Plagiarism Purge #283
Comments
|
Hey thanks for the submission! Can you explain the marketplace component and how it would work? Also what content are you looking to curate? This looks like a general purpose curation market of sorts but please clarify what your aim is here. |
|
The idea is to be a general purpose component which would essentially compare media against other known works of it's type and report back on the uniqueness. I imagine it may have to actually be broken down into multiple components each for its own media type. I know it could work for written documents and code. So possibly the easiest example would be plagiarism on written documents, same as most universities use. When you turn in a paper, professors generally submit them through services which check for plagiarism. This is not a black and white decision, but generally a value return based on % found - mostly because many papers quote other papers, so of course some matches are expected, and most professors have different levels they are willing to tolerate (perhaps 20% or if its a research paper 30% or 40% since much content might be quoted). The module is not necessarily a curator, though perhaps it could be set to mod content above a certain threshold and then simply tag the rest. Say a written document is an 85% match to a known, existing, document. The component could be configured to automatically drop/hide content with >80% match. Other than that, it would tag the work with a match % in order for users to gain knowledge on content they are voting for. As an example, say I wished to create a bounty for a short story; I would like a 30 page short written as a long lost episode of Seinfeld, mostly featuring Kosmo Kramer on a field trip to Yankee stadium to visit George where he ends up giving batting lessons to Derek Jeter. People could submit their work and users would vote on which piece wins the bounty. In order to help the users vote, each submitted work would go through the component and return a value of content plagiarized to ensure that the work is indeed a unique work and not a plagiarized copy of existing Seinfeld work. It would not be responsible for determing winners, or whether or not a piece is plagiarized, it would just be a supporting component designed to give users the information they may need to vote the most deserving piece to the top. |
|
Gotcha, the reason I brought up the curation market aspect is because to be a district it needs to be a marketplace. I could see this working as a plagiarism bounty platform. Where people could earn bounties or possibly even just use a TCR to give the financial incentive to curate and those people would manually check for plagiarism before making a vote. The issue here is that a TCR is objective to the individual rather than subjective and "correct data" about it being plagiarized. The token distribution would need to be with identified parties or groups to work right. I believe Messari is doing this with their internal group where a TCR is being used but they have a reputation to keep so they do try to make their curation efforts "correct" to the best of their ability. So to make your idea a marketplace I think this would be the best strategy, your next goal would be to figure out how you would distribute a token and to whom, to make the data as "correct" as possible. I think an academic group or students would be a great place to start a case study on a testnet and see how it goes. I'm labeling Curation market because this isn't so much a marketplace on it's own but a great potential TCR and academic based or journalism focused bounty marketplace. Would love your thoughts here, if I'm missing other community marketplace components. |
|
I understand. Thanks for thinking through this with me. I agree that it seems a TCR is a good approach here. Perhaps it could be set such that users wishing to utilize such a feature could place a bounty on the results. In this way it could offer rewards for those who participate where the reward is split between the "winners" in a way similar to Augur. In this way, too, users could simply determine how accurate they wish their result to be based on the bounty set. If the user sets too low of a bounty, they may not have any takers to curate the results. If they place a larger bounty, they will have many takers all placing predictions. One hiccup I can think of right off the bat is how to determine whether users are actually curating, or simply "placing wagers". |
|
That could be an issue for sure. This can however be different for every TCR. There has to be some type of system in place for reputation, ratings, token controlled access, or other incentive system. This ensures the data is more accurate and community standards can emerge. Especially if there is an added layer within governance surrounding a community based mission, vision, or bylaws. This becomes more about building community and culture than anything else. |
|
Perhaps user "whitelists" could become a thing. You could even have curators of curators; like pihole lists, where users can use lists maintained by others and combine them for improved lists. So as the community members gained rep and trust, other members could maintain lists of their most trusted members per TCR. Users could then use those whitelists to limit curators to specific members. As members gained rep, they could increase their fees - thus the marketplace would create benefits for users to work harder and gain more trust. Then if you wish to use a TCR you could place limits on who you wish to trust for curating purposes. Perhaps curating 'members' would actually be teams on the backend similar to mining pools. Where users do the work on the backend and the group splits the bounties based on who does the work. That 'member' would then quickly gain rep since it would be a conglomerate of users working around the clock and producing more than a single user alone. |
|
Ultimately I thin group based TCRs are a great road to go imo. It solves a lot of issues with TCRs for sure. I can't wait until after Meme Factory is running for a while so we can test some of this out. There is no reason we couldn't experiment with this idea now with an Aragon entity. You could have that entity holding the DANK and the members can choose whatever model they want to who get's the ability to vote. |
|
I sent a reward for this proposal as well. |
|
I'm not sure if you're aware but we recently launched a new bounty to migrate your proposals to the actual District Registry: https://registry.district0x.io/ We're replacing the old voting app with the registry. Let me know if you need help but I would love to see you migrate this over so you can claim your DNT. It does take a 10,000 DNT submit to submit your proposal but this gives you an extra 2000 you can stake in the registry beyond the deposit amount. Hit me up on Telegram or Discord if you need help or have questions. Telegram: https://t.me/district0x/75217 PS, please excuse the canned response. I am encouraging everyone here to start migrating so they can claim their 12000 DNT. |
Name:
PlagiarPurger
Purpose:
A district module to provide a decentralized market-style platform, with automated tools as possible to help voters, to prevent plagiarized content.
Description:
The district would be built as a module, or plugin, to be used by other districts to facilitate duplicate or plagiarized content such as re-uploading previously uploaded content as one's own and attempting to make profit off of other's works. Tools such as Sherlock Plagiarism Detector and reverse image search results could be integrated in order to help voters by providing additional information on an automated basis.
Ethereum Address:
0xE8332043e54A2470e148f0c1ac0AF188d9D46524
The text was updated successfully, but these errors were encountered: