Our journey is over, we showed our tool on the demo day and talked to a lot of people about it. However, our tool is still a proof-of-concept and not ready to be deployed anywhere.

When we were creating the architecture behind our tool we came across a huge discussion about the naming of the ‘things’ we were generating. When can we speak of:

• a building?
• a service?
• an organisation?

Publishing Linked Open Data isn’t all about data. You also need to describe every piece of data for other users. To describe your data you can use something called: vocabulary, but where do you get these?

The problem

Even when cities do use linked open data, it’s not always completely the way it should be. For example, one of the cities that publishes open data forgot to add Cross-Origin-Resource-Sharing (CORS) headers to its website, making the data inaccessible from other browsers when using javascript.

Using other people their open data is a nice idea. However, what if their an error in their data? In this blog post we will take a look at several examples and what you can do this to fix the errors in the data.

Using the stuff of somebody else might be a challenge if there’s no documentation around. This blog post is trying to explain the workflow of the Smart Flanders tool we’re building to help the civil servant to publish its data as Linked Open Data in a good way.

Finding and accessing open data can be a challenge

The current situation

ID’ing is a problem

The problem

We’re the Smart Flanders team for open Summer of Code 2018 and we have only one purpose: making Flanders smart again! Our team consists of only three people: