iiinova — where your digital self keeps evolving

READ-COOP has been a great long term supporter of our Winter School :). Thank you! @transkribus.bsky.social

Reposted by Cato Minor

DigiLab, Institute for Medieval Research, ÖAW @digilab-imafo.bsky.social · 8h

For the next few weeks, we are teaching a new generation of young scholars how to use handwritten text recognition at our HTR Winter School in Vienna, which is this year is online only. Many thanks to @transkribus.bsky.social for supporting us. 🙏 #HTR #DigitalHumanities #Transkribus

Presentation slide titled "Transkribus" from the ÖAW Institute for Medieval Research. Bullet points say Transkribus is the oldest platform (since 2013) and the easiest to use, with a huge number of models and a large community, and a sustainable infrastructure (cooperative plus fees). Trends are building an ecosystem and advanced features such as smart extraction. Below, a diagram shows a workflow in three steps: ScanTent for scanning, Transkribus for transcribing, and Transkribus Sites (Read&Search) for publishing.

Screenshot of the Transkribus web editor during an online Zoom session. On the left is a page from a medieval Czech manuscript (OENB Cod. 1175) in two columns of dense Gothic script, with a decorated initial and a miniature of three crowned figures. The text lines are marked with green region boxes. On the right is the transcription panel, where the first line reads "We can start transcribing!"

053

Cato Minor @catominor.bsky.social · 21h

Photocopillage? Tell us more :)

000

Reposted by Cato Minor

Martin Roček @martin.rocek.dev · 11/10/2026

We've added a few more tweaks to Translucens' algorithm with @vanlinduist.bsky.social. Some parts might be damaged beyond repair, but overall I'm quite happy with the update. #DH #Translucens

091

Reposted by Cato Minor

Regesta Imperii @regestaimperii.bsky.social · 08/10/2026

Wir freuen uns auf unsere neue digitale Veranstaltungsreihe "RI colloquium digitale"! 📅 jeden zweiten Donnerstag, 16 Uhr 🌐 digital via Zoom (Link bei Anmeldung) 📬 regesta-imperii@adwmainz.de Wir starten am 22.10. mit einem Beitrag von Andreas Kuczera zu den "Regesta Imperii im Graphen" #medievalsky

3118

Cato Minor @catominor.bsky.social · 08/10/2026

Source: O slově robot ("About the word robot). Published in Lidové noviny, 24. 12. 1933.

120

Cato Minor @catominor.bsky.social · 08/10/2026

Actually (sorry for that!), Karel Čapek wanted to call them "laboři" (from the Latin "labor") originally, but he himself did not like that word. So he asked his brother Josef Čapek who without much thought suggested "robot", "while continuing to paint". (K. Čapek wrote this recollection in 1933.)

231

Cato Minor @catominor.bsky.social · 29/09/2026

You give us great informative pictures :). Thank you.

000

Cato Minor @catominor.bsky.social · 29/09/2026

That looks amazing! I wish I could experience it :).

110

Cato Minor @catominor.bsky.social · 29/09/2026

This is especially true for under-resourced langauges where there is simply not enough existing training data. Such challenges, however, don't make the technology itself unsound or useless. :)

000

Cato Minor @catominor.bsky.social · 29/09/2026

And I would like to add: while imperfect, ATR can provide much better results in the informed hands. There are many different models, platforms and technologies. We are still not at "it just works for everything"-level. One may need to test a couple thing and maybe train models or fine-tune them.1/2

100

Cato Minor @catominor.bsky.social · 29/09/2026

If the author decides to cover (mention) ATR/HTR in her guide, I encourage her to send it to her peers who are knowledgeable about the matter. The ATR community is very welcoming, and there are many people who would be happy to check its accuracy. :)

110

Cato Minor @catominor.bsky.social · 29/09/2026

I understand the intention but it is unfortunately a very uninformed post. One should argue from the informed position not from impressions.

110

Cato Minor @catominor.bsky.social · 29/09/2026

3/3 Der Österreichische Bibelübersetzer, and many more. If you had any question, please, don't hesitate to ask :). This a fascinating field - but as almost anything in our research - one needs to learn about it. :)

100

Cato Minor @catominor.bsky.social · 29/09/2026

2/? where scholars write articles, publish books, and present at dedicated conferences. Indeed, ATR has been part of major ongoing projects dealing with medieval manuscripts that produce tangible results: The Paris Bible Project, the edition of Decretum Burchardi,

120

Cato Minor @catominor.bsky.social · 29/09/2026

1/? I don't doubt the author's expertise. I would just like to point out that the blog post can be summarized as: “I Googled a tool, pasted an image into it, and got bad results.” As researchers we should go beyond the naive approach. Automated Text Recognition is a research field

120

Cato Minor @catominor.bsky.social · 26/09/2026

(And I still miss the buttons of remarkable 1... but alas, buttons are almost non-existent among larger e-ink devices).

000

Cato Minor @catominor.bsky.social · 26/09/2026

2/2 it also uses screws so I think / hope that the battery replacement should be easy. And yes, the colors of Pro and Pro Move are the best ones among color e-ink devices, IMHO, but I think that if you don't really need colors, it does not outweigh its quirks.

100

Cato Minor @catominor.bsky.social · 26/09/2026

Now I am a proud owner of all three modern reMarkable models: Pro, Pro Move and Pure. And to be honest, I think the Pure is the best of them, if you don't need colors and front light. It refreshes the display less often, it has better contrast, it is quicker, it is much lighter than Pro.1/2

110

Cato Minor @catominor.bsky.social · 14/09/2026

I am looking for people using AI coding in digital humanities for a workshop I am organizing in December 14-16 in Vienna :). If you are one of those or you would recommend someone, please, write to me.

000

Reposted by Cato Minor

DigiLab, Institute for Medieval Research, ÖAW @digilab-imafo.bsky.social · 20/08/2026

Are you interested in Automated/Handwritten Text Recognition of historical sources? This September 7-9, we are organizing the 2nd Exchange Meeting of SCOOP (Sources Codes of the Past) in Vienna, bringing together researchers and developers. More information: www.scoop-network.org#programme #htr #atr

scoop-network.org

SCOOP: Source Codes of the Past

A conference and exchange network for the international community working on Handwritten Text Recognition (HTR). Vienna, 7-9 September 2026.

032

Cato Minor @catominor.bsky.social · 20/08/2026

Glad to hear that :). Actually, I returned to Roam in the past months and I like it! There is still something magical about it. Regarding other note-taking apps beside Roam, I am now more into e-ink note-taking devices, like reMarkable or Onyx Boox.

000

Cato Minor @catominor.bsky.social · 31/07/2026

A fun fact: A couple of months ago, I actually returned to Roam Research as my PKM, back from Obsidian. I just opened it one day and it was like before - even my weird CSS still worked. Everything since I left it in 2021 (?) is still in its place, it simply works.

020

Cato Minor @catominor.bsky.social · 30/07/2026

I am happy it works for you :). It seems to be quite a solid device. I just disliked that there was no proper synchronisation with the computer and phone. Maybe they have fixed meanwhile.

000

Cato Minor @catominor.bsky.social · 30/07/2026

A fun fact: I have, indeed, later bought and sold again ViWoods AI Paper. :D

100

Cato Minor @catominor.bsky.social · 29/07/2026

There was Transkribus at the beginning and then eSriptorium, Pero, OCR4All, Teklia, Calfa Vision but it was still calm. Now, we have a new vibe-coded transcription app like every month? Unsubstantiated claims, simply using some VLM. #ocr #htr #atr

010

Cato Minor @catominor.bsky.social · 28/07/2026

Tweeting about your minor find in 19th century.

001

Cato Minor @catominor.bsky.social · 26/07/2026

After one and half hour of vibecoding with Github Copilot Pro. Okay.... That became really unusable for some small experiments.

000

Cato Minor @catominor.bsky.social · 24/07/2026

Does anybody have this book? :)

000

Cato Minor @catominor.bsky.social · 24/07/2026

Definitely. E.g., do more expensive manuscripts burn quicker or slower?

010

Cato Minor @catominor.bsky.social · 24/07/2026

My idea for a quantitative study of the flammability of medieval manuscripts doesn’t sound so outrageous now, does it?

120

Cato Minor @catominor.bsky.social · 21/07/2026

And for models on a specific platform, a link could be shared, be it Transkribus or Hugging Face or elsewhere. Don't take me wrong: great that you are expanding. I just think one platform could more fulfill the expectations of the users.

100

Cato Minor @catominor.bsky.social · 21/07/2026

But would it be any problem? Models that could be shared, would be shared. Models which could not be shared, would not be shared. It seems not a big issue to me.

100

Cato Minor @catominor.bsky.social · 20/07/2026

I understand that they are separate entities. My question would go rather, why have you decided to have them separate? It would feel for me very natural for HTR-United to have also links to models.

100

Reposted by Cato Minor

Thibault Clérice @ponteineptique.bsky.social · 20/07/2026

Hey, we are happy to announce HTRMoPo Web at @inriaparisnlp.bsky.social ! htrmopo.inria.fr HTRMoPo is a HTR Model Repository that allows you to describe your Kraken Models. It has been built by Ben Kiessling, the researcher behind Kraken. What does the web version do ? Let's see:

A screencapture of the home page of HTRMoPo with the sentence "Document & Publish Kraken models"

2145

Cato Minor @catominor.bsky.social · 20/07/2026

Just a small suggestion: It would be great, if the creators of models were more visible without the need to follow the Zenodo link. How should it coexist with HTR-United? :)

211

Cato Minor @catominor.bsky.social · 17/07/2026

So it seems that Google became totally unusable for searching images. I wrote "medieval scribe" into the search field and maybe 50 % of images is AI generated.

020

Cato Minor @catominor.bsky.social · 15/07/2026

If the manuscript contains medieval logic, then a scholarship for a logician, if theology, then for a theologian, if necromancy, then for a necromancer.

010

Cato Minor @catominor.bsky.social · 14/07/2026

I love PDFs for academic publications. The things simply stay at the same place and don't move. You can also still open a very old PDF without any issues.

110

Cato Minor @catominor.bsky.social · 11/07/2026

My night train from Brussel to Vienna is already six hours delayed and we still have to go through half of Germany. Night trains going through this country are so bad that their purpose (you save time by travelling at night) is completely obliterated.

000

Cato Minor @catominor.bsky.social · 08/07/2026

One of the interesting developments at #imcleeds2026 is the number of presentations generated by Claude. I have seen already four, I think. And to be honest ... they look fine. Just that it is difficult to unsee them.

010

Cato Minor @catominor.bsky.social · 08/07/2026

Today I learned that Chinese made facsimiles of letters already in around 3rd/4th c. CE (if I understood the speaker well). Fascinating. Do we have facsimiles from elsewhere that would be so old?

010

Cato Minor @catominor.bsky.social · 07/07/2026

Digital humanities seem to have one key advantage: one has always something to present: A graph, a network, equations, a bunch of numbers. This seems to be much easier than with traditional research. But sometimes I am left to ask myself: what real knowledge are we gaining? #digitalhumanities #dh

110