Steeds meer onderzoekers halen hun data terug uit de VS. “Ze willen niet afhankelijk zijn van Amerikaanse regelgeving,” zegt Magchiel Bijsterbosch. Bij SURF herkennen we die zorg.
Samen met onze leden bouwen we aan betrouwbare, soevereine voorzieningen voor opslag, analyse en deling van onderzoeksdata. Voor nu en in de toekomst.
Lees het NOS-artikel: https://nos.nl/artikel/2564520-wetenschappers-halen-stilletjes-al-onderzoeksdata-weg-uit-vs-om-toegang-te-houden
AT #SWAT4HCLS #Biohackathon, we are already exploring ways to utilise AI to generate SPARQL queries and curate ontologies. For example, by enriching metadata collection using VOID headers. Can these metadata headers be used to feed data to AI approaches and consume AI results?
The #SWAT4HCLS #Biohackathon started. We have just inspected the co-hacking spaces available for the participants. There is the hack lounge where we can relax and hack and the Biohackathon room where people can claim a table to work.
#SWAT4HCLS #Biohackathon pitch #1 "Publishing FAIR Datasets from Bioimaging Repositories". This proposal aims at making Bioimage repositories, more notably OMERO, FAIR. This probably by building a FAIR Data Point on top of these repositories.
Who is joining #SWAT4HCLS?
I am organising the #biohackathon that is part of this exciting conference where we mainly discuss the application of the semantic web in the life sciences.
Please reach out if you want to propose a topic to be addressed during the biohackathon which will be held on Feb 29. There will be a co-working space assigned during the other days (from the 26th onward) where you can either prepare or start the hacking.
Last year, we renamed the #SWAT4HCLS hackathon to Biohackathon #SWAT4HCLS. This year, we are extending the format. The last day remains the official hack day, However, during the tutorial and conference day there will be a biohackathon room where we can prepare our activities for that hackathon day.
Please join us for #SWAT4HCLS (https://www.swat4ls.org/) the Biohackathon (https://lnkd.in/eNmFJK7B)
You are also cordially invited to submit biohackathon proposals/pitches: https://lnkd.in/e5GzVVBn
#guessr on dutch municipalities: https://guessr.blinry.org/?Q2039348
One very nice example of a data driven game an a great with to start the weekend is guessr. Where you have to guess the actual location of an object. In this link you have to guess the location of university hospitals: https://guessr.blinry.org/?Q1059324 (Q1059324 is the wikidata id for http;//www.wikidata.org/entity/Q1059324. Please share other Qids if your find other fun games to guess. We can use the inspiration at the #swat4hcls #biohackathon.
During the preperation meeting for #swat4HCLS 2024 we decided to introduce a game theme at the hackathon, that is developing biohealth games being fed by knowledge graphs and/or wikidata. Ofcourse it is also an excuse to play games, but who is game for this game plan?
You can register at https://www.swat4ls.org
@vrandecic @vrandecic despite the political ramifications, it is an interesting development of which it, as you say, would be interesting to follow. I think forking of wikidata in different branches is inevitable and in away a logical step following the wikidata wikibase federation we have also been seeing in the past years.
Next week Andra Waagmeester (@Andrawaag) will defend his PhD thesis on January 16 on the "Biological Pathway Abstractions: From Two-Dimensional Drawings to Multidimensional Linked Data" (https://www.maastrichtuniversity.nl/events/phd-defence-andra-sachinder-waagmeester). In the morning there will be a mini symposium (see image). Register here today: https://forms.gle/oqMbsVhurBiEQeca7 ! #wikipathways #biology #rdf #wikidata
@egonw @wikipathways If they are covered by #wikicite that shouldn't be necessary right. All you need is a federated SPARQL query.
@jtwcornell91 @mart1oeil @rafa_font @trains @Ruth_Mottram No reservations are not required, but most travellers take seat reservation and with an interrail that is not trivial if at all possible. So I ended up using the interrail in Germany and Denmark and had to switch seats often
@rafa_font I am jealous of SBB. I live in Belgium, which we could consider the heart of the EU, yet the Belgian railways and their unions, contrary to SBB, are actively blocking any cross-country collaboration. See e.g. https://www.brusselstimes.com/406943/tri-state-train-between-belgium-the-netherlands-and-germany-gets-green-light. Any progress and chance in cross-border rail connections is met with fierce resistance by both management and the unions. As long as this remains, trains in Belgium will remain difficult.
@rafa_font @trains Regarding the point of having to deal with different providers who don't talk to each other, 100%. However, I sometimes rely on https://www.thetrainline.com/, who somehow seem to have figured this out. I don't know what happens if your connection is disrupted since, so far, I was always able to make the connection.
@mart1oeil @rafa_font @trains @Ruth_Mottram
That is not entirely accurate. #interrail does not work well when you have to deal with seat reservations in, for example, Germany or Denmark. Do expect to change seats often. In other countries, interrail comes with hefty reservation fees.
How to render publication records on #wikidata using #scholia. This is specifically written for the #biohack23 but should work on any organisational unit: https://osf.io/preprints/biohackrxiv/py5jq/
@davidho That is primarily virtue signalling by local governments. Most cycle lanes in Belgium remain extremely dangerous and are even called "murderlanes" or "moordstrookje" , which coincidentally once won the word of the year award. I agree we need to talk more about Belgian bike infrastructure, but mainly the lack of it. (source of photo: Door Bert76 - Eigen werk, CC0, https://commons.wikimedia.org/w/index.php?curid=111023183)
@thoasm The interesting part is that in France, it argues that it should not give results because patient data is sensitive. It could also be explained by different interpretations of the word "arbitrary", which does not necessarily mean fictional.
"Give me a arbitrary FHIR record for a patient with COVID admitted in hospital X"
To chatGPT from Belgium: https://chat.openai.com/c/968d05e4-dcc5-4745-b2ab-b7b25aaba2a6
To chatGPT from France:
https://chat.openai.com/share/9d3648cf-33f0-4b46-98ba-8ff1c9670879
Same prompt, different results, any one have ideas why this is?
