Sign in

Jörg Lehmann

@jrglmn.bsky.social
113 followers 36 following 33 posts

Digital humanism | machine learning | digital cultural heritage | Berlin State Library | „Name a bias – we have it!“

PostsRepliesMedia
Jörg Lehmann @jrglmn.bsky.social · 26/12/2025
That’s my favourite dedication I’ve seen in 2025 …
„Dedicated to Donald Lau, fortune cookie writer for 30 years“
100
Reposted by Jörg Lehmann
Stabi Berlin @stabiberlin.bsky.social · 19/12/2025
Seit wann gibt es das Spiel "Stadt, Land, Fluss"? Antworten auf diese Frage hat Dr. Tarek Saier gefunden - mithilfe von Datensätzen, die bei uns an der #StabiBerlin publiziert wurden und dem klugen Einsatz von #KI. Mehr zu seinem Rechercheprojekt lest ihr hier 👉 blog.sbb.berlin/stadt_land_f...
blog.sbb.berlin
Seit wann gibt es „Stadt, Land, Fluss?“ Mit Stabi-Daten und KI Antworten finden - SBB aktuell
Das Spiel „Stadt, Land, Fluss“ lässt sich in seiner heutigen Form auf die 1930er Jahre datieren; der Name ist spätestens 1937 nachweisbar. Das zugrunde liegende Spielprinzip – Antworten zu mehreren Ka...
041
Jörg Lehmann @jrglmn.bsky.social · 17/12/2025
Two results from the project "Human.Machine.Culture" at @stabiberlin.bsky.social published Guidelines for the Documentation of Ethical, Legal and Social Issues (ELSI) in Cultural Data doi.org/10.5281/zeno... Guidelines for the Publication of Cultural Data for AI Research doi.org/10.5281/zeno...
101
Jörg Lehmann @jrglmn.bsky.social · 12/12/2025
A note on the cost-benefit ratio of #chatgpt - a user of one of our datasets (doi.org/10.5281/zeno...) calculated the cost of what the use of GPT-5 would have been if the 5 million pages provided by us would have been processed using #OpenAI 's service: $1.2 million (gasp!) @stabiberlin.bsky.social
doi.org
Berlin State Library (2023). Fulltexts of the Digitized Collections of the Berlin State Library (SBB)
The motivation for creating this dataset was to enable research on the basis of fulltexts which are available in a cultural heritage institution on a large scale. Libraries such as the Berlin State Li...
101
Jörg Lehmann @jrglmn.bsky.social · 11/12/2025
In the poster session at #CHR2025 starting at 5pm, our dear colleague Michał Bubula will present our short paper entitled "How Scalable is Quality Assessment of Text Recognition?" anthology.ach.org/volumes/vol0... Go there and talk to him since we aim to provide your high-quality research data!
anthology.ach.org
How Scalable is Quality Assessment of Text Recognition? A Combination of Ground Truth and Confidence Scores
011
Reposted by Jörg Lehmann
CRETA e.V. @cretaverein.bsky.social · 10/12/2025
Morgen, am Donnerstag den 11.12.2025 sind Nils Reiter & Janis Pagel vom CRETA-Verein bei der Poster-Session der CHR2025 Conference in Luxemburg anzutreffen. Sie stellen einen Ansatz zum Thema "Automatic detection and classification of literary character properties in German narratives" vor! #CHR2025
023
Jörg Lehmann @jrglmn.bsky.social · 10/12/2025
#FF2025 @danielvanstrien.bsky.social has published his workshop slides on #opensource AI for #GLAM 's danielvanstrien.xyz/slides.html I consider his readworthy contribution as the AI complement to my reflections on open data (and their shades) published here: mmk.sbb.berlin/2024/06/21/o...
danielvanstrien.xyz
Slides – Daniel van Strien
020
Reposted by Jörg Lehmann
Christof S. 🇪🇺 @christof.fedihum.org.ap.brid.gy · 20/11/2025
Very cool! The Proceedings for #CHR2025 have already been published, now at the new and slick #ACH, the "Anthology of Computers and the Humanities", developed and maintained by ACH, the "Association for Computers and the Humanities". As an example, you can […] [Original post on fedihum.org]
Screenshot of the landing page of one article in the ACH, in simple tones of gray.
1512
Reposted by Jörg Lehmann
Daniel van Strien @danielvanstrien.bsky.social · 09/12/2025
Just posted my slides from the AI4LAM #FF2025 workshop on open source AI for GLAMs. Probably slides on their own aren't that useful, but they do feature one of my growing collection of libraries-and-AI memes, so there's that danielvanstrien.xyz/slides.html
Here's alt text for the meme:

Alt text: "Flex Tape meme format. Top panel: Phil Swift (labeled 'Library Systems Vendor') aggressively spraying water representing 'Outdated systems, metadata issues, disjointed search and complex user needs.' Bottom panel: A hand slapping Flex Tape underwater, with the tape labeled 'AI-powered chat interface.'"

