iiinova — where your digital self keeps evolving

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 07/09/2026

"AI" is everywhere nowadays, but 30 years ago similar hype was about everything "computer". I stumbled at this Italian dictionary from 1986 proudly stating on its cover that words in it were sorted by a computer 😁

For the first time, 20,000 Italian words computer-organized into semantic groups and fields of reference.

260

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 31/07/2026

Reminder: two weeks are left to apply for the PhD and postdoc positions in my lab. See more info here www.aesthetics.mpg.de/en/career/jo...

Postdoc and PhD positions at MPIAE in Frankfurt

13535

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 29/06/2026

📣 I'm hiring! Two positions (PhD student & Postdoc) in my ERC project Macroevolution of European Literature. Let's study the cultural evolution of literature using massive data 📚📈 📍Frankfurt | Apply by August 15 Details: www.aesthetics.mpg.de/en/career/jo... Please repost to help spread the word!

Job announcement of two positions at the Max Planck Institute for Empirical Aesthetics in Frankfurt: postdoc (Macroevolution of European Literature) and PhD student (Computational Literary Studies). Successful candidates will be affiliated with the research group Cultural Evolution of the Arts.

08082

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/05/2026

Cultural evolution of the arts is a new academic community, and in this paper we try characterizing it: its object of study, its unique challenges, and its sub-areas (shown below ⬇️). (3/3)

020

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/05/2026

There is something unusual about a "scientist of artistic evolution", but we noticed that, in recent years, increasingly many scholars become such "cultural evolutionists of the arts" – despite this unusual academic profile. We explain this in more detail in the opening paragraph: (2/3)

140

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 30/04/2026

My group at the MPI for Empirical Aesthetics now has a name: Cultural Evolution of the Arts! You can read a bit more about the idea behind the group at the group's web page: www.aesthetics.mpg.de/en/research/... And here's the obligatory door nameplate :)

Door name plate with the name of my research group: Cultural Evolution of the Arts

2419

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 11/03/2026

Yay! This week I started working at the MPI for Empirical Aesthetics in Frankfurt as a group leader. Lots of work ahead, but super exciting kind of work 🔥

Happy me in front of the MPI for Empirical Aesthetics

9945

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 18/12/2025

I got wonderful news: I was granted a position of Max Planck independent group leader at the MPI for Empirical Aesthetics ⬇️ (@ae.mpg.de) in Frankfurt, which I'll join in early 2026. It's a huge honor: being trusted with academic freedom offered by MPG. My group will study the evolution of arts. 1/3

The building of the Max Planck Institute for Empirical Aesthetics in Frankfurt

1212413

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/12/2025

Here is the first page of my PhD thesis published in 2018. There's eight. And I'm not alone in writing like this. Styles are different, you know

100

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 25/11/2025

AI was so close to getting it right...

Fake Steven Pinker, drawn by an unknown AI model

040

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 21/11/2025

📚 MajinBook: an impressive new dataset by Antoine Mazières & @tpoibeau.bsky.social. Metadata for 500,000 books from Library Genesis & GoodReads. Mostly books from the last 100 years or so, see the plot ⬇️ arxiv.org/abs/2511.11412

Temporal distribution of books in the corpus

0102

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 15/09/2025

My copy of the Oxford Handbook of Cultural Evolution has arrived! Looks gorgeous – and massive. And somewhere on page 554 (which is roughly the middle of the book 😅) you can find my chapter. Big thanks to the editors for organizing this!

Front cover of the Oxford Handbook of Cultural Evolution

The chapter "Evolution of Modern Literature and Film" in this volume

0241

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 25/08/2025

I happened to accidentally find this edited volume on quantitative approaches to literature, from 1969. Never heard of it - even though it was co-edited by one of the big figures in narratology: Lubomir Doležel (another surprise). The minimalist plots here are ⚡️

0171

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 22/03/2025

This week I had a pleasure to present my research at Huygens Institute in Amsterdam. What a wonderful place, really! I preached my usual gospel about the cultural evolution of the arts. Mostly – about the decomposition analysis of literature (see royalsocietypublishing.org/doi/10.1098/...)

2281

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

The #OpenAccess paper in @royalsociety.org Proceedings B can be found here: doi.org/10.1098/rspb... Also, shoutout to my co-author @babeheim.bsky.social ! (8/8)

Sobchuk O. & Beheim B. (2025) "Does literature evolve one funeral at a time?" Proceedings B.

0101

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

Our finding: cohort turnover (🟦+🟥) is stronger than individual change (🟨): i.e writers don't change their topics much mid-career. This aligns with previous studies by @tedunderwood.me, @stephenvaisey.com, @oliviermorin.bsky.social, @wenyishang.bsky.social, @kkiley.bsky.social & F.Moretti. But… (6/8)

1224

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

Next, we borrowed a method from evolutionary ecology: decomposition analysis. It allows “dissecting” a trendline of a topic into underlying forces. We looked at the impact of new authors starting their career (🟦), established authors retiring (🟥), & authors changing their topics mid-career (🟨) (5/8)

1242

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

We used a corpus of 23,000 books by 6000 authors. Then we algorithmically extracted 200 major topics that these authors were writing on. Such topics have clear temporal trajectories ⬇️ (4/8)

The historical trajectories of three LDA topics in literature

1130

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

So, we considered two possibilities. (1) Individuals – say, writers – always learn, changing their topics, e.g. in response to significant events, like the moon landing. Or (2) authors don’t change much over their careers, and the topics of their books are shaped during their formative years (3/8)

Two possible scenarios of why authors change the topics of their books: due to a disruptive event (e.g., war) or due to demographic turnover

1111

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

Max Planck (this guy ⬇️), who wasn't only interested in quantum physics, once wondered: “Does science advance one funeral at a time?” Literary scholar Franco Moretti also hypothesized that the rise and fall of book genres (e.g. gothic novel) is driven by the change of generations of writers (2/8)

190

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 05/02/2025

Change over time is often depicted as a trendline. But what does shape a trendline? Which forces? Our new paper presents a method allowing to “decompose” trendlines into constituent forces. Also, we tackle an old puzzle: Does culture change “one funeral at a time”? 🧵(1/8) doi.org/10.1098/rspb...

a schematic depiction of a trend line and several causal forces that give it its shape

211446

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 06/01/2025

Holy crap. I took a peek at what's happening on former Twitter. Is Musk drunk on power, or what? I was resisting removing my Twitter account, but, yikes, staying there isn't different from having an account on Truth Social – maybe even worse.

A visual representation of Elon Musk's ambitions to become Hitler

221

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 29/12/2024

Why do some academic journals still have the section "Author Contributions" even when the paper is single-authored? Maybe this section is a good place to introduce one's ghostwriters? 😅

A screenshot of author contributions section from a random single-authored paper

061

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 30/08/2024

Hi everyone! You may know that sometimes I collect donations to buy things ⬇️ for the Kremenets battalion 🇺🇦, which is protecting the eastern front of Ukraine & which includes my relatives & schoolmates. Now they need help again (1/3)

What we bought for your donations: medical supplies, drones, and even a car

1109

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 14/06/2024

Strange reality of collective research: colleagues from our institute just published a really cool paper in Nature. But because the paper has 90+ authors (well, genetics...), no author names – not even the first author's name – were put on the front page... I'm not sure what to think about it

Front page of this paper https://www.nature.com/articles/s41586-024-07546-2

220

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

Our findings? The vast majority of actual genres (🔵) exhibit first-mover properties, while mixed-genre samples (🟡) do not. So *yes*, there does seem to be first-mover advantage in music /end

