I attended the Benelux Digital Humanities conference. There was a lot to take in, so I’ll let my tweets do the talking.
It was a beautiful summer weather in Amsterdam 7–8 June, but I could barely enjoy it: I attended the Digital Humanities conference for the Benelux for the first time and this was my second DH conference overall. My first DH conference was the one for the German language-area (DHd, held in Cologne), during which I got the hang of tweeting out my conference. I’ll let my tweets lead this short summary of the DHBenelux, but I didn’t tweet everything I saw. For everything and everyone I neglected to mention (sorry!), take a look at the conference program.
First conference day, 7 June
Just saw cool screenshots of a digital editions of Late Greek ‘mashups’ from the epic cycle by @indoukas. Inline commentary, CTS integration with CapiTainS and the maps from Pelagios.— Redmer (@redmer) June 7, 2018
Currently, I’m working on a project that works with Linked Open Data, CTS and CapiTains, so this presentation was a great opener. CTS and CapiTainS enable the easy integration of canical texts (like from Greek or Latin literature) in a website, that work with unambiguous passage identifiers and web APIs. Luckily, the slides were published afterwards.
Keynote speaker @jfwinters now shows a bit of @webrecorder_io, that archives web pages in your own little private collection #dhbenelux Useful for giving workshops and classes, where you want an interactive screenshot— Redmer (@redmer) June 7, 2018
The fragility of web resources is something that comes back time and time again in this field. Just as I’ve given a guest workshop to a group of Archiving students, the example server broke. I saved some pages from my cache, but it was less than ideal to demonstrate to a group of students.
“Museum collection numbers change more often than you think”—Mark Depauw at #dhbenelux on @TrismegistosTM that provides an alternative for such numbers. It also includes unpublished materials from a variety of partners.— Redmer (@redmer) June 7, 2018
Trismegistos was presented as a great base to link data for the ancient world. It creates stable IDs for items that partner institutes give to them. There is however no API available, no data downloads and no clear license.
We did the FAIR Data Principles chant:— Redmer (@redmer) June 7, 2018
Support for DH is @UWuttke’s topic: The Library as a center for expertise for data science. #dhbenelux Before discussing, we are voting live with the audience on some divisive questions about DH https://t.co/g9xN4Sqm2M— Redmer (@redmer) June 7, 2018
.@MaxKemman noticed that according to a source, when librarian make collection, it’s neutral, but when historians do, it’s an argument. 😂 He closes with the statement *Digital Humanities: The librarians are the third wheel* #dhbenelux https://t.co/g9xN4Sqm2M— Redmer (@redmer) June 7, 2018
Finally, @schambers3: The role of libraries in curating research results. Shout out to European Open Science Cloud (EOSC)—though the actual tender looks disappointing #dhbenelux https://t.co/g9xN4Sqm2M— Redmer (@redmer) June 7, 2018
The final poll question was “Researchers should always involve librarians in plans for sustainability of research results” #dhbenelux Reminded me of that the data for my BA thesis “Ucuchi” (https://t.co/eBDMivVUEp) is still offline. Perhaps @UCLA_DH could help rebooting it?— Redmer (@redmer) June 7, 2018
Above tweets came from a round table discussion about the role of libraries in DH and data science research. The live polls were a nice touch to not only involve the audience passively, but also have a point to start talking about in the discussion afterwards. The questions asked and the propositions by the presenters were strongly-worded to spark discussion.
Second conference day, 8 June
Netwerk @Oorlogsbronnen (@LizzyJongma) put together collections from 75 institutes, linking them with topics (SKOS), locations, dates, people (their journeys in WW2 #gdpr).— Redmer (@redmer) June 8, 2018
Openly available/usable resources foster inter-institute collaboration https://t.co/ivLkmnM8Eb
#dhbenelux @afelonnedoek: photo keyword tagging is context sensitive: photo of people walking is tagged ‘tramway, strike’ [so a strike by-effect]. Inter-institute search engines don’t have that context, so @IISG_Amsterdam works on make that more explicit https://t.co/ivLkmnM8Eb— Redmer (@redmer) June 8, 2018
#dhbenelux @ivozandhuis talks about @AdamNet_NL, that improves [access to] CHI information, by combining data sources in a freely available #RDF #Sparql endpoint. Example: https://t.co/WCs5PQJe0Y https://t.co/ivLkmnM8Eb— Redmer (@redmer) June 8, 2018
#dhbenelux in the final discussion: data vs interfaces; more respect for data creation, curation.— Redmer (@redmer) June 8, 2018
The separation of data from UI is useful, I think. Even better when you make your Data API (freely available for everyone) the basis of the interface https://t.co/ivLkmnM8Eb
✨S✨P✨A✨R✨Q✨L✨ #DHBenelux— Redmer (@redmer) June 8, 2018
This round-table “Applying DH techniques to a cultural heritage environment” was a really nice one. I saw examples of projects working with Sparql and RDF, that first provide an API, on which other institutes, projects and people can build. This seperation of concerns helps the sustainability of the data (a recurring issue even in this report :laughcry:).
Linked Open Data #linkeddata #dhbenelux— Redmer (@redmer) June 8, 2018
– Disambiguation of names
– Marriage, baptism registries, probate inventories
– Did some exploration via Neo4J
– ‘Lenticular Lenses:’ an app for link discovery and refinement
Linked Open Data #linkeddata #dhbenelux— Redmer (@redmer) June 8, 2018
– Migrant workers in NL 1700–1800
– Matching people from monsterbrieven
– wage, career progress, origins
– some differences between Dutch and migrant workers
Sabrina Sauer talks about insights from the #clariah #Nardis project on #linkeddata exploratory search and #narratives for disruptive events. #dive @neocarlitos @oana_inel @laroyo #dhbenelux pic.twitter.com/7QEG0iGqt7— Victor de Boer (@victordeboer) June 8, 2018
Linked Open Data #linkeddata:— Redmer (@redmer) June 8, 2018
– by @drkmurch
– 958 medieval English manuscripts of Norman-French lit
– metadata available as #FAIRData
– I’m really thinking of profileDesc/langUsage here
– Sweet 3-line Latin, French, English poem (too late for pic)
This session about Linked Open Data gave some clear examples how the Open aspect of LOD enables cross-referencing between projects.
Next came the poster and software demo session. It was unfortunate that the demos were in a side-room and that I had too little time to talk about all posters (e.g. the one about Syriaca).
Nevertheless I saw a demonstration of ‘Example Based Search’, a way to search through text corpora by means of an target sentence. That target sentence is parsed and the user indicates what parts of that sentence are relevant to the query: the part-of-speech, the literal word, the case or nothing at all. I think it’s an inspirational way of searching through data, that especially makes searching in text corpora easier.
Oh wow I can only say the closing keynote blew my mind. Unfortunately, what was presented was still under wraps, so all I can say is: I had no idea that [REDACTED] was possible to this extent. At the Lowlands music festival, both presenters will show off their rap flow work for a live audience. Go and see that!
I’ve met a lot of people working in the Benelux digital humanities area, which was exactly the reason I came to DHBenelux. There were really cool presentation, both professionally useful for me and simply technologically amazing. The food was good, I made some friends and I can’t wait for the next (DH2019 in Utrecht?).
Many other people have also tweeted with the official™ hashtag #dhbenelux. Take a look at those for the other tracks, as well as the blog posts about the conference below. If there are more, please let me know.