This captures the joke that vendors are positioning AI chat as a quick fix for deep-seated library infrastructure problems—a bit like slapping tape on a leak rather than fixing the plumbing.
273
Jörg Lehmann @jrglmn.bsky.social · 08/12/2025
#FF2025 pickings: This year has been extremely productive with regard to the AI & commons debate, as well as in view of the publication of open, public domain datasets. Paul Keller & Europeana Foundation: Publishing cultural heritage data in the age of AI, Dec 2025 openfuture.eu/publication/...
openfuture.eu
Impulse paper: Publishing cultural heritage data in the age of AI – Open Future
This paper proposes a framework to help cultural heritage institutions decide when and how to share collection data for AI training, balancing open access with managing large-scale AI reuse aligned wi...
200
Jörg Lehmann @jrglmn.bsky.social · 02/04/2025
A colleague from the UK and I bring in our two cents on a highly divisive issue, written from the perspective of European CHIs and research libraries. A Position Paper on AI and Copyrights in Cultural Heritage and Research (EU and UK) doi.org/10.5334/johd.290 #genAI #commons #openness
doi.org
A Position Paper on AI and Copyrights in Cultural Heritage and Research (EU and UK) | Journal of Open Humanities Data
110
Jörg Lehmann @jrglmn.bsky.social · 14/03/2024
Sigh. This topic #openness, intellectual property rights #IPR, #genAI is getting really complicated for #GLAM institutions. Wrote a blogpost to chart what's up in the EU and what we currently need: mmk.sbb.berlin/2024/03/13/o... Currently, there is no technical solution to implement an opt out...
mmk.sbb.berlin
Orientation in Turbulent Times – Mensch.Maschine.Kultur
140
Jörg Lehmann @jrglmn.bsky.social · 26/01/2024
New post on the "power hungry magic" of contemporary artificial intelligence published on the blog of the HumanMachineCulture project: Energy, CO2 intensity and sustainability as mostly overlooked issues in the deployment of GPTs. mmk.sbb.berlin/2024/01/26/p... #LLMs #ChatGPT #metaverse
mmk.sbb.berlin
Power Hungry Magic – Mensch.Maschine.Kultur
010
Jörg Lehmann @jrglmn.bsky.social · 18/01/2024
Copyright is but one indicator of the value of digital texts, which have gone through a quality filter called ‚publishing houses‘. The same can apply to texts in the public domain, and GLAM institutions should reflect on this. Texts in open access are as well valuable, see doi.org/10.54900/zg9...
doi.org
Large Language Publishing
The New York Times ushered in the New Year with a lawsuit against OpenAI and Microsoft. The paper covered the suit, fittingly, as a major business story. OpenAI and its Microsoft patron had, according...
000
Jörg Lehmann @jrglmn.bsky.social · 14/01/2024
Dutch National Library restricts access for commercial AI Blocking is done via the robots.txt. Crawlers are thus excluded regardless of copyright. Consequently, public domain material is not accessible to the crawlers. Restriction is selective: Googlebot-image, dataforseo.com, GPTBot, ChatGPT-User
kb.nl
KB restricts access to collections for training commercial AI
The KB is taking precautions to prevent the crawling of digital sources for training generative AI. Read what this means here.
21912
Jörg Lehmann @jrglmn.bsky.social · 12/01/2024
New post "Feeding the cuckoo" published on the blog of the MMK project, focusing on privacy issues in large language models, especially Google's Bard (my friend, the poet). mmk.sbb.berlin/2024/01/12/f... #LLMs #privacy #ChatGPT #ethics #elsi
mmk.sbb.berlin
Feeding the Cuckoo – Mensch.Maschine.Kultur
000
Jörg Lehmann @jrglmn.bsky.social · 09/01/2024
Power Hungry Processing Luccioni, Jernite & Strubell, November 2023 "the most efficient text generation model uses as much energy as 16% of a full smartphone charge for 1,000 inferences, whereas the least efficient image generation model uses as much energy as 950 smartphone charges (11.49 kWh)"
doi.org
Power Hungry Processing: Watts Driving the Cost of AI Deployment?
Recent years have seen a surge in the popularity of commercial AI products based on generative, multi-purpose AI systems promising a unified approach to building machine learning (ML) models into...
033
Jörg Lehmann @jrglmn.bsky.social · 23/12/2023
I wrote a blogpost on LLMs and anthropomorphism for the blog of our project: mmk.sbb.berlin/2023/12/20/h... People who are a bit lonely before Christmas may want to read it…
mmk.sbb.berlin
Human-Machine-Cognition – Mensch.Maschine.Kultur
010
Jörg Lehmann @jrglmn.bsky.social · 22/12/2023
Datasheets for Digital Cultural Heritage Datasets: doi.org/10.5334/johd.124 What are the characteristics of digital cultural heritage datasets? How would dataset documentation look like? We formulate a series of recommendations and propose a datasheet template, see: doi.org/10.5281/ZENODO.8375033
0166