I think you're halfway there. The problem is not the feed view—that's a matter the client can handle. The problem is the feed itself. It's the wrong data structure.
Around the time that the phrase "social media" started getting bandied about, the Internet went from topic-focused indexing to person-focused indexing. Makes sense, because topic indexing mirrors the real world. Anywhere you're physically present can be considered a small topic-indexed island. When you're at the grocery store or the post office, the collection of people in that space is an (ephemeral) aggregation over a topic/interest. So is a house party.
When you're at a house party, there is no person-based index that anyone who's there can consult to cross reference the other house parties that the host or another guest has been to. (And there's certainly no index that people who are not even at the party can consult.) At best, you're limited to a view that consists of the house parties where you've seen that person in the past or heard that they attended.
Like I said, topic-based rather than person-focused indexing is the way the Internet used to work. The only aggregation of all my posts across all mailing lists is the one in my mail client, on my machine. It's one not available to the general public. This is a good thing, because person-based indexes are an invasion privacy. We'll probably eventually find proof one day that they cause cancer (of the conversation).
I think people building out services are stuck in a metaphorical rut that retards their vison. As projects crop up to try to address the "surveillance" and "tracking" that the media reports Facebook is doing, these would-be pioneers keep reaching for the same wrong data structure that ultimately sabotages the intent. This happens because they look at their forebears that they're trying to displace, but Facebook, Twitter, etc. have poisoned the well of thought, so people end up recreating alternatives built on the same flaws. I look at this and see it in the same way that I see people who have only started programming in the GitHub era—and a bunch of people who predate it but have had their vision clouded by what they've been keeping in their immediate environment most recently. They cannot conceive of a way to host code repositories or any development model that doesn't work the way GitHub does. (Which happens to record activity in the same type of data structure—although it does so in a way that it's presented to anyone who asks for it, with even fewer privacy controls than Facebook or Twitter.)
You are ignoring the why, though. Person centered feeds simply make us spend more time on them. They are more engaging.
And unfortunately until people start being a lot more conscious about whether this is a good thing, companies with person centered feeds will win in the market. Twitter was going bankrupt before it introduced their recommended feed, now its not.
This will be like smoking. It's great, and cool, and also addictive. People will have to make a decision to stop doing what their emotions tell them to do. I think eventually it will happen, but smoking is still around, most people have smoked a bit in their lives. You just stop eventually because you know its bad for you.
In the end we are all vain creatures and want to compare/contrast with others. Topic-based was great to spread information. People-based is great to spread envy, which leads to commercial buying habits, advertising interest, corporate interest.
Huh? I don't know how you arrived at this conclusion, or at the conclusion that it needs to be pointed out that "They are more engaging". I don't know how pointing it out changes anything about what I wrote.
I wrote about this approx 2 years ago: https://www.colbyrussell.com/2019/02/15/what-happened-in-jan...
Around the time that the phrase "social media" started getting bandied about, the Internet went from topic-focused indexing to person-focused indexing. Makes sense, because topic indexing mirrors the real world. Anywhere you're physically present can be considered a small topic-indexed island. When you're at the grocery store or the post office, the collection of people in that space is an (ephemeral) aggregation over a topic/interest. So is a house party.
When you're at a house party, there is no person-based index that anyone who's there can consult to cross reference the other house parties that the host or another guest has been to. (And there's certainly no index that people who are not even at the party can consult.) At best, you're limited to a view that consists of the house parties where you've seen that person in the past or heard that they attended.
Like I said, topic-based rather than person-focused indexing is the way the Internet used to work. The only aggregation of all my posts across all mailing lists is the one in my mail client, on my machine. It's one not available to the general public. This is a good thing, because person-based indexes are an invasion privacy. We'll probably eventually find proof one day that they cause cancer (of the conversation).
I think people building out services are stuck in a metaphorical rut that retards their vison. As projects crop up to try to address the "surveillance" and "tracking" that the media reports Facebook is doing, these would-be pioneers keep reaching for the same wrong data structure that ultimately sabotages the intent. This happens because they look at their forebears that they're trying to displace, but Facebook, Twitter, etc. have poisoned the well of thought, so people end up recreating alternatives built on the same flaws. I look at this and see it in the same way that I see people who have only started programming in the GitHub era—and a bunch of people who predate it but have had their vision clouded by what they've been keeping in their immediate environment most recently. They cannot conceive of a way to host code repositories or any development model that doesn't work the way GitHub does. (Which happens to record activity in the same type of data structure—although it does so in a way that it's presented to anyone who asks for it, with even fewer privacy controls than Facebook or Twitter.)