Moving away from algorithmic curation

2023-04-24 15:28:04 -08:00

This was originally posted as a tweet thread back in November 2019, which is why it starts off with some suggestions about how best to use Twitter that are irrelevant now, since Twitter was killed by a dipshit billionaire with more money than sense and it took out the third-party clients in its death throes. But the rest of the thread holds up and I felt it worth resurrecting.


  • Tired: Helping Twitter refine its algorithmic profiling.
  • Wired: Switching to reverse-chronological timeline, as persistently as Twitter makes necessary.
  • Inspired (but admittedly not available to all): Switching to a good third-party client like Tweetbot or Twitterrific.

A few weeks ago, I saw a tweet from someone who’d switched to the algorithmic timeline experimentally and saw absolutely nothing about a then-current major news event that folks they followed had been tweeting about.

I still think about that.

It increasingly seems to me that the best things you can do with these services—recommendation engines, algorithmic timelines, and such—is (1) don’t use them when you can help it, and (2) lie to them at every opportunity.

Poison the well, and don’t drink from it.

I say this because we need to re-learn how to find each other, to recommend things ourselves, and to try each other’s personally-offered recommendations.

These are things that we should not give up to the control of companies, nor any other unknowable, unaccountable entity.

This also comes out of my thoughts about Twitter itself. And the degree to which social media has replaced RSS as our means of receiving fresh content.

It’s been good in some ways. Some of us have learned a lot, met new folks.

But we can’t depend on this.

We can’t depend on getting more of what we’ve expressed we want, if the algorithmic timeline can override that.

We can’t depend on discovering new things (and good ones, not bad ones) because the algorithm is unaccountable, built on profiling, and only seeking engagement.

A discovery algorithm’s job is to introduce people to things they don’t know they want or need.

How do you do this without introducing them to fascism, outrage fuel, shock content, or other trash? Without humans seeing that shit to screen it out?

How do you do this ethically?

Assuming the answer is “you can’t”, we then need to take up the mantle ourselves.

Spread positive things. Things you’ve made. Things you’ve learned. Skills, ideas, thoughts, actions.

This must include anti-fascism, the only other alternative being silent neutrality.

And we’ll need to use social media as best we can as long as we can, because of its amplifying nature, but we must also re-learn the other ways, the older ways. Online and off.

The old ways still work.

Print still works.

Person-to-person still works.

It’s gonna be hard to break dependency on social media, because of network effects and because of the addictive nature of it.

We probably need to start DMing each other email addresses, for a start.

And regularly contacting each other, Christmas-card style.

We’re going to need to make some changes in order to not keep heading down the same directions we’re currently going.

Not just “we” in the first person but “we” as in society. What “we” in the first person do must be chosen with that goal in mind.

I do hope, though, that whatever we ultimately replace social media with, it still has cats.

3 Responses to “Moving away from algorithmic curation”

  1. g Says:

    Cool captcha

  2. nisser blahman Says:

    Dipshit billionaire speaks very highly of you sir.

  3. undefined $user_ID in …/comments.php on line 74 Says:

    We probably need to start DMing each other email addresses, for a start.
    i once did that and never heard anything from them again

    ‘>alert(‘hi dont mindme’)

Leave a Reply

Do not delete the second sentence.