A: Density curves of the relationship between year of release and songs’ success in 100 random genres and in the samples from the mixed-genre dataset. B: Blue dots: the relationship between time and total playcount for each of the 10 preregistered genres. Yellow distributions: the same relationships, but for the samples drawn from the mixed-genre corpus of songs (1000 samples per each actual genre). Panel C shows the same, but for 100 randomly picked genres. On panels B and C, full circles (●) on y-axis mean that the first-mover hypothesis is confirmed, empty circles (∘)—that it is not

040

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

Another key piece was to figure out how to debias our data. For this, we compared actual genres to mixed-genre samples. Our guess was: actual genres must show first-mover properties, while similarly structured mixed-genre samples must not /5

Our approach to debiasing music data: comparing actual genres to mixed-genre samples. The figure shows several plots illustrating how it's done.

120

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

We picked 110 genres – new & old, big & small. 100 were chosen randomly, while 10 were intentionally selected (& pre-registered) genres that we knew well & could confirm that all the bearers of the genre tag are actual members of this genre (secondary data – gotta be careful!) /4

A: Density curves showing the sizes of genre populations in three datasets used in our analysis. While the median population size of our 10 preregistered genres is somewhat larger (144 artists per genre), the median population of genre sizes in the random sample of 100 genres (77 artists) is almost the same as the median across all the genres on Every Noise at Once (80 artists). B: Adoption curves of 100 random music genres. C: Adoption curves of 10 preregistered music genres

120

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

To answer this, we needed several pieces of the puzzle. One key challenge was getting high-resolution genre tags (a rare thing!). We found them on everynoise.com, which has TONS of detailed genres – not just “rock” or “pop”, but “Ukrainian rock” or “country pop” /3

120

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

Take "emo rap" as an example. It's a new genre that rose to prominence in the 2010s, with depressive superstars Lil Peep ⬇️, XXXTenacion, and Juice WRLD as its central figures. Could it be that the fame of emo-rappers is related to time they joined this genre? /2

Lil Peep, photo from: https://www.rollingstone.com/music/music-news/lil-peep-cause-of-death-revealed-202399/

130

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 17/05/2024

Our new paper w/ @masonyoungblood.bsky.social & @oliviermorin.bsky.social. Are early members of music genres more popular than late adopters? Or: Is there first-mover advantage in music? We tested this question using ~1 mil songs from Spotify 🧵/1 epjdatascience.springeropen.com/articles/10....

Abstract of the paper "First-mover advantage in music": Why do some songs and musicians become successful while others do not? We show that one of the reasons may be the “first-mover advantage”: artists that stand at the foundation of new music genres tend to be more successful than those who join these genres later on. To test this hypothesis, we have analyzed a massive dataset of over 920,000 songs, including 110 music genres: 10 chosen intentionally and preregistered, and 100 chosen randomly. For this, we collected the data from two music services: Spotify, which provides detailed information about songs’ success (the precise number of times each song was listened to), and Every Noise at Once, which provides detailed genre tags for musicians. 91 genres, out of 110, show the first-mover advantage—clearly suggesting that it is an important mechanism in music success and evolution.

23221

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 22/03/2024

What are the best ways of clustering literary works by genre? In our new paper, @artjomshl.bsky.social & I compare 291 combinations of algorithms to find this out. Many findings are surprising: say, k of topics in LDA doesn't matter, or Euclidean distance is terrible 🤷‍♂️ www.nature.com/articles/s41...

Four steps of the analysis. The workflow includes two loops. Big loop goes through various combinations of thematic foregrounding (Step 1a), feature type (1b), and distance metric (1c). For each such combination, a smaller loop is run: it randomly draws a genre-stratified sample of 120 novels (Step 2), clusters the novels using the Ward algorithm (Step 3), and validates the clusters on the dendrogram using the Adjusted Rand Index (Step 4). As a result, each combination receives an ARI score: a score of its performance in detecting genres.

Raw distributions of ARI scores for all the combinations of thematic foregrounding, feature type, and distance metric. Boxplots are colored by feature type.

23223

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 18/03/2024

On April 2nd I'll be giving a talk about my recent work for the #digitalhumanities center at Potsdam University – and online – so feel free to join if interested. (Note that registration is required.) www.uni-potsdam.de/de/digital-h...

Poster for my talk at Potsdam University. Title: The Driving Forces of Literary Evolution. April 2, 6:30 pm.

0155

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 06/03/2024

Rewatched Dune, part 1. Why are bad guys in movies so often... bald? All of the Harkonnens (top left ⬇️) have perfectly shiny heads. But so do Voldemort, Bane, orcs, and so many others. Are we... afraid of the bald? 😅

340

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 23/02/2024

A very cool new preprint – about trees, art, and trees in art. Trees are governed by fractal patterns – hypothesized by Leonardo da Vinci himself. But are trees in the art following the same fractal patterns as trees in the real world? That's the question arxiv.org/abs/2402.13520

075

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 08/01/2024

Finally, it was a pleasant surprise to find an actual paper copy of the journal in my mailbox. Journals don't often do this, in my experience. (3/3)

040

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 08/01/2024

📄 Why are fictional monsters often hybrids – that is, consisting of several parts of various creatures? And why are these hybrids so aggressive-looking – with sharp claws & teeth? @oliviermorin.bsky.social & I explore these questions in a new paper (1/3) read.dukeupress.edu/poetics-toda...

Cthulhu image from https://lovecraft.fandom.com/de/wiki/Cthulhu

22915

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 03/01/2024

In the past, many of you helped massively – by donating money, for which we bought medical supplies, satellite terminals, a drone, and even a car for the battalion! I really hope that you could help again. The war isn’t in the headlines, but the stakes are as high as before (2/3)

This was bought for the donations in the past: 2 Starlink terminals, a drone, 2 night vision devices, 2 Motorola radios, many medical supplies

This is a car we bought for the donations! It was bought almost a year ago, and still serves well at the frontline

130

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 03/01/2024

🇺🇦 You may know that sometimes I collect donations for the Kremenets battalion: battalion from my hometown ⬇️, which is defending the eastern border of Ukraine. The battalion includes my relatives & schoolmates. Now they need more funds – for a rather expensive drone (1/3)

A photo of Kremenets from above. By A. Lukashchuk, Wikimedia Commons.

11311

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 14/12/2023

Out of curiosity, I checked eLife's accounting (it's public), and the numbers there are insane. For example, they pay £ 2,2 million (!) as salaries. Just let it sink. To whom? Probably to editors and mangers – but I'd like to know more. elifesciences.org/inside-elife...

020

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 13/12/2023

There's a nuance. They require "final, peer-reviewed manuscript". This contradicts the policies of many journals, who stress that preprint is an article before any of the journal's involvement (review and proofreading). In reality, ofc, one can put the article after peer-review

110

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 13/12/2023

Journal fees for open access are becoming obscene. But what are we, scientists, paying for? I made a simple plot with journals in my research area(s). Clearly, we are paying for prestige: a shockingly clean correlation between the impact factor and journal fees

A plot showing prestige (impact factor) on vertical axis and fees charged by journals on horizontal axis.

21303179

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 11/12/2023

@andreskarjus.bsky.social captured me MCing at the CHR conference. What about? DH/CH people often use trendlines, but we rarely explain their underlying mechanisms. @babeheim.bsky.social and I use decomposition analysis to "dissect" trendlines into pieces and explain the forces behind each piece

A little silhouetto of a man, talking about science #1

A little silhouetto of a man, talking about science #2

090

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 08/12/2023

I truly enjoyed Computational Humanities 2023 – huge thanks to all the organizers! Looking forward to the next one, which will be in Aarhus. Everyone who hasn’t joined us this year – come next time!

0151

Oleg Sobchuk 🇺🇦 @sobchuk.bsky.social · 06/12/2023

Digital Richard McElreath gave a plenary today at the CHR conference about the “analog” tools for the digital humanities. My favorite quote: “Big data can be big wrong” 😁

Photo of Richard McElreath plenary at the CHR conference

0110