Modification: Restricting Datacap Allocation for Newly Created Accounts (Trust & Reputation Improvement)

[herrehesse]

Issue Description

Accounts that are newly created on GitHub or have no history of using datacap should not be permitted to request 1 to 15 PiB+ of datacap on their applications due to the absence of trust and reputation. Instances of abuse involving new unaccountable accounts have been observed countless times, posing a risk to the integrity of the Filecoin network. To uphold network integrity, it is necessary to implement a more stringent process for allocating datacap to new accounts.

Impact

Allowing new accounts to request significant amounts of datacap without demonstrating their value to the Filecoin network or proving their ability to handle and distribute datasets has resulted in severe cases of abuse over the past six months. This undermines trust within the network and jeopardises the overall integrity of the system.

Proposed Solution(s)

To address this issue, the following steps are proposed:

• Limit accounts with no trust in the Filecoin Plus ecosystem to a maximum of 100 TiB of datacap.
• Enable them to showcase their skills, responsibilities, and adherence to the program's rules.
• Once they gain the trust of the community and notaries, allow them to increase their datacap request by opening a new LDN application with a higher datacap limit.

Benefits

By reducing the influx of large requests from unknown and reputation-less entities, we can improve the focus of notaries on tasks that are important and enable them to perform due diligence on proper clients and datasets who really add value to the Filecoin ecosystem. This ensures that the allocation of datacap is directed towards trusted entities, enhancing the overall efficiency and effectiveness of the notary system.

Timeline

The proposed solution will be discussed during the next T&T community call, followed by the implementation of new rules and guidelines.

Examples

filecoin-project/filecoin-plus-large-datasets#2103
filecoin-project/filecoin-plus-large-datasets#2100
filecoin-project/filecoin-plus-large-datasets#2098
filecoin-project/filecoin-plus-large-datasets#2094
filecoin-project/filecoin-plus-large-datasets#2092
filecoin-project/filecoin-plus-large-datasets#2090
filecoin-project/filecoin-plus-large-datasets#2089
filecoin-project/filecoin-plus-large-datasets#2083

[MRJAVAZHAO]

Another situation that should be considered is the introduction of new accounts from other accounts, which has established reputation in the community.

[herrehesse]

As long as they can proof (KYC) who they are, and explain why a different account is used. Perhaps we should allow someone after community consideration.

[herrehesse]

@raghavrmadya @dkkapur @kevzak

I would like to know your thoughts and opinions on this, as we currently face another major increase in new accounts and tens of PiB's of datacap requests with zero reputation or trust.

[ipollo00]

I'm afraid I disagree with this proposal.
Notaries and the community should focus on the background and description of the data request and whether the status of sealing is the same as the description.

1. The length of time of the account cannot be the criterion for judging whether it is credible, which will increase the barriers for external personnel to join;
2. People with "old" accounts may also be "untrustworthy". In this case, what reason should be used to judge?
3. Adding this restriction will make the review process in the community too complicated. The current rules are just taking shape, and that's a good sign, so let's keep it up.
4. (In my opinion) It is the most correct way to exercise the notary's ability to review and focus the review on the DATA itself.
   I believe that a notary who understands the rules and is mature enough can judge whether the account is credible through the description in the ticket and the situation of data usage.

[herrehesse]

Thank you for sharing your perspective on the proposed proposal. While I appreciate your input, I would like to address some of the concerns you raised.

• Firstly, it is important to note that the proposed restriction on new accounts or accounts with no history of datacap usage is not solely based on the length of time the account has been active. It is primarily focused on the absence of trust and reputation within the Filecoin network. While an "old" account does not guarantee trustworthiness, it could provide a track record of activities and interactions within the network, which can be used as a basis for assessing credibility.

• Implementing a restriction on new accounts does not aim to create unnecessary barriers for external personnel to join. Instead, it seeks to ensure that the allocation of large amounts of datacap is directed towards trusted entities who have demonstrated their ability to handle and distribute datasets responsibly. This not only protects the integrity of the Filecoin network but also fosters an environment where newcomers can gradually build trust and reputation within the community.

• By implementing a more stringent process for allocating datacap to new accounts, we can encourage account holders to showcase their skills, responsibilities, and adherence to the program's rules. This provides an opportunity for new users to prove their value to the Filecoin network, gain the trust of the community and notaries, and gradually increase their datacap allocation over time.

• While focusing on the background and description of the data request is important, it is equally crucial to consider the credibility of the account making the request. Trustworthy and reputable accounts are more likely to provide accurate and reliable information about the datasets they intend to handle, which enhances the effectiveness of the notary system in reviewing and approving requests.

In conclusion, implementing a restriction on new accounts or accounts with no history of datacap usage is a measure to ensure the responsible allocation of datacap and protect the integrity of the Filecoin network. It provides a pathway for newcomers to gradually build trust and reputation within the community, while allowing the notary system to focus on reviewing both the data itself and the credibility of the accounts making the requests.

By striking a balance between these considerations, we can strengthen the overall efficiency and effectiveness of the Filecoin ecosystem.

[ipollo00]

The situation of data usage will be given real feedback and can be evaluated objectively. However, "account credibility" is always a subjective indicator, and subjective indicators cannot be defined as rules.
Besides, Take a step back.
I think your proposal can be kept until at least a period of implementation of the results we have just achieved (#925) to see if it is necessary to raise it.
Proposing it at this point will undoubtedly increase everyone's pressure and complexity.

[kernelogic]

GitHub is censored in China. I think only experienced IT professionals (maybe) have an account created long time ago, not all. Might better off validating age of an WeChat account than GitHub.

[Tom-OriginStorage]

I don't think this is a good plan,

Github accounts are only available to technical personnel, while others may not have this account. However, applying for LDN is not a matter for technical personnel in enterprise companies, and the account itself is not the core criterion for evaluation. This will only deter companies and people who want to enter. The ecology of FIL+needs to be injected with fresh blood, rather than just flowing between us enterprises and these people, so the road can only become narrower and narrower

[herrehesse]

"This will only deter companies and people who want to enter." Why? You do not need datacap to start using Filecoin?
"The ecology of FIL+needs to be injected with fresh blood" No we need experienced members who know what they are doing.
"the road can only become narrower and narrower" It should, giving away a few EiB of abuse is more than enough.

[Tom-OriginStorage]

I have already made it very clear that the people who want to apply for LDN are not technical personnel. They do not necessarily have Github accounts. FIL+requires more real and useful data, and there is a real need for retrieval. They are willing to pay for data storage for retrieval in the future. If you do not allow them to apply now, increasing the threshold for their application means blocking them out, right? I think LDN can provide technical partners or assume retrieval and data authenticity, and make more rules at the robot level to complete the inspection. For example, if the previous 500T quota does not meet the rules, it will be temporarily determined

[liyunzhi-666]

What Fei says does happen, and if our KYC review is strong enough to determine that the client is a real person and that he represents his company, then the age of the GitHub account doesn't matter as much.

The problem right now is that even if someone has had a GitHub account for many years it's still someone who's lied, and it's happened before that #924. So I think the discussion of this proposal should focus on Gov team and Notary improving the KYC/KYB review process.

[Aaron01230]

I don't think it's a good plan either, let's focus on notaries perfecting their KYC process