Northernlands 2 - Why is collaboration around data important in 2020 and beyond?

Description

Leigh Dodds shares his knowledge and experience about data collaboration

Transcript

This transcript comes from the captions associated with the video above. It is "as spoken".

Good afternoon everyone and welcome to this Northernlands

session on collaborating around data. My name is Leigh Dodds.

I'm the director of delivery at the Open Data Institute.

I'm going to introduce the session, put a bit of context to the

discussion and then we'll hear from 2 speakers: Sam Nelson from Open

Data Manchester and Lindsey Marchessault from the Open

Contracting Partnership. So the question that we are considering

in this session is

"why is collaboration around data important in 2020 and beyond?"

Well, we're living in

interesting times. We're dealing with a global pandemic.

Whilst being faced often daily with the issues and

problems of unethical collection and use of data.

All set against the backdrop of a rapidly changing climate.

So we have some big challenges to face in our

societies, in our economies, and in the environment.

Data can help us to address those challenges, but open only

if we are open and transparent in how that data is being

collected and used.

And only if that data is fairly representing the communities

that will be benefiting from it or be potentially impacted

through its use.

And we need to ensure that data is being used in responsible and

ethical ways. And to do that we need to collaborate. We need to

work together to try and maximize the value that we can

create by using data whilst minimizing its potential harmful

impacts. And it's only through that that we will create a world

where data works for everyone.

So in this session, we're going to look at collaboration around

data in two ways, firstly around collaborating on its collection

and use. So collection of maintenance. And secondly around

its publication and use.

So let's jump into the first of those - collaborative maintenance.

Now when were typically introducing open data, we focus

on the definition that open data is data that anyone can access

to use and share.

And while that is true and useful, it doesn't get into the

practice of open data which, even in 2020, is still largely

around public and private sector organisations publishing data

that they have already collected, so increasing access

to it to allow others to use the data in new ways. But it doesn't

have to be that way.

We can work together to collect and maintain data

and there are many projects that exists that have explored that

approach, so I think it's important to recognize that

collaborative maintenance is an option that communities can work

together to collect and maintain data, which is then are made

available under an open licence.

When we look at the kinds of work that is involved in

maintenance, it turns out that we can share quite a lot of it.

We can decide on what data should be collected in the first

place and how it should be

stored and managed. We can share the work of collecting

and maintaining the data so it can maintain its use over

in value over time. We can work together on managing access to

data and improving its quality.

And we can collaborate on building the tools and the

communities that can help us to unlock value from that dataset

once it's been collected.

So there's a range of different activities we can be involved with

So let's look at a few

brief examples. So the first and perhaps most widely known is

Open Street Map, which is very successful collaborative

maintenance project. Within Open Street Map, one particular

initiative is the Humanitarian Open Street Map Project, which

brings together local mappers working on the ground with

remote mappers, working with satellite imagery

to help to improve responses to disasters around the world.

By working together, those mappers can rapidly update and

improve the map in areas where humanitarian aid is needed to

ensure that aid is directed in the most useful and most impactful ways

Another example would be the Mozilla Common Voice Project

which is a project to try and develop a more diverse

and representative data set that can support the development

of voice recognition applications

So the project encourages people to contribute their

voice to the dataset and get involved in transcribing

recordings that have been contributed to it.

The third example we can look at is the work of Open Data Manchester

in Stockport. So let's hear from Sam Milsom about that project.

I'm Sam, I work for Open Data Manchester. We're a

not-for-profit formed from a diverse group of open data

advocates back in 2010, and we support organizations to release

data and we help people to use it. We build support good data

practice through expert advice, advocacy, events, research.

That sort of thing.

So back in 2019 part of the ODI's open GIS call, we ran

a project which we called Mapping Mobility Stockport which was a

program about trying to map the experience of people with

mobility and accessibility impairments. Looking at the roots

and the strategies that they used to travel in and around

Stockport Town Center.

Quite often these sort of communities can be overlooked in

town planning, and you know whether that's because the urban

environment deteriorates overtime or just doesn't account

for their experience. So it could be things like not having

dropped curbs at the pavement at busy crossings, for example.

These things, historically, can

be overlooked. And again, moving forward in terms of planning an

redeveloping the Town Center.

Quite often again, these things are quite hard to sort of map

and to take into account often because the GIS systems

themselves don't. They have a very sort of basic way of

describing these things like drop curbs or tactile paving,

but they don't take into account the vast range of experiences

that people have, so they might

not take into account quite acute neurological conditions,

for example. So how do you sort

of plan for that? So.

With Mapping Mobility Stockport we had some funding

from the ODI in which we sort

of developed a methodology for capturing and mapping this data.

So we worked with Stockport Town Center and some community groups

in Stockport, so Disability Stockport and also Age UK,

Stockport and we did some ransom workshops with them where we got

Maps of the Town Center from the council and we literally just

started scribbling on the maps. The routes that people took to

get where they needed to get to go around the Town Center and

started to identify

problem areas, but we also started to identify the routes

that people took regularly. Because quite often people do

have their own routes and strategies to move around the

Town Center, but these kind of remained tacitly within those

communities, so it was a really good exercise for us and for

Stockport Council to be able to actually map these routes and to

think about why do people take these routes? Why not this

particular route when it's more direct? Maybe because there's

you know it's too steep for a wheelchair to pass down safely.

That kind of thing. So it was a really, really good exercise for

Stockport Town Council because they were well,

redeveloping their, doing a lot of redevelopment of the Town

Center. The bus stop they are currently redeveloping. So a lot

of the experiences of the bus station area we... some of

the information that we captured has fed directly into the

redevelopment of the bus station

So making it more accessible.

With the workshops themselves, yep we started off doing

workshops where we just scribbled down problem areas.

But then we also went out onto the streets and conducted

walkabouts; 101 walkabouts with people so wheelchair users,

people with visual impairments, elderly and let people just walk

around and with a dictaphone and taking photos and we just took the

routes that they took and got them to describe their experience.

And in the end we were able to build a sort of

actual digital map

of these roots and of the good areas and of the problem areas.

And as I say, the project was really, really successful.

Stockport Town Center you know have used the data that we captured

and mapped in the redevelopment of their bus station, but

actually moving forward. At Open Data Manchester we've been

thinking about how can we capture these experiences?

Because there are so many more experiences and I think that a

lot of GIS systems don't have

the... let's say the language or the structures through which we

can describe these experiences. Quite often you're looking at

urban areas from the point of view of the buildings or from

the streets rather than the experience of the human being;

the embodied experience of the human in that space. So from the

Mapping Mobility Project what we're hoping to do is we

actually going to take the data that we collected with these

groups and start to build a framework. A sort of really

basic. I suppose trying to look at is there a... can we create a

sort of basic standard for describing experience, so

describing and mapping areas from the human being, the human's

perspective, and from that it can also incorporate a multitude

of different experiences. So we're actually hoping this year

to start developing that project further. So taking the data and

the stuff that we collected and developing a kind of prototype.

Whether that's going to be a

questionnaire or some kind of standard. We're going to sort of

hopefully work on that and develop it further, and to see

whether we can, you know

yeah, create a kind of a standard or a framework for capturing

Capturing more people's experience of moving around

urban areas which we think could make town centers, city centers

much more inclusive and just a lot safer and easier for all

people to move and travel around and live in.

So thanks, Sam. What's really interesting to me about that

project is not only were they able to create some local

impact by working with the local community on collecting

missing data, it's also led to some reflections about better

ways to collect data in future so that more... it's more

representative of those communities.

That's something that we're really interested in at the ODI

We recently published a guidebook to help people design and

build systems and tools that

will support collaborative maintenance.

We'd really love to get some feedback on this, so

please take a look and let us know what you think.

So let's move now onto the second way in which we can think about

collaboration. So collaborating on the publication and use of

data. And again, I think it's illustrative to look at what is

happening in the open source movement and think about how we

might apply that to open data initiatives.

So some of the more successful open source projects

have recognized that in order to be impactful, they need to get

people to contribute more than just code and bug fixes.

They need a wider range of skill sets. They need people to help

produce documentation. They need user researchers to make sure

that the software is well designed. They need designers to

help improve the user interfaces and the user experience of using

those tools. So there's been a move to try and get more

people involved in a whole variety of different ways.

I recently came across this framework I called BASEDEF that

had been developed by a number of people in the open source

community. It tries to spell out a variety of ways in which we

might contribute to

an open source project and in helping it achieve some impact

So we might blog about it to help other people

understand how it could be used. We might try it in a project so

that we can provide useful feedback about how it might be

used. We can file bugs. We can contribute bug fixes, but we

could also just work to try and improve the documentation to

help other people, perhaps with less technical skills to get

some value from the software.

Now, what if we applied the same framework to open data?

We could write about new datasets as they're published to

help others to understand both the value of the data, perhaps

also its limitations.

We could write tutorials and produce code that can illustrate

how a dataset can be used to help apps to show how it could

be used in specific tools, or to answer specific types of

questions. We can give feedback to the publisher, maybe to

suggest how they could improve, how the data has been published,

Perhaps to align it better with other datasets, or to adopt a

specific set of standards.

And through the freedoms that are given to us by open

licences, we can enrich and improve those datasets. Perhaps

to add in additional data or to link it with other sources.

And increasingly, we need to make sure that data is well

documented so we can all contribute to that. We can help

to fill in the gaps and support publishers in making sure that

their data is well described and

well published. Increasingly, there's a... there's a range of

open data initiatives that are realising that this kind of

broader support is necessary in order to achieve impact.

One example of those is the Open contracting Partnership.

So let's hear from Lindsey about the work

that they've been doing.

Hi, my name is Lindsey Marchessault and I'm the Director

of Data and Engagement at the Open Contracting Partnership

where we support governments and other stakeholders around the

world to improve their public services by improving their

public contracting. Public spending on contracts adds up to

more than 10 trillion dollars of public investment per year. So

it's very important to get it right. And by that I mean value

for money for government. Fair access to opportunity for

business and high quality goods works and services for everyone,

which includes everything from buying textbooks to maintaining

power plants. Unfortunately, in many cases, we don't have the

data that we need to ensure an effective procurement process

or to diagnose the performance of the system as a whole.

These data challenges might be because the data is not

collected, not digitized, or because of poor data quality.

Unusually, there's an element of collaboration in this.

I'll give an example. Currently in the COVID-19 emergency response,

many governments are currently purchasing or have purchased

ventilators. And the process to effectively buy ventilators,

involves a lot of needed data as input. This includes

understanding what is our current stock and supply of

ventilators. Where are the ventilators? Are the ventilators

functioning? Are they the appropriate ventilators that we

need? What do we expect in terms of the amount of ventilators

that we need and when we're going to need them and where

we're going to need them? What are best predictions and our

best advice? Finally, what do we know about the market for

ventilators and where are they

available? What are market reference prices and who might be able to

supply these things. Without the process to understand and

collect this data, digitize it, and use it effectively

the quality of the procurement outcomes will be affected. And

that's just one example of how data is important in all

different types of public purchases at all different

stages. Whether that's the planning of the procurement

the procurement process itself, or how we want to effectively

manage that that contract delivers value for what was

needed, and intended.

So this situation is why we support partners to collect,

publish, improve and use open data about the planning,

procurement and implementation of public contracts.

And to do this, we maintain a variety of tools and guidance,

most notably the Open Contracting data standard.

And we provide both policy and technical advice, including on

change management strategies. And we also provide a free

global help desk for the Open Contracting data standard in

both Spanish and English.

Over the years, we've supported hundreds of partners from dozens

of countries, and we are working to build an international

community of Open Contracting

practitioners. And one thing that we've learned is how

important multi-stakeholder collaboration is to

achieving measurable results. Collaboration has

been even more important in 2020, as the entire open

contacting community has focused on ensuring the

effectiveness and accountability of the

COVID-19 response procurement. So this is to

ensure we have the necessary medical equipment,

facilities and services to address the crisis.

And I'll share one example.

The government of Paraguay, through their procurement agency

wanted to ensure transparency of the COVID-19

emergency purchases. Because of the emergency government

agencies were allowed to buy quickly without a competition

without an advertised process. But just because something has

to be done quickly doesn't mean that it can't happen without

accountability, and so our helpdesk advised them on how they

could use the Open Contracting data standard to identify the

COVID-19 related procurement and ensure that it was referenced

to the COVID-19 emergency budget lines as well. Our

engagement leads for Latin America also worked closely with

the government and users of the data to ensure that the data was

being presented to people in a way that they could use it, and

one way that this happened was by a tool developed by the Inter

American Development Bank that we advised and supported on

which reuse the data and combined it with other

information including budget

graphical data and something called an "investment map" and the

data included information on the buyers, the prices, the suppliers

and information about the items and services being

purchased and their delivery.

And this data is being used by a wide range of stakeholders

within the country, including the government who improve their

own processes. By business to understand the market and the

opportunities. By civil society, academia and media. In fact,

journalists have reported about several irregular procurements

in the media, including a case in which large volumes of

tonic water were purchased at roughly five times the market

price as a COVID-19 emergency procurement, and as a result of

this case coming to light

the government issued a new requirement that buyers must

report and published their market reference prices to

justify the prices they're paying, which will hopefully

avoid other inflated emergency

procurement. So at the Open Contracting Partnership, we're

also supporting similar efforts to publish and use procurement

data all around the world. Some specific to the COVID-19

response, but also starting to prepare for the coming spending

associated with the economic recovery. And, for example, we

recently launched an action research program where we're

supporting researchers from 12 countries to use data and

investigate the affectiveness/ integrity of the COVID-19

emergency response procurement. And we hope that their findings

will lead to similarly

actionable policy recommendations which can be

implemented through collaboration with government

I'll stop there, but I'm happy to answer any questions

about our work, our tools, or our support.

So thanks, Lindsey. Again, I think there's something really

powerful there about demonstrating how, by bridging

between the publishers and the users of the data, the

partnership has been able to increase the impact, increase

the value that has been gained by sharing that data in the

first place. So to return to our original question,

It's because we need to work together to

get more people involved in understanding how

data is being collected. Get more people involved in

maintaining that data over the long term so that we have a

stronger data infrastructure. And we need to work together

as communities to make sure that we can maximize the value

of that data while minimizing its potential harmful impacts.

And through that work we can create an open, trustworthy,

data ecosystem.

So thank you for listening. If you'd like to learn

more about the work of our speakers or the work that we

have done at the ODI, then please follow the links in the slides

Thank you very much.

  • Leigh Dodds

    Director of Advisory
    The Open Data Institute

    Leigh Dodds
    © The ODI 2020

    Leigh has 17 years of experience working in a variety of technology focused roles including software engineer, product manager, technical consultant and CTO. Leigh spent 10 years working in the publishing industry dealing with data integration and management issues. At Publishing Technology whilst CTO of their scholarly division, Leigh was responsible for designing an innovative publishing framework based on semantic web technology. On moving to Talis Group in 2011, Leigh became responsible for programme manager for their “Data as a Service” products, overseeing product development and launching their Linked Data consulting business.

    Recently Leigh has been working as an independent technical consultant helping businesses explore technology and best practices that support the integration and publication of Linked Data and Open Data. Leigh has worked with a variety of organisations ranging from small startups through to large multi-national businesses. Leigh also enjoys writing and has worked as a freelance author publishing articles and training materials for O’Reilly and IBM.

Sponsors

Nothernlands 2 is a collaboration between ODI Leeds and The Kingdom of the Netherlands, the start of activity to create, support, and amplify the cultural links between The Netherlands and the North of England. It is with their generous and vigourous support, and the support of other energetic organisations, that Northernlands can be delivered.

  • Kingdom of the Netherlands