Made some Python to convert MIDI signals from a Yamaha FGDP-50 to play Hang Drum samples: instagram.com/p/C-IuQ4loS0f/ Very satisfying dulcet sounds wafting through the house on this summer evening.

Modern SQLite #1: STRICT Tables I'm starting a series of short notes about handy SQLite features you may not have heard about. And the first one is "strict" tables. As you probably know, SQLite's type system is very flexible (some people even call SQLite the JavaScript of databases) — you can store any value in any column type (e.g. create an INTEGER column and store text values there, or a REAL with blob values). Some people love SQLite for its flexibility, others hate it for the same reason. So at some point the SQLite authors introduced STRICT tables. They check types the same way other DBMS do (see the screenshot). Even with strict tables, you can still explicitly declare a column as ANY — such columns can hold values of any type. So you can have the best of both worlds — strict type checking and type flexibility. Available since SQLite 3.37 (Nov 2021).

Super happy to see this announcement. I have been considering making a super-simple read-only *.ipynb viewer for the console in Textual for months now. But voila! The cool froods over here have done one better. I am not the only one wanting this...

Hooray! Our team is the featured image... ;-)

Een week geleden was het oude stadhuis van Gouda het episch centrum van de 7e HackaLOD, de hackathon voor het erfgoedveld. Lees in dit artikel wat alle 7 teams aan creatieve toepassingen met erfgoeddata hebben ontwikkeld winnaars & filmpjes bit.ly/3QAfYjA

Yesterday‘s DC at @iswc_conf #iswc2023 was such fun! Grateful to be able to present my research on representation of time in Cultural Heritage Knowledge Graphs. The slides are available here: zenodo.org/records/10081703

At #SWODCH2023 I had the honor to present: A Data Model for Linked Stage Graph and the Historical Performing Arts Domain 🍿 🥳 #iswc2023 co-authored by @sashavses @lysander07 Slides: zenodo.org/records/10077479 Paper: ceur-ws.org/Vol-3540/short1.… @FIZKarlsruhe @fiziseka

See the whole thread, but this little clip tells a whole story. Fabulous tool for Book Historians

Despite the lack of sleep, team FIZ had an amazing time at @HackaLOD in Gouda! @fiziseka @FIZKarlsruhe @sashavses @MahsaVafai @epoz #culturalheritage

Blog 4 voor de Preservation Watch van @DigitaalErfgoed over Large Language Models, digitaal erfgoed en preservering is nu online te lezen. Deze keer mocht ik Etienne Posthumus @epoz@blog.epoz.org en Bob Coret @coret interviewen over hun ervaringen

Pluimpje voor de klantenservice van @hema . Ik ben zeer vriendelijk en bekwaam geholpen. Thx!

If you grew up in the eighties, do yourself a favour and listen to this 8-bit chiptune: open.spotify.com/track/1KKyJ…

#arthistoCast Folge 6 ist online: Normdaten in der Kunstgeschichte. Mit Angela Kaius @docgrid & Julia Rössel @JuliaR2Punkt0 spreche ich über #Normdaten, #Vokabulare und den Nutzen für die #Kunstgeschichte arthistoricum.net/themen/pod…… @digiarthist @kunstgeschichte @nfdi4culture

Shout-out to the vocabularies team at @GettyMuseum - the vocab.getty.edu/ LOD endpoint is a joy to use for the power-user. 🙏 💪 Using it for some German/Dutch/English museum description reconciliations and it works a treat.

Gaussian Splattings are absolutely amazing in themselves - and here is a whimsical yet beautiful abuse of them. It gives me the same sense of wonder as the first time I saw a bullet time rendering.

How do y'all handle (large) n-triple files with bogus URIs in them? Most triplestores accept URIs with unencoded characters like [ and ] in them, but certain ones are all strict and barf. I could iterate over the input and "fix" it with a script, but this is tedious. (and slow)

Want to learn more about #knowledgegraphs? We recommend the Open HPI course by Prof. Dr. Harald Sack @lysander07 of our co-applicant @FIZKarlsruhe: Knowledge Graphs - Foundations and Applications: open.hpi.de/courses/knowledg… #NFDI4Memory #FIZKarlsruhe #BigData #AI

2 open PhD & PostDoc positions at FIZ Karlsruhe. Help us to solve challenging research questions in #knowledgegraph based research data management and knowledge extraction in the domains of historical sciences, mathematics, and more #nlp #nfdi #PhD fiz-karlsruhe.de/en/stellena…