Henrik Karlsson
@phokarlsson · joined 02 Sep 2011
I write essays at https://t.co/Tf2yohMyM1 Right now thinking about writing-as-thinking, relationships, self-cultivation, LLMs
17 670Followers
437Following
5 755Posts total
1.5MViews on collected posts
Últimas publicaciones
Buscando publicaciones nuevas…
Any recommendations of really good interview books?
I just finished reading the three that Martin Gayford did with David Hockney; they were fun! Another interview book, which I like even more, is the one Paul Cronin did with Werner Herzog.
3.6K views · 30 likes
· 2 reposts · 9 replies
Open on X →
which are the most labor intensive essays ever written?
4.5K views · 33 likes
· 0 reposts · 4 replies
Open on X →
@phokarlsson i think artists are on some kind of a spectrum with regards to how open they are to being buffeted around by the requirements of the work. Some seek to impose themselves. But what you describe is (for me) the kingly school of songwriting.
426 views · 4 likes
· 0 reposts · 0 replies
Open on X →
there will be a “right” way to extend the melody, add harmony to it, etc: it will tell you if you listen carefully. So it is really about letting go of ego and submitting to this structure outside of yourself.
1.4K views · 51 likes
· 1 reposts · 1 replies
Open on X →
Strangely, though, at the same time it requires you to be deeply self-attuned, because the only way to feel the latent structure of the song is to listen in to how the various possibilities resonate with you. But there is little space for choice.
1.2K views · 47 likes
· 1 reposts · 1 replies
Open on X →
It is interesting how little freedom there is in creativity if you take it seriously. If I’m writing a song, I have some freedom when I make the first one or two choices—picking the first section of the melody, say. But after that the form really demands what comes next:
9.6K views · 170 likes
· 7 reposts · 7 replies
Open on X →
A portrait of a person written purely by describing the imprints they have left in other people, the ways others are different because of their encounter.
1.4K views · 29 likes
· 0 reposts · 0 replies
Open on X →
@phokarlsson “The poet only asks to get his head into the heavens. It is the logician who seeks to get the heavens into his head. And it is his head that splits.”
Chesterton
It’s dangerous to your own health to treat people otherwise!
21 views · 2 likes
· 0 reposts · 0 replies
Open on X →
@phokarlsson (How) Will doing this benefit me/them?
6.8K views · 2 likes
· 0 reposts · 2 replies
Open on X →
@phokarlsson Frisch: Thinking that you know the other is always the end of love. (...) But it's not b/c we know the other that our love ends, it's b/c our love ended that we think the other is complete. https://t.co/Oy2RaU5JOu
8.2K views · 107 likes
· 9 reposts · 1 replies
Open on X →
treat people like they are too complex to fit in your head (link in next tweet) https://t.co/n4tiWEMoUV
577.1K views · 11K likes
· 1.5K reposts · 14 replies
Open on X →
I collected my thoughts in more detail here. https://t.co/DhuqFip3Uh
1.9K views · 8 likes
· 3 reposts · 2 replies
Open on X →
The data is from Games of Go on Disk which transcribes games from pro tournaments. I suspect that it is pretty hard to cheat on a live pro tournament, but not sure what precautions they take. https://t.co/pr8OqmxIoJ
7K views · 24 likes
· 2 reposts · 4 replies
Open on X →
There is a parallel here to cognitive apprenticeship theory, ie the idea that the reason cognitive skills are hard to learn is that we can't see them so you should externalize them for students. For humans to learn from AIs, perhaps making reasoning explicit is needed.
10.3K views · 124 likes
· 16 reposts · 4 replies
Open on X →
@phokarlsson That paper does not even once (that I can see) mention the hypothesis that better go AIs allow for easier cheating.
13.1K views · 60 likes
· 1 reposts · 2 replies
Open on X →
The significant improvements seems to have come not after May 2016 when AlphaGo was released but after October 2017 when Leela Zero was open sourced.
13K views · 107 likes
· 12 reposts · 1 replies
Open on X →
Ok, so this is interesting: it seems maybe the quality improved not after AlphaGo, but after the release of open source versions. Because they allowed players to look under the hood and understand the strategy. Just seeing the AI moves was not enough. https://t.co/8d1B1dsSzB http
23.8K views · 249 likes
· 39 reposts · 2 replies
Open on X →
I think this is the root source of the graph. https://t.co/XZ1cuiC0X4
14K views · 37 likes
· 4 reposts · 2 replies
Open on X →
"we estimated the time trends of decision quality only for the decisions that differed from the optimal AI decisions." which I take to mean that the graph above shows only the improved decision quality of moves that are not direct memorizations of moves done by AIs?
15.2K views · 48 likes
· 5 reposts · 1 replies
Open on X →
I saw the graph here, which links to original study: https://t.co/5Q6ZTiyPn1
19.4K views · 105 likes
· 8 reposts · 1 replies
Open on X →
Graph of decision quality among professional Go players. A sudden increase after AlphaGo. It is not only because they are learning from the AI. Players are suddenly inventing new moves at a faster rate too! https://t.co/sOyIcuvqDt
774.8K views · 3.2K likes
· 479 reposts · 53 replies
Open on X →
Frente a cuentas del mismo tamaño
8 publicaciones de los últimos 90 días, junto al rango de 10K–100K seguidores. llega mucho más allá de su propia audiencia, y esa audiencia responde.
Visualizaciones medianas1 388esta cuenta924mediana de 10K–100K
Alcance, %7.86%esta cuenta3.62%mediana de 10K–100K
Interacción, %2.02%esta cuenta1.52%mediana de 10K–100K
| Métrica | Esta cuenta | Mediana de 10K–100K | Proporción |
|---|---|---|---|
| Visualizaciones medianas por publicación | 1 388 | 924 | 1.50× |
| Alcance (visualizaciones ÷ seguidores) | 7.86% | 3.62% | 2.17× |
| Tasa de interacción | 2.02% | 1.52% | 1.33× |
Otras cuentas de este rango → Comparar con otra cuenta → Cómo se construyen estas referencias →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
10.3K22 Jan
7K23 Jan
1.9K
577.1K25 May
8.2K
6.8K
215 Aug
1.4K9 Aug
9.6K
1.2K
1.4K
426
4.5K
3.6K10 Aug
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
1.42%22 Jan
0.43%23 Jan
0.69%
2.17%25 May
1.47%
0.06%
9.52%5 Aug
2.08%9 Aug
1.96%
3.95%
3.84%
0.94%
0.82%
1.14%10 Aug
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes64.0%15 451 in total
Reposts8.6%2 066 in total
Replies0.5%112 in total
Quotes1.3%319 in total
Bookmarks25.7%6 211 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
Followers by day
4 Sep
Daily snapshots since 04 Sep 2026; the dashed line is the starting count.
Cuentas similares
NickRochefort
@thugthief
35.3K followers
· 476 posts
SERB
@serbonline
35.3K followers
· 2.4K posts
Tomas Davila ✓
@DavilaTomi
35.3K followers
· 21.5K posts
RTV Noord Sport ✓
@RTVNoordSport
35.3K followers
· 32.8K posts
DISTRICT MAGISTRATE (DM), Auraiya
@DMAuraiya
35.3K followers
· 10.4K posts
KritiSanon 24x7
@KritiSanonZone
35.3K followers
· 27.3K posts
Bhopal Commissioner
@bhopalcomm
35.3K followers
· 12.4K posts
Hugo Dixon
@Hugodixon
35.3K followers
· 43.8K posts