Are the Open Data Warriors Fighting for Robin Hood or the Sheriff?: Some Reflections on OKCon 2011 and the Emerging Data Divide

Posted on July 3, 2011


(I’ve taken the liberty of shifting the continuation of this most interesting discussion (below)  and particularly my response to Peter Murray-Rust’s comments to a new blogpost.)

I spent the last couple of days at a fascinating (and frightening) event in Berlin—OKCon—a convention for the (in this case mostly European) uber-geeks who are in the process of recreating governments and potentially governance itself in Western Europe (and beyond).

The ideal that these nerdy revolutionaries are pursuing is not, as with previous generations—justice, freedom, democracy—rather it is “openness” as in Open Data, Open Information, Open Government. Precisely what is meant by “openness” is never (at least certainly not in the context of this conference) really defined in a form that an outsider could grapple with (and perhaps critique).  Rather it was a pervasive and animating good intention—a grail to be pursued by warriors off on a joust with various governmental dragons. Their armaments in this instance (and to an outsider many of them are magical indeed) are technical skills and zeal sufficient to slay any bureaucrat or resistant politician’s rationalizations and resistances to being “open”—i.e. not turning their information treasure chests into universally accessible nodes in a seamless global datascape.

If I seem a bit skeptical/cynical – less than true believing – its not because I don’t believe in this goal of “openness” (who could be churlish enough to support things that are closedclosed systems, closed doors, closed minds—you get the picture), its just that I see a huge disconnect between the idealism and the passionate belief in the rightness of their cause and the profound failure to have any clear idea of what precisely that cause is and where it is likely to take them (and us) in the very near future.

To start at the beginning… the “open data/open government” movement begins from a profoundly political perspective that government is largely ineffective and inefficient (and possibly corrupt) and that it hides that ineffectiveness and inefficiency (and possible corruption) from public scrutiny through lack of transparency in its operations and particularly in denying to the public access to information (data) about its operations. And further that this access once available would give citizens the means to hold bureaucrats (and their political masters) accountable for their actions. In doing so it would give these self-same citizens a platform on which to undertake (or at least collaborate with) these bureaucrats in certain key and significant activities—planning, analyzing, budgeting that sort of thing. Moreover through the implementation of processes of crowdsourcing this would also provide the bureaucrats with the overwhelming benefits of having access to and input from the knowledge and wisdom of the broader interested public.

Put in somewhat different terms but with essentially the same meaning—it’s the taxpayer’s money and they have the right to participate in overseeing how it is spent. Having “open” access to government’s data/information gives citizens the tools to exercise that right.

And (it is argued), solutions are available for putting into the hands of these citizens the means/technical tools for sifting and sorting and making critical analyses of government activities if only the key could be turned and government data was “accessible” (“open”).

Through partially technical and partially political processes of persuasion, lobbying, arm twisting and ultimately policy development and intervention, governments everywhere are in the process of redeveloping internal technical systems so as to make at least some of their information available –opening  this to the folks such as those attending this conference to work on and design means to make useful and accessible.—and the conference heard from enthusiastic young people who are effecting these changes in various parts of Europe, the US, Brazil and so on.

A lot of the conference took place in specialized workshops where the technical details on how to link various sets of this newly available data together with other sets, how to structure this data so that it could serve various purposes and perhaps most importantly how to design the architecture and ontology (ultimately the management policies and procedures) of the data itself within government so that it is “born open” rather than only liberated after the fact with this latter process making the usefulness of the data in the larger world of open and universally accessible data much much greater.

Again, so far so good… But as I sat through the first day of the conference and as the time came for my own presentation I suddenly realized that there was a dog, and a very large and important dog that wasn’t barking… During that first day and with only one or two exceptions on the second what I didn’t hear even indirectly was a discussion of who the ultimate users would be of this data (the beneficiaries of this ”openness”) and what ultimate uses this open data was being designed towards.

Some might wonder why I think that this non-barking dog is of such significance—why does it matter who the user is—what is important is that they/we have access to the data and the best approach is to effect a design that “anyone” can use i.e. for a universal user—the argument being that what is being built is not a vehicle but a platform and it doesn’t matter who the drivers are as long as everyone can use the highway.

So, in the absence of any articulated expression of who the (assumed) user is let’s speculate a bit about who this phantom figure might (or might not) be.  Given that in instances like these one tends, in the absence of other influences, to default to the known and familiar. Thus here one can almost certainly assume that the user is expected to be more or less like the folks in this room–young and bright, speaking English well, very well educated, overwhelmingly male, few or no minorities of colour or race, with firm middle class backgrounds, very very technically skilled and with the set of values and assumptions that go with the above i.e. strongly individualistic, slightly competitive, and not suffering fools (or the non-technical) easily.

My assumption then is that the anticipated user for this “open data” and for the kind of measures (policies, procedures, programs) which are being lobbied for and designed into government policy and practice looks very much like these folks at this conference–which scares me a very great deal…especially when combined with the VERY fragmentary evidence that is coming out on who is actually using this “open data” (which corresponds quite closely to my assumption) and what benefits are being realized as a result of its use.

Perhaps the most significant example to date of a national “openness” policy is the Government of India’s Right to Information law which by any standard is one of the strongest pieces of legislation supporting “open” government anywhere in the world. But it turns out that there were  flaws (and it appears quite fatal flaws) in the legislation/program that are now coming to light—the most significant of those flaws being a lack of enforcement mechanisms and perhaps most importantly the lack of a strategy for widespread broad based implementation focused on the end user.

What has happened in India is that by making the (quite false) assumption that the end users i.e. citizens would have the means to use this law to realize their right to information without additional support or intervention India has created a circumstance where citizens themselves need to engage in an often quite unequal struggle to access and use the information and the result has been a rash of murders of those wishing to use the information to expose corruption, self-dealing and misuse of pubic funds.

The legislation did not provide mechanisms for enforcement and thus individuals and groups had to take it upon themselves to attempt to gain access to desired information through individual action.  Thus rather than having legislation that focused on the potential end user in their Indian multitudes it simply provided for a notional “access” and left the rest to the individual citizen with these results:

First Right to Information Murder

Another Right to Information Activist Murdered in India

Three Right to Information (RTI) activists were murdered in this country in three months

But why should this matter to these enthusiastic young people five thousand miles away from village India.  Well if we take a look at one of the very few detailed studies of the end users (Escher) of an “open data” project (and a project that has been reproduced in a number of other national jurisdictions) that of the online citizen democracy tool we begin to see a pattern:

The overall demographics of these users extend the traditional biases in political participation:  In the “ audience people above the age of 54 tend to be over-represented, while those younger than 45 are under-represented in comparison to the Internet population. In terms of demographics there is a strong male bias and a strong overrepresentation of people with a university degree that also translates into strong participation from high income groups…One in five users (21%) of the site has not been politically active within the last year,” This means, if I am understanding this that 79% of the users of this site (and the related expense information) have been politically active within the last year!

So this attempt to enhance democratic participation has ended up providing an additional opportunity for those who already, because of their income, education, and overall conventional characteristics of higher status (age, gender etc.) have the means to communicate with and influence politicians. The additional information and an additional communications channel thus has the effect of reinforcing patterns of opportunity that are already there rather than widening the base of participation and influence.

Similarly with the case that I quoted in an earlier post which examined the outcome of a program to digitize land records in Bangalore and which had the quite perverse and unanticipated effect of providing a means for the wealthier land owners to extend their holdings and thus their wealth at the expense of the poor because they had the knowledge in how to use the information newly made available as well as the resources to hire the professionals to help them interpret the information in the way which was most immediately useful.

Thus it matters very much who the (anticipated) user is since what is being put in place are the frameworks for the data environment  of the future and these will include for the most part some assumptions about who the ultimate user is or will be and whether or not a new “data divide” will emerge written more deeply into the fabric of the Information Society than even the earlier “digital (access) divide”.

In each of these instances, by NOT paying attention to (and thus intervening to redefine) who the ultimate users of the “open data/information” would be, the effect has been to reinforce or even extend existing structures of power and influence rather than to have this newly open data be the basis for more inclusive and democratic participation. In the absence of making explicit the model of the ultimate user and thus designing appropriate processes of opening the data and making it available for the widest (and least implicitly discriminatory) range of users, the result will be as we have seen which is a user who is already in a position to make use of the information because of prior existing skills, knowledge, power, or status.

For these processes to NOT have these outcomes the data designer must base his work on an implicit model of a user who is NOT technically skilled, who is NOT financially well off, who does NOT have the characteristics of colour, gender or class which automatically gives them influence and power.

I’ve dealt with the matter of how to ensure opportunities for a broader base of effective use (and users) elsewhere but in this context as a recommendation to the folks espousing and doing Open Data could I suggest that there be a formal commitment to devote 10% of project (and programme) resources including time and funds to ensuring Open Data use by groups and individuals who are not technically skilled, are not middle income and above, who are not currently active in the political process but who might ultimately make the most beneficial use of the resources now being made available.

For anyone interested in my thoughts on the “hot to’s” of this I can refer them to the earlier blog post an edited version of which appeared in a recent issue of First Monday.

(Anyone re-reading this post may notice that I’ve removed my reference to “World of Warcraft warriors”.  In light of the recent terrible events in Norway and the association of the perpetrator with WoW I think any suggested link of the extremely well-intentioned activities of the OGD activists with WoW, even in jest, is I think very inappropriate and highly misleading!)