[LIP-28] UOF Feed Space.md #58
Conversation
|
by introducing pay-to-win mechanics in your feed algo? how do you imagine that will help promote high quality content and not users with deep pockets?
as a user, why would i want to pay any attention to promoted content if i know for a fact it's gonna be inferior to what my algo can offer? regarding abuse, the only way i can imagine ad posts not being filtered out from normal content to farm revenue from 100% of the feed space is if it's indistinguishable from normal content. that's deceiving and still doesn't provide advertiser any guarantees that their content was seen by a human user and not by an ad revenue farm. i think advertisement markets would be a good addition to the protocol, but they must be managed by apps, not by algos. cutting out the middleman that establishes trust between the end user and the advertiser doesn't empower anyone, only promotes abuse. |
|
Most of these qs can be summed up in just the free market will solve the issues in case you don't want to read my rambling. "by introducing pay-to-win mechanics in your feed algo? how do you imagine that will help promote high quality content and not users with deep pockets?" This question is answered best by any blockchain if you just look at transaction fees. But, I would argue that if content is good, then it will be desired by users and included in algorithms. However, if you have good content, but no network effects then having the ability to pay to get your content propagated in the algorithm will help you achieve network effects. Obviously, this allows for undesirable content to be propagated using money; however, UOAs do not have to accept these offers as they would likely hurt their client base. As for quality, if for example a 3rd party sells space to low quality content then they will lose users and revenue. I would keep in mind that in this model there is not one algorithm that is being bid on there will be thousands + so unless you spam them all the idea of a pay-to-win mechanic isn't as powerful as you'd think imo. "as a user, why would i want to pay any attention to promoted content if i know for a fact it's gonna be inferior to what my algo can offer?" If someone is willing to pay you to incorporate content into your algorithm you may want, but not need may be able to subsidize other parts of a users social experience. Although, they don't need to opt-into anything. I don't really know or understand the logic of "cutting out the middle man promotes abuse." I have always found unnecessary middle men to be the abuse themselves; however, in no way does adding this functionality force anyone to include anything in their UOA, it simply provides the option. If it is in fact going to lead to a worse outcome then with a so called "trusted middlemen" then why not let the free market prove it yk? Side note: maybe todays Lens apps are well trusted, but would you consider apps like Facebook, Instagram, Tiktok, ext to not be abusive when they design their algorithms to keep you engaged for over 2+ hours per day, which has conclusively harmed the generation that grew up with them producing conclusive data on depressive, anxiety, and eating disorders. (In case its unclear, the point of keeping your attention is to serve you more ads). I would argue that apps like these have proven to not possess trustworthiness from at least the user perspective. I personally think that users should get to own whether or not something like an advertisement is included in their feed. Thank you so much for your comment :) (I didn't proof read this so sorry if anything comes off too blunt or was repetitive). |
|
Just to add a response to the 3rd paragraph is that users can create and run their UOA themselves hence kind of the point of a user owned algorithm token, so maxing out the sale of their feed space would be a decision they can decline to make - not the decision of some overlord company that needs to get their quarterlies up haha. |
|
Do I understand correctly @ZKJew that to achieve LIP-28, what is needed is actually to agree upon metadata standard to support feeds to be owned by the users? And when you mentioned about the metadata set by the user, you mean that the user sets what they want in their algo to be included, not the app? |
|
"Do I understand correctly @ZKJew that to achieve LIP-28, what is needed is actually to agree upon metadata standard to support feeds to be owned by the users?" Yes, there may be a more intuitive way to do it. But essentially, there needs to be as standard where posts can be inserted, and read by front ends in a way that is accessible to all parties involved. There would also need to be a standard for agent/third party auction interactions. This is also necessary for LIP-26 or having a UOA in the first place. "And when you mentioned about the metadata set by the user, you mean that the user sets what they want in their algo to be included, not the app?" "Not the app" yes and no. Potentially the app could be hired as an agent to produce the feed stored in the NFT, but likely this would be a specialized actor that is solely aligned with user. Does the user set it? Again yes and no. I would propose when the user hires an agent to generate the feed that is the user doing it not the app, but it's passive so I am not exactly expecting a user to control any settings or do anything > 1 click. |
Updated vision
title: Introducing User Owned Algorithm Feed Space
description: Feed space is the commoditization of an algorithms feed
author: <@ZKJew>
discussions-to: Lens
status: Draft
type: Lens Metadata Standard/Lens Open Algorithm Standard
created: <(2024-06-29)>
requires: LIP-26, LIP-27
Abstract
Every User-Owned Feed (UOF) will generate specific content for a user based on the posts and preferences inscribed in its metadata. This means that the content viewed by the user is determined by what is defined in the UOF's metadata on all apps that use a particular UOF (note: a user might have multiple UOF's for differenct types of apps). The Lens Algorithm Marketplace enables the trading of inclusion of posts in a users feed. However, it also introduces an opportunity to commoditize the content within the UOF itself, known as "feedspace". This refers to the ability for users to profit from the inclusion of posts in their UOFs.
Motivation
User-Owned Feeds (UOFs) have the potential to be a decisive factor distinguishing Web3 platforms from traditional social media. By commoditizing the metadata within UOFs, we can level the playing field for apps who lack the resources needed to compete in today's sophisticated algorithm-driven environment. Instead of optimizing content purely to appease algorithms, creators can concentrate on producing high-quality content to foster growth, leveraging the ability to purchase feed space within UOFs and for algorithm generators to reward content in a less clear way. This shift allows for a more equitable opportunity for all creators to thrive based on content quality rather than algorithmic manipulation.
Specification
The availability of feed space should be determined by the owner of the User-Owned Feeds' (UOF) agent, or actor that is generating the inscriptions of the users feed though an algorithm with the users delegation. An agent can determine what can or cannot be included in a users feed based on a users preferences and market conditions. A user should have the ability to override an agent through forced inclusion or exclusion on their app. An agent filling a users feedspace must workwithin the metadata standards of a UOF and can be undelegated by a user that wishes to switch agents (even if this is to themselves).
Rationale
The purpose of this LIP is to enhance users' ownership of their social media experience. User-Owned Feeds (UOFs) empower users to control the content they see in their feeds. The Lens Algorithm Marketplace facilitates the trading of these UOAs in a competitive market, aiming to generate advantages for users. Additionally, the concept of feed space enables users and creators to gain financial or social benefits through the inclusion of curated content in users' feeds.
Backwards Compatibility
No backward compatibility issues found.
Security Considerations
Needs discussion.
Copyright
Copyright and related rights waived via CC0.