💾 Archived View for sakurina.flounder.online › 20201115r.gmi captured on 2021-12-17 at 13:26:06. Gemini links have been rewritten to link to archived content

View Raw

More Information

⬅️ Previous capture (2021-12-03)

-=-=-=-=-=-=-

2020-11-15 (replies)

re: "Is This Aggregator Idea Good?" by ew0k

I think it's a neat idea, and I wouldn't be opposed to being a consumer of some of these feeds to discover new content, but I would be hesitant to publish a feed of my own (putting aside how I would do so while hosting on flounder)

the recursive nature of this is an issue. the more popular shannon becomes, the likelier it is that everyone's feeds converge towards being the same, and at that point, it just becomes a decentralized alternative to the main CAPCOM and spacewalk instances for discovering newly posted content on gemini. that's great if you're optimizing for raw quantity of posts discovered, but not great if you're aiming towards quality and personalization.

what happens if someone six degrees away from you posts things that are deeply offensive? That's when the recursive design starts being dangerous. those posts will propagate across the entire chain up to you, and by then you will also be propagating it to anyone following you.

if shannon is a passive recursive aggregator that automatically shares things on your behalf, then you need to be extremely careful with the messaging surrounding it. phrases like "what I follow" or "what I like" just make it seem like anything that is on my shannon feed is an implicit endorsement, and if I'm not intentionally putting it there or don't at least have some level of control over the depth of recursion, I don't think that's an honest representation of what it is.

maybe an alternative route to take would be lightweight social bookmarking. if my feed was just my own posts and posts I've explicitly chosen to recommend, then it truly would be representative of "what I like". yeah, that means in aggregate you are discovering significantly fewer posts in a given day than with the recursive approach, but the signal to noise ratio will be much higher assuming you follow people with heavy overlap with your interests.

while it may seem like I'm shooting down the idea in this post, I am definitely interested in seeing where this goes!