Conversation

Why does social media make us memoryless? One thesis is: each day we see the most important thing ever. But then yesterday’s most important thing is less important. And last week’s viral post? Already forgotten. Perhaps we want more consistency and less novelty.

164

1,156

Balaji Srinivasan

@balajis

Dec 22, 2021

I first saw the Feiler Faster thesis in a piece by

@kausmickey

20+ years ago. More recently,

@Noahpinion

and

@micsolana

have written on the phenomenon. But I feel it’s been going on for longer than that, perhaps accelerating with media decentralization.

slate.com

Faster Politics

My friend Bruce Feiler had a thought that he said I could steal. It's this:

Here’s

’s recent piece on the topic.

Quote Tweet

Mike Solana

@micsolana

· Dec 15, 2021

america has amnesia. nuclear power, space, and who remembers when the virus came from china? memes, by way of viral misinformation, can be dangerous, but nothing is so dangerous as the antimemespace consuming our reality. piratewires.com/p/variant-xi

Show this thread

Balaji Srinivasan

@balajis

Dec 22, 2021

If we are interested in learning and building and focusing, the complete opposite of social media could be mnemonic media. See

@michael_nielsen

and

@andy_matuschak

’s phenomenal results below. Would love to see an update if it’s available.

Quote Tweet

Michael is back in San Francisco, soon in Santa Fe

@michael_nielsen

· Jun 26, 2019

Here's the demonstrated retention versus the number of times each question in the essay has been reviewed. It takes a little work to unpack, but the basic point is: demonstrated retention appears to be increasing exponentially with the number of reviews.

Show this thread

Andy Matuschak

@andy_matuschak

Dec 22, 2021

Current versions of the data MN showed look basically the same, except even more steeply exponential (i.e. actually more efficient now). Have been working to understand the dynamics of review more precisely; lots of confusing/interesting details there not ready for publication.

Balaji Srinivasan

@balajis

Dec 22, 2021

Are there any specific domains where it works best? Is this all quantum country data?

Replying to

and

This is all QC data. I've run experiments in a variety of domains now with a variety of reader backgrounds. Most worked much less well than QC (speaking generally). I think I understand a big part of why that is; working on it. See patreon.com/posts/55309960 and siblings for more.

patreon.com

Revamping the mnemonic medium around reader control | Andy Matuschak on Patreon

Join Andy Matuschak on Patreon to get access to this post and more benefits.

8:00 PM · Dec 22, 2021TweetDeck