. So instead of one giant swarm, there are many tiny ones. Yes, this will result in a slower download, but will be more private and safer because everyone’s IP address is only given out to a few other peers. Because the IP addresses are filtered from the peerlist before peers are returned to the downloader, the chance of coming in contact with an anti are greatly reduced.

@shelia –

Please list the Usenet provider where you “pay $50 a year for all the movies music and porn i can imagine” – i think we’ll all be lining up for that one.

The cheapest of the cheap charge several times that price for unlimited usenet service.

If there is, then god help us all..

Rock on Pirate Bay, rock the hell on.[/quote]

As you consume your electricity, are you aware of that of every electron there’s a downside…A electron has to get to earth based plugs…and where does those go? go figure…in the earth which gives of more heavy weather and more lighting storms and so on…etc…

so hell is broken loose now :)

The only real way to get security is to go Private.

Just need to know the right trackers, with the best community’s. No ads, No fake shit, No RIAA/MPAA

Good luck TPB, sounds great. But for me, privet trackers are where it’s at.

should see something like: hxxp://tpb.tracker.thepiratebay.org/announce or similar somewhere, and its currently down atm for some.

on topic: great their trying to make a better tracker ;) always good!
and hopefully they will win their law cases so they wont have trouble getting their new nice shiny tracker spammed down by lobby organisations..:p

evil DHT = **AA logging your downloads
______________________________________

TPB are making some errors

1. tracker ipfiltering and logging at the db level has a VERY VERY VERY high cost. it is infinitely better to do the ipfiltering in the kernel and if necessary design services to poll the kernel logs
2. their current (published) server specs are a jumble of computers the best of which is barely one step up from a multi-media desktop system … they cannot be serious. please please please get yourself a hardware specialist to give you some advise on what a real server looks like
3. there is plenty redundancy in clustered services with dedicated hardware buffers (slack in the system) ie. use a 8x dual-core cpu server with 64GB RAM instead of 12 individual boxes

i wish the guys luck but it is not a good start. TPB tracker has been unreliable and slow for years and they need a NEW plan, NOT a plan which builds on their past mistakes

The solution to 1 & 2 is decentralization and large scale distribution (e.g. DHT). For some types of attacks (e.g. tracker DoS or take down) it is also the solution to 3. For other types of attacks like spoofing or corrupt data poisoning it would seem far more effective to be able to automatically use information from ‘good’ peers to determine ‘bad’ peers than to try ‘manually’ maintain lists on the trackers (a la peerguardian). And again, this works better in a distributed setting.

Consequently, if you are a developer it would seem more prudent to spend time on implementing DHT-search and solving the spoofing and poisoning through some sort of distributed trust network than to work on this tracker.

Go TPB!!!

ThePirateBay CAN however, be the shit!
And keep the fight going as long as possible!
When I get paid again, I can’t wait to buy some merch from them!
To help support them!

I hope they get an island, that would be sick!