Monthly Archives: October 2019

Open Data Camp: 7 up!

Sarah Roberts, who is a member of the team at Swirrl, one of our Bronze Sponsors, has written about Open Data Camp 7 on Swirrl’s blog. Here’s what she said.

We’re happy to be sponsoring Open Data Camp again, which heads to London the weekend of Nov 2nd and 3rd. In honour of the seventh edition of the camp, here are seven recent bits of data discussion which have caught our eye …

In at Number One

Image Source

The recently published Northern Ireland Open Data Strategy, is straight in at Number 1. The strategy runs from 2020–2023 and has four main themes:

  • Increasing the rate of publication and sharing of government information as open data
  • Promoting original, useful ways of using open data to show its benefits.
  • Engaging with the public sector to promote Open Data through events, setting up an innovation fund and running open data competitions
  • Training so Open Data’s a business function, rather than an ad hoc activity.

There’s a lot of interesting content in the strategy, and it’s especially great to see ‘making data accessible and easy to use’ as one of its key principles. And a mention of tidy data. Be still my beating data heart.

Two: Sweet Harmony

And tidy data makes me think of another good thing going on with open data : The Connected Open Government Statistics project. This is a GSS project which aims to harmonise data across departments so it’s interoperable. There’s a lot of good stuff going on here to tackle the disparate data landscape in a standardised way and a lot of organisations involved. More on this here.

Three: Stand(ards) and Deliver

Stand and Deliver by Adam and the Ants. Image Source

I don’t go to many conferences, but those I have been to this year have all included AI as a topic. I was talking about this to my colleague, Jamie Whyte, who said one similarity between AI and data science is that they both depend on the quality of data they operate on. Bill Roberts spoke about the same thing from a different angle at our Power of Data event a couple of weeks ago: the idea that if you want to do data science you need some data, most likely from lots of data sources. And if that data’s not in a good shape when it goes it, then the risk is that automation will generate poor quality results. Standards at the point of publication are the answer here — not Adam and the Ants, really. But enjoy the ear worm.

Four: Never Ending Story

Storytelling with data. I know, it’s not new but it’s still something that hooks me in. At the recent GSS conference in Edinburgh, Mark Robinson of NHS Health Scotland presented a really engaging talk on the use of data in interventions for health inequalities. He showed this Shiny app (screenshot below) in his talk, which allows users to select the area & health outcome they want to look at:

5ive

I’m totally showing my age now, I know. Image Source

In at number five are the Indices of Deprivation. This data was released last month by MHCLG and Bowie Penney covered some of its uses at the same GSS conference. The uses are wide ranging, from contextual funding use to estimating life expectancy, and people are already making stuff with it, like Open Data Manchester’s lava lamp visualisationRob Fry’s interactive map and Alasdair Rae’s work. Get to the IMD postcode apparea app and the Indices of deprivation datasets here.

Six: Stop, Collaborate and Listen

Slide from Jeni Tennison’s talk at Power of Data 2019

The Open Data Institute have been doing a lot of work around the idea of data institutions for the new data age. I love the Data Access Map (shown) and other related work includes research on Data Trusts and a joint paper with the RSA & Luminate on data rights and data ownership. Open Data Camp’s a huge collaborative effort and there’s another opportunity to meet up with people collaborative style at the next cross-government API meeting in Newcastle next month.

Seven: One Vision

Not the catchy Queen song (although, what a song) — the National Data Strategy. We were lucky enough to have Gaia Marcus, who heads up the strategy at DCMS, chair a panel on this at our event a couple of weeks ago. Panelists included Rosalie MarshallSiân ThomasGavin Freeguard and our very own Bill and questions ranged from opportunities to improve public services through effective data use to priorities of data standards and data infrastructures. The videos will be up very soon — worth a watch!

This time Swirrler, and long time Open Data Camp supporter, Jen will be there — say hi if you see her there!