Thoughts on Linked Open Data following REEEP/NREL Washington D.C. LOD workshop

My favourite thing about being REEEP’s Community Facilitator is the people I get to meet – intelligent, concerned people working across a diverse range of issues, in all types of organisation from small NGOs to enormous International Organisations and governments. The kind of committed people who give up their Federal holiday to attend a workshop on the potential of Linked Open Data (LOD) to support the transition to a clean energy system. That’s exactly what happened last week when, as Denise Recheis reported in the previous post, REEEP and NREL ran a LOD for clean energy workshop in Washington D.C.

REEEP workshops and seminars are always interactive so there’s plenty of chance to hear from all participants. We were particularly excited that Bernadette Hyland, CEO of 3 Round Stones and a pioneer in semantic web technologies, came along. She concluded at the end of the workshop: “It was really exciting to see the first workshop dubbed as a ‘Linked Open Data’ only workshop in DC (and I follow these things!).  I commend you for organizing such a useful program, format, venue and the nice ‘take aways’ for the workshop. Thank you.

My second favourite thing is that I get to learn so much. As I said to the participants, I’m not an expert in LOD – address

Jon gets technical

your questions to Florian, Denise or Jon Weers over at NREL if you want to get technical – but I’ve facilitated a couple of these workshops now (the first was in Masdar City in January; more on that in this post) so I’ll try to explain briefly what it’s about:

Open Data is data that is made available with an open licence so others can re-use it freely.

Linked Open Data is when that data is made available on the internet in a machine-readable format, including an explanation of what the data is, enabling other people to link directly to the dataset, or a specific part of it.

A key point is that data is not simply LOD or not LOD – there are various steps on the way. Tim Berners-Lee has suggested a five step plan: http://5stardata.info/

If data is available in a 5-star way it offers a number of benefits:

  • the opportunity to enrich your datasets by linking relevant or related data from other organisations, e.g. reegle is a consumer of open datasets from NREL’s OpenEI, as well as the UK Government and institutions such as the UN and the World Bank, which we combine with our own information to create detailed country energy profiles. We are not copying their info to reegle; rather we are linking directly to their database and pulling out the information we want in real time, so if they update their info, reegle will automatically show the most up-to-date info;
  • clarity, since it needs to be completely clear what the data is. At the workshop Jon explained this via ‘the Georgia problem’: if you say Georgia to someone in the US, most people will think you are talking about Georgia the US State, but if you say Georgia to someone else, they might think of Georgia the country in the Caucasus region. In LOD, the data would include the explanation of which Georgia it referred to;
  • efficiency, since you can access the specific data you want, rather than downloading the whole dataset;
  • reduced duplication – there’s no need to store multiple copies of data on different servers if you can link directly to someone else’s dataset on their server. Reducing duplication also reduces the risk of error and confusion if one copy becomes out of date.

There’s a good explanatory animation on LOD here.

So why is REEEP promoting the idea of LOD? Our overall mission is to accelerate the uptake of renewable energy and energy efficiency. We believe that by increasing access to, and reducing costs of, energy data and information we can

Participants' thoughts on key organisational challenges to implementing LOD

support the effectiveness of government, industry, academia and NGOs in their respective roles in the energy transformation. We know from our network that organisations experience difficulties in finding the data and information they require in a suitable format. Due to the small number of organisations currently providing Open Energy Data, we are already reaching the limits of how far we can develop reegle to meet users needs.

Of course a lot of energy related datasets are already out there but unfortunately most of the time the data is stored in different places, and structured and presented in different ways. It is often not in a machine readable format (e.g. PDF documents) and there is rarely a clear indication of the license and its requirements so users cannot be sure what they can use it for. This results in datasets in silos that are hard to link, compare and combine, which prevents innovation and presents a hurdle for project implementation in clean energy.

Our tame experts respond on how to overcome the challenges

We’ll be doing further workshops in different locations in the future. Next up will be Germany early next year with our colleagues at BMU, who kindly supported the Washington event. So watch this space if LOD interests you; we’re on a bit of a mission!

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s