Against Linked Open Data
Linked Open Data may sound good and noble, but it’s the wrong way around.
It is a truth universally acknowledged, that an organization in possession of good Data, must want it Open (and indeed, Linked).
Well, I call bullshit. Most cultural heritage organizations (like most organizations) are terrible at data. And most of those who are good at collecting it, very rarely use it effectively or strategically. Why then, this fetishization of Linked Open Data (LOD) for collections when just getting good Data is a struggle for most cultural orgs. At the very least, it’s the wrong way around. Data (yes!), Open (if you explain how to access it and keep it up to date) and Linked (only in very rare circumstance, when there is a clear ROI*). If your institution isn’t using the Linked Data, and it’s your Data, why would outsiders be interested?
* “I can’t wait to see what people do with our data” is not a clear ROI.
It took me some time to come around to this (potentially unpopular) way of thinking. I’ve been a long time proponent of LOD, I’ve written and spoken in favour, and attended multiple LOD events and conferences. I’ve attended so many events, in fact, that I’ve started mapping participants and their affiliations to better understand who is contributing the field such as LOD LAM 2015 and LDCX 2017 (TLDR: it’s an awesome community of super smart people — primarily made up of digital library departments in wealthy US universities, big art museums, gov agencies from Europe, Australian and New Zealand, platform developers, activists and artists).
Just uploading your data to the internet is the digital equivalent of dumping folders of paper and Polaroids into the street. Sure it’s public, but most people don’t know where to find it, or how to use it.
Even Tim Berners-Lee’s 5 star data isn’t an appropriate barometer for success for cultural sector data, and Tim is a god. It has LOD as the 5 star standard, without points for applicability, sustainability, explanation and accuracy.
I propose an alternative anagram, and an alternative order of importance.
- D. Data. Step one, collect the data that is most likely to help you and your organization make better decisions in the future. For example collection breadth, depth, accuracy, completeness, diversity, and relationships between objects and creators.
- U. Utilise. Actually use the data to inform your decisions, and test your hypotheses, within the bounds of your mission.
- C. Context. Provide context for your data, both internally and externally. What’s inside? How is represented? How complete is it? How accurate? How current? How was it gathered?
- S. Share. Now you’re ready to share it! Share it with context. Share it with the communities that are included in it first, follow the cultural heritage strategy of “nothing about me, without me”. Reach out to the relevant students, scholars, teachers, artists, designers, anthropologists, technologists, and whomever could use it. Get behind it and keep it up to date.
I’m against LOD, if it doesn’t follow DUCS first.
If you’re going to do it, do it right.