May 5, 2020 by Audrey Lobo-Pulo
As the Scottish Cities Alliance invest in new technologies to create ‘Scotland’s eighth city – the smart city’ to accelerate and transform service delivery, Datopian’s open source, collaborative and custom developed modules made it a top choice for future projects.
Established in 2011, the Scottish Cities Alliance is a unique collaboration between Scotland’s seven cities – Aberdeen, Dundee, Edinburgh, Glasgow, Inverness, Perth and Stirling – and the Scottish Government, working to build on the economic potential for the country. By working together, the alliance partners share knowledge and create projects of scale, thereby offering great investment prospects across the business spectrum.
Our CEO, Paul Walsh, spoke with Doug Young, who coordinates the Data Cluster, a working group within the Scottish Cities Alliance, with a goal of progressing open data in Scottish cities to drive innovation, increase transparency and improve the use and re-use of data within local authorities.
Paul: Doug, we’d like to find out more about what the Scottish Cities group are doing – but first could you tell us a bit about your role?
Doug: I’m effectively a coordinator of the collaborative components of six projects between the seven Scottish cities, all working towards the publication of open data. This is also connected to a wider Smart Cities program: the ‘8th City Programme’, which itself is supported by European Regional Development Funding (ERDF) – covering a wide variety of smart cities projects.
# Data is very much at the core of this, because one of the key outputs for all of these Smart Cities projects, be it smart waste, intelligent street lighting, etc., is the requirement to publish open data. So, data is very much at the heart of the whole program. – Doug Young
I work between all the cities, but I effectively coordinate individual projects and look after the collaborative elements of that. As a collective working group, we’re looking at things like publishing data to particular standards, publishing a range of common datasets between us, and a couple of other things too.
Paul: So this is very much driven by a smart cities type of program rather than a more traditional government transparency program that often drives open data initiatives – is that right?
Doug: Absolutely – the key indicator for these projects is opening up datasets for innovation, so it’s very much about looking at data sets from that perspective, although transparency does play a role as well.
Paul: So, tell us how you chose CKAN for the open data portal component?
Doug: When the cities joined the ‘Open & Agile Smart Cities’ network, one of the recommendations was that when we’re publishing data, it’s on a CKAN platform, and that certainly influenced our thinking. In addition, the sheer flexibility and the open source nature of CKAN was appealing. The modular nature of CKAN was another thing that was very attractive to us. All of this led to us going down the CKAN route rather than looking at a more proprietary approach.
Paul: Great. And how did you come to work with Datopian?
Doug: So, when we started, five of the seven cities did not have an open data platform – so a requirement for a platform was our starting point. So we said, “Look, we actually need somewhere to publish our data.” Rather than carry out a series of individual procurements, we came together as a ‘kind-of consortium’ and used our own existing procurement consortium to jointly procure a platform. We used the UK government’s GCloud framework for cloud services, and through that process – and there were a few competitors there – but at the end of that procurement process, Datopian came out on top. It’s worth noting that the other competitors were offering more of a traditional Software-as-a-Service (SaaS) model.
With Datopian, there was more of a collaborative element there – in terms of developing custom modules and the opportunities around that, as well as the open source nature which was quite attractive to us. That was a big reason why Datopian came through the procurement process.
Paul: Sure. In terms of the wider program, have you got any insights you could share – like the impacts you are seeing from the open data publications or as part of the Scottish Cities programs in general?
Doug: So, one thing we’re going through at the moment is collecting those case studies, because our approach to publishing so far has been kind of ‘off the cuff’. A number of datasets have had many thousands of views, and that’s often been with relatively little promotion – but we’ve increasingly seen the platforms as a good place to have and publish interesting information, or even to collate existing information and have one place for local authorities.
For example, most of the cities have published school catchment areas – so depending where you live you’ll tend to go to a particular school – and there’s different boundaries there. And that’s something local authorities traditionally haven’t published up here – at least to my knowledge. So that’s something we’ve been putting out, and that’s something that, quite understandably, has got a lot of interest. People are going to buy a house in a particular area or looking to move to a particular area would probably want a sense of what schools their kids are likely to be going to. So that’s been quite a popular dataset.
A certainly more niche case, one that’s been surprisingly popular: the City of Stirling published data on their cemeteries. The local authorities of Scotland tend to look after larger cemeteries, so Stirling Council decided to publish records of burial plots and where they are etc. What was interesting is that they actually got a few people get back in touch with them, and say things like:
“I think that information for that burial plot is incorrect, and that’s actually the wrong date…”
So in addition to re-use of the data, they’ve received feedback that can improve the quality of their data, just from publishing it openly – which is quite interesting as well. I think there’s often an expectation that if you publish open data you might expose issues with data quality, and there’s a risk that doing so is ‘bad’ from a reputational perspective – but this shows that there’s opportunities to actually improve the data quality by publishing it.
Paul: I think that’s a really important insight there. Maybe, there’s a tendency to publish data that you already know is of ‘perfect’ quality or has been verified – but some new inputs into that data can come only by publishing, and so it can be enriched and transformed through the act of publication and engaging with the audience. So this sounds like a really interesting scenario that arose from that dataset!
So, in relation to the open data publication work, what’s next for the programs that you’re working on?
Doug: Good question – I think the big focus at the moment is just getting more data out – at least for a couple of cities, a lot of the work has been just engaging stakeholders internally and saying, “look, we’re keen to publish data”, and getting through the traditional concerns around publishing data. For starters, as a result of this engagement we’re seeing a lot more data get published by some cities – on top of that as well, I think that we’re keen to carry out more of the collaborative common datasets work – that’s publishing more similar datasets, and looking at how we can publish them to similar standards – so similar fields, looking at metadata standards and making sure that we’re vaguely aligned there. So that’s something for us that’s been a bit more ad hoc – looking at the metadata quality. But again, the platform has been great for that. So several cities have made good use recently of the data dictionary function.
Doug: And the contexts of datasets – so that’s something we’re getting better at. So we’re publishing more and we’re getting better at it!
Paul: Great – so is there anything you’d like to see or anything you’d like to see developed as an offering going forward?
Doug: That’s a good question – it’s one of those tricky ones – if I went to the city leads and spoke to them, I’m sure they’d offer very different ideas. I think the main thing, I’m not sure if it’s more about CKAN’s core functionality or more of a module thing, is possibly some more upfront data previews. The current data preview functionality is quite deep, and if we want to publicise it more widely, future users will probably want to click on a dataset and immediately see some sort of visual representation. So that’s very specific, but maybe it’s having the option to say, “okay, here’s a json file – can we get a visual representation of that?”
Paul: I do hear things around visualisation – and a lot of that comes to building ‘showcase apps’ on top of CKAN and so on, where you’re inputting data from multiple datasets…
Doug: And that’s something we’re keen to make use of more as well. It’s not something we’ve done a huge amount of, but it’s certainly something we’re looking to pick up as we go forward and make more use of that functionality that’s already there.
Paul: Is there anything else you’d like to add to finish?
Doug: I suppose the main thing is that we’re all keen to see where CKAN goes from here, and what Datopian is up to going forward. It’s been really interesting to get involved at this stage where open data is already quite a mature concept – and it’ll be fascinating to see what the next stage is.
Datopian delivers outstanding solutions that enable your organization to realise your data’s potential. From hosted data portals powered by CKAN to specialised data engineering, from agile data practices to data strategy development, Datopian empowers you to transform data to insight.
© Datopian (CC Attribution-Sharealike (by-sa))