Partnered With:

Durabit: The new generation of BitTorrent, seeking solutions to the "seeding" problem.

Read this article in 12 Minutes

Durabit uses time-locked Bitcoin transactions and Chaumian eCash mints to create economic incentives for BitTorrent users to seed files.

Original Title: "DURABIT: INCENTIVIZING TORRENT SEEDING WITH BITCOIN"

Original Author: Shinobi

Translated by: Luccy, BlockBeats

Editor's Note:

BitTorrent has played a critical role in internet data transmission for 22 years. However, BitTorrent has always faced the problem of users not wanting to continue seeding after downloading, making files difficult to obtain.

Recently, there has been a new discussion in the community about storing data on the Bitcoin chain. Durabit, as a Bitcoin-native data incentive mechanism, uses eCash and OP_RETURN to solve the seeding problem of BitTorrent. Shinobi, a self-taught educator in the Bitcoin field, believes that this design seems to provide an effective solution to ensure the continuous availability of data, especially when user demand weakens. The emergence of Durabit brings new possibilities for the long-term storage problem of BitTorrent, providing users with more incentives to ensure the continuous availability of files in the network.

BitTorrent has been around for 22 years. In many ways, it is a technical protocol that is almost as important as Bitcoin, as it has had a significant impact on changing the way data is transmitted over the internet. If Bitcoin is a currency used to send funds when people don't want you to do so, then BitTorrent is a mechanism used to transfer data when they don't want you to do so. However, it has always had a big problem, and I believe anyone who has ever used it is very familiar with it. This problem is the seeding problem.

How many people immediately close their torrent client after completing a download and do not continue to seed the file? Almost everyone has had this experience. Without users online and providing downloads for others, BitTorrent cannot operate, and most users do not continue to seed after obtaining the complete file. Especially when the demand for a file is high, people will seed various parts of the file while downloading, but once they finish downloading, they will stop seeding. However, at the same time, other people come online and start downloading, and they are also seeding while downloading. As long as the group that goes through this cycle is large, this method is effective, but if not, the torrent will gradually disappear and become unavailable due to people stopping seeding.

This brings about a problem for the long-term storage of individual seed files. Although it is an excellent protocol for circulating data during high demand, the data often becomes inaccessible when demand decreases due to people stopping seeding. Durabit is a recently proposed solution to this problem, which is relatively simple but seems to provide a reliable incentive mechanism for people to continue seeding files.

This system relies on Chaumian ecash mint to drive the incentive mechanism for file seeders. Third parties who wish to ensure the availability of files have entered into contracts with the electronic cash mint, using a series of pre-signed transactions with time locks. Each transaction is time-locked for a period of two weeks and pays a small amount to the Chaumian ecash mint. Each payment is a time-locked UTXO that cannot be spent until the next transaction becomes valid, with the remaining funds eventually returning to the address controlled by the initiator of these transactions. The next transaction in the chain will spend the change output.

In this series of transactions, the first transaction associates the contract with the file that the issuer wants to encourage seed operations on by recording a specific seed magnet link in the OP_RETURN output. After mint holds these presigned transactions, it submits the first transaction to the chain and begins monitoring the seed swarm for the specific magnet link. Then, mint waits for any seed client running the Durabit client to contact it. If a Durabit client sends a request to mint from the same IP address that it sees in the seed swarm, it will maintain the connection through a separate channel.

During this process, mint monitors and tracks seed users who have registered with it. In the two weeks prior to the latest expenditure becoming available, mint distributes Chaumian ecash tokens to each registered seed user to ensure the continued availability of data. Mint can allocate tokens proportionally based on the amount of data, or randomly distribute tokens among registered seed users through a lottery. Once the expenditure output becomes available, it can declare and open an exchange window to exchange the issued Chaumian tokens for actual bitcoins during the sowing cycle. This cycle will continue until a series of pre-signed transactions are completed. For the issuer of the contract, the total amount of bitcoins contributed to the contract and the amount paid per cycle are entirely up to them.

I am convinced that most people are wondering: "What prevents Chaumian mint from simply collecting these expenditures without distributing a portion of them to those who are planting?" This is the beauty of this proposal: pure incentive. Each transaction pays a small amount of money to Chaumian mint, which is output in a time-locked manner, and the remaining funds are returned to the issuer of the contract. At any time, the party issuing this contract can effectively revoke it by spending twice the output, and all subsequent presigned transactions will be invalid from that point on. The mint understands this and must weigh the potential loss of all future income from any individual contract by agreeing to a percentage of each expenditure collected for itself, against the potential benefits of retaining the entire expenditure but losing a percentage fee of all future expenditures.

On the other hand, the issuer of the contract was initially motivated to issue the contract in order to maintain the availability of a specific document by incentivizing people to seed it. If they truly want the document to remain available, the best course of action is to not revoke any contracts they have already issued unless the mint behavior fulfilling the contract is malicious. This arrangement allows for incentives to be properly aligned so that for the mint, monitoring the seed network and honestly distributing funds to seeders should be in their best interest; and for the contract issuer, it is best to avoid double-spending and revoking contracts as long as the mint continues to operate honestly.

This proposal focuses on the issue of actual audit integrity, whether it is from the perspective of mint auditing the distribution of tokens and payments to seeders, or from the perspective of issuers auditing mint. In the case of mint auditing seeders, they can choose to download random blocks of seed files on a regular basis. This should provide sufficient assurance that any individual seeder actually owns and provides the file to other users. In the case of issuers auditing mint, indirect monitoring of the seed network should provide sufficient basis for evaluating the integrity of mint. Once the contract starts, mint begins to distribute payments, and the seed network should establish a traffic baseline proportional to the economic incentives provided by the contract. If the issuer notices a significant drop in seed network traffic at any time, this is a good indicator that mint has not honestly handled the distribution, and the contract should be revoked.

Neither of these methods is 100% reliable, especially when auditing seed users in mint, but they should be good enough. Ultimately, if a planter is essentially just getting data from other planters in response to mint challenges, then the data does need to be available enough for them to get any random blocks generated by mint challenging them. Therefore, in this case, while an actor may be able to dishonestly collect payment from mint without hosting and providing files, they will not be able to manipulate the system in this way if the files are actually unavailable. I do not believe this is a fatal flaw, as it still achieves the overall goal of ensuring file availability.

Overall, Durabit is a very simple system with Chaumian mint as a trusted and convenient party, but I believe that simplicity is its advantage. The amount of funds that can be maliciously escaped from mint is very small, and if such an event occurs, the issuer of the contract can simply revoke the existing contract and reissue it with another mint. I think it provides a very simple and elegant solution to the incentive problem of using BitTorrent to seed files, even when user demand drops significantly.