Modification: Improving the Guidelines of "Valuable Data" on the Filecoin+ Network

[herrehesse]

I am submitting a proposal for a modification to the current guidelines regarding what data is considered valuable and eligible for datacap LDN requests.

In my opinion, if we do not limit our view of what quality data is and continue to rely on the opinions of people who simply want to obtain datacap, we will never move towards the quality phase. It is important that we prioritise the quality, scientific value, and public interest of the datasets being stored on the Filecoin network, rather than just focusing on growth.

Issue Description

The current description of valuable data is extremely vague, which leaves room for potential abuse of sets that have no actual value. This can lead to fraud and a decline in data quality, which is contrary to the goals we want to achieve, which is to make the Filecoin+ program a hub for valuable data storage and increase the value of the network by offering high-quality data sets.

Impact

The current approach of allowing any data, regardless of its value (due to opinions), into the Filecoin+ ecosystem can lead to a decline in data quality and misuse of the program. This undermines the reputation of the program.

Proposed Solution(s)

I propose that we create a clear and reasonable description of valuable data, and only allow data that aligns with Filecoin's core values of storing "world's most valuable data" to receive datacap. All data that does not meet this criteria should be subject to paid / regular deals, rather than FIL+ deals.

During the quality phase of the Filecoin+ program, we should focus on the quality and scientific value of the datasets, proper indexation, easy retrieval, usage, and visibility on the Filecoin network.

A perfect example, and a potential candidate to limit datacap requests to, is the website https://openpanda.io/.

Timeline

The implementation of this modification should begin as soon as possible to prevent further decline in data quality and misuse of the program.

Technical dependencies

The implementation of this modification may require changes to the existing governance framework the Filecoin+ program. The team should assess and evaluate the feasibility of the proposed changes.

Risks and mitigations

The proposed modification may face resistance from some members of the community who may disagree with the proposed changes. To mitigate this risk, the team should engage in extensive community outreach and education to explain the rationale behind the proposed changes.

[dkkapur]

This was definitely part of feedback we captured in the last months and resulted in us proposing / crafting a definition with various members of the community. I'm personally supportive of this line of thinking / philosophy and would love to further the conversation.

Let's start with this baseline and further specify from there? What are your specific thoughts on this:

Quality data is all content that meets local regulatory requirements and

• the data owner wants to see on the network, including private/encrypted data
• or is open and retrievable
• or demonstrates proof of concept or utility of the network, such as efforts to improve onboarding

pulled from https://filplus.storage/ and https://medium.com/filecoin-plus/announcing-the-quality-phase-for-filecoin-plus-2890a9797456

[herrehesse]

I disagree with the points mentioned above as they appear to be overly vague and remain open to potential abuse.

Instead, I propose that we strive for a 100% quality phase that showcases the datasets available on openpanda. This would involve demonstrating exceptional retrieval capabilities, efficient indexing, rapid availability, and aesthetically appealing dashboards. By emphasizing these aspects, we can exemplify the Filecoin network's true potential to large clients.

My thoughts below:

• the data owner wants to see on the network, including private/encrypted data
  The determination of value should not rest with the data owner. Instead, the community as a whole should have the authority to assess the value of data, considering they bear the burden of the multiplier. Given that the multiplier represents a cost to the community, it is only fair that they hold the power to evaluate and determine the value of the data being stored. This approach ensures a collective decision-making process that aligns with the interests and needs of the broader Filecoin community.

• or is open and retrievable
  I fully support the concept of "open" and suggest lowering the "encrypted/private" FIL-E multiplier accordingly. Additionally, I wholeheartedly agree that retrievability is a critical factor that requires immediate attention. It is disheartening to witness a significant portion of data stored with datacap being non-retrievable. Many storage providers neglect the responsibility of storing an unsealed copy. Unfortunately, the absence of removal tools for storage providers exempts them from facing repercussions and provides little incentive for adherence to the established rules.
  Furthermore, it is essential to acknowledge that the current retrieval process is far from perfect, necessitating immediate development efforts to enhance its functionality. By prioritising the improvement of retrieval capabilities, we can effectively progress towards the desired quality phase.

• or demonstrates proof of concept or utility of the network, such as efforts to improve onboarding
  I am not sure what you are trying to say here, can you give me an example? (do you mean Estuary?)

Currently, it is evident that the utilisation of FIL+ is primarily driven by a growth-focused mindset, with little regard for the actual data being stored. This approach undermines the true purpose of the network and program. It is imperative that we shift away from this mentality as soon as possible and prioritise the quality and value of the data.

It is crucial to maintain the understanding that the Filecoin network allows anyone and everyone to store data without any restrictions or limitations. However, it is important to emphasize that random data should not be granted datacap. This valuable resource should be treated with utmost respect and not merely utilized as a means to maximize profit within mining operations, as it is currently being done by some Storage Providers.

[The-Wayvy]

I suggest phasing out Filecoin Plus.

[The-Wayvy]

Permissioned, decentralized storage is an oxymoron.