The Aspiration To Publish All Public Service Data Must Be Balanced

The publication of open data is not resource free, the aspiration to publish all public service data must be balanced by the resources needed to publish/sustain such data and a genuine benefit to the public. There needs to be a clear recognition that it will take some time and cost to deliver.

It is important to recognize that the publication of open data is not the only tool to deliver greater transparency and accountability in public services. A valuable addition though it may be, it should not become a substitute for existing democratic scrutiny and governance arrangements. Our experience of open data publication to date has been that any associated scrutiny has, at best, been superficial and at worst, unnecessarily confrontational and destructive. Reliance on open data publication alone, without proper contextual information and in-depth analysis has the potential danger of undermining the democratic process, unless proper controls are in place.

The notion that the greater publication of open data may be in the business and strategic interest of public service organizations has been underplayed; substituted by a presumption that some form of regulatory compulsion is necessary to ensure that this happens at all. The examples of local authorities demonstrate that progressive public service organizations are willing to progress towards greater transparency of their own volition.

There are data sets where a charge has been made for data in order to cover the costs of collecting and providing that data. There is a danger that the effect of removing the ability to charge in these cases either results in genuine hardship (particularly for smaller public service organizations) or the ceasing of the collection of that data as it is no longer viable. This loss of revenue could have a significant impact on organizations during a time when other funding sources are being cut. There appears to be a misconception that a significant number of Freedom of Information requests are for data sets. Although data sets are requested by some areas of the media and lobby groups, the vast number of FOI requests are for answers to specific questions. There is a clear distinction between the wishes of special interest groups and the media for sets of data and the general public who are asking for useful, meaningful information.

We welcome however the emphasis on the publication of new data. We welcome the commitment to provide greater guidance on assessing the balance between costs and benefits of publishing particular data sets, however we would suggest that it is linked to the existing public interest tests required within the FOI Act regarding the use of exemptions, in order to ensure that any guidance and the requirements of the Act are in harmony. There should also be a recognition that public service organizations should take into consideration the need for future open data publication when making future IT investments to ensure that systems and infrastructure are geared towards efficient publication.

There are genuine concerns that although personal data may have been removed from particular data sets, that through the aggregation of a number of datasets personal data may be revealed. Although the guidance issued on redacting personal information in FOI requests is helpