Thanks to Rollie Colle for bringing a new development at the US Department of Labour to my attention.
eWeek.com introduces the development as follows:
The U.S. Department of has launched a set of new online tools for software developers to tap into DOL data for use in applications.
The DOL’s new first-of-its-kind federal Website is aimed at making it easier for software developers to incorporate Labor Department data into online and mobile applications. The site features published APIs (application program interfaces) and SDKs (software development kits) that enable developers to remotely access data collected by the department.
The tools, available at http://developer.dol.gov, are part of the Labor Department’s ongoing efforts to increase transparency, participation and collaboration through the administration’s Open Government Initiative, Labor Department officials said.
The eWeek site goes on to quote DoL as follows:
“While a handful of other federal agencies are making data available through one or more APIs, the inclusion of SDKs is a federal first,” said Deputy Secretary of Labor Seth Harris, in a statement. “By doing so, we’re lowering the technical barriers and providing developers of all experience levels the opportunity to turn good ideas into powerful software applications for the American public.”
What I find so positive about this is that the DoL is taking the issue of a potential Data Divide seriously and is devoting some of its development resources to responding by providing tools that those with more limited technical experience can use to design applications for using DoL data. (If this were happening a couple of months from now I’ld even try to claim credit based on my suggestion that 10% of the investment in an Open Data initiative be assigned to ensuring a basis for broad opportunities for “effective data use” which this clearly seems to be. 
)
Also, following on from my earlier blogposts on the Data Divide and Open Government Data, Tracey Lauriault introduced herself to me and I discovered through her that there was a very active “community data movement (consortium” and a new NGO (Community Data Canada) right under my nose here in Canada. Tracey has done an excellent but as yet unpublished report documenting some of this but I’ld like to point to two specific cases that she highlights, in advance and then I’ll at a later point, when Tracey’s paper is out, do a longer discussion.
The sites are EMIS — Espace Montréalais d’information sur la santé (in French) and Saskatoon’s CommunityView Collaboration. Both of these are designed to provide opportunities for active data use and some manipulation by the non-specialist end user.
The Saskatoon site particularly, provides effective (video based) training programs for the reasonably data literate user. Through these there is access to interactive maps and other data sets specifically of interest to those at the community level (primarily I would guess professionals like planners or social workers) interested in local level planning and particularly social planning. Not surprisingly the site was developed by the local social planning council and is an extremely good model for how to start some of those local processes of data use.
The EMIS site is sponsored by the Province of Quebec and focuses primarily on health data. This site probably provides more opportunities for “activist” use particularly around health and environmental issues but doesn’t seem as easy to manipulate or as visually user friendly. As well, there isn’t so far as I could see, any indication on the site as to how the site might actually be linking in practice with the broader potential user community in Quebec.
What is interesting about all three of these sites is that they are taking the issue of making their data available and usable by the non-professional end user very seriously and are putting significant resources in that direction. Tracey tells me that in fact there is something of a “community data movement” in Canada which has been around for a number of years (focusing particularly I believe, on maps and mapping). My plan is to start monitoring those developments more closely and hopefully start to bring about some sort of active linkage between the community informatics community and the community data communities as their interests and objectives seem to be converging at a fairly rapid rate.
One thing that all three sites seem to lack is an evident involvement by end users in the design of the data, the sites or the tools being presented through the sites. It would be very interesting to, for example, work with a local anti-poverty or anti-homeless activist group (in the case of the Saskatoon site), health or environmental community (or activist) organizations in the case of the EMIS site, or labour or organizations of the unemployed (in the case of the DoL site) and have their input into the design of the site, the selection of the data being provided, the design and selection of the tools and so on. Any of this may have occurred on any of these sites but there isn’t, based on my casual review, any evidence that this has happened. My guess is that the result might have been even more user useful than what is there now, although in all three cases these are excellent beginnings.
One final observation based on a review of the above is that expecting governments to take Data Divide considerations into their planning should be much less costly and onerous than for example, engaging them in responding to Digital Divide issues. Developing applications or application support software, community/grassroots oriented data access/use websites, non-professional oriented video training programs all go quite some way to overcoming the Data Divide while being very much cheaper and easier to accomplish than the quite significant infrastructure and site developments that have been the primary approach to resolving the Digital Divide in many regional and national environments.
I would be very pleased to get information on any other sites or strategies for responding to the “Data Divide” as others might come across these.
timgdavies
July 21, 2011
Hey Michael,
Interesting to see. From a quick look at the DOL site, I’m not sure though they’re really focussing on end-users: rather, they have a pretty complicated and not fully open (you only seem to be able to get the data (a) from their API; and (b) after registering to access it) API, and they’ve provided some code resources for developers to build things with this data.
The average user who wants to look up key data and information still needs to wait for a developer to build something that will meet their needs, and even then, unless the DOL site takes care to make sure it links to those apps, users will have to discover those ‘infomediary’ services.
Providing support for developers to use open data turns out to be quite important (as lots of the things between having data, and building applications, are common tasks that it doesn’t make sense for everyone to duplicate. That’s why with projects like AidInfoLabs we’re trying to collate and document all the different source code that’s been written for working with the International Aid Transparency Initiative Data as open source (building on insights that came from writing this paper about the importance of shared source around open data).
But focussing on users needs to go beyond that. Including to the sort of community engagement that the Community Data Movement has been exploring.
A few of the approaches I’m exploring to bridge the data divide:
#1: The Open Data Cook Book – collating user-friendly ‘recipes’ for working with data – trying to show non-technical users what can be done with raw data using free tools in a step-by-step way.
Not sure we’ve quite got the tone and pitch right yet – but ultimately if data is to be democratising we need to at least widespread basic skills for working with it, not just new infomediaries sitting between data and democratic debate.
#2: The IATI Explorer Toolkit – there’s lots of work to do on this before it is really user-friendly, but we’ve been trying to put together a suite of tools that allow people to get hold of International Aid Transparency Initiative data in the form and format they want – even though the original data comes as fairly tricky-to-work-with XML.
So far we’ve got a tool that allows users to query and explore sub-sets of the data, and tools that allow people to get at flat spreadsheet CSV versions of the data. And I’ve been trying to write some recipes on aidinfo labs for how to use tools like Google Refine to explore the data.
#3: Open Personae – a very early stage project for aidinfo labs where we’re putting together a collection of ‘personae’ (imagined profiles of potential users) to help us think about whether the different applications, tools and services being developed to use aid data are meeting a diverse range of needs.
In the long run I’d like to think about how we can get virtual panels of users engaged in reviewing proposals and plans (it would be interesting to think about the 10% funding you suggest being spent by users for their benefit, rather than on what organisations think users need), but the personae I hope will give us a first step to bringing a clearer notion of users into our discussions.
#4: Curating data use – trying to make sure we think about linking from the source of data to tools, analysis and stuff that works with it.
Not big projects, but rather, trying to find small approaches to slant what we’re doing more in a user-centred direction. Really look forward to getting inspiration from other initiatives…
Michael Gurstein
July 21, 2011
Thanks Tim for your very very useful comments. I think that these comments take the discussion in the direction I would like to see it going which is towards some practical specifications on how to make government data most easily useful to the public (as the end user). Your focusing on the software developer as the intermediary here is I think crucial.
Only one other comment which is that while I’m sure you are right in you assessment of the DoL initiative, what I was reacting positively to was the fact that they were taking the need for ensuring that their data would be usable and useful to the less sophisticated end user and appeared to be making some steps in that direction. You are much more qualified to assess the significance and effectiveness of those steps than I am but I think that they are even making these kinds of public declarations is very useful and can be pointed to in other contexts as the kind of commitment that should be made in these circumstances (as well as something against which they themselves can be held to account).
timgdavies
July 22, 2011
Thanks Michael.
Just in case I didn’t capture it quite in the comment above, I think my view is that software developers are one amongst many possible intermediaries of data: and that we need to make sure we are creating eco-systems where data is accessible, and skills are available, for a diverse range of actors to find what they need in data.
Tim
Daisy
July 21, 2011
Anything to diminish the technical divide is a good thing. It’s interesting that Canada is able to promote health care and environmental activism–something many would argue is necessary in the US
Michael Gurstein
July 22, 2011
Good point Tim, I should have said the software developer would act as “an” intermediary, not as “the” intermediary since in an ideal situation/ecology there should be a number of intermediaries with different roles/skills–content skills, technical skills, social animation/organizing skills and so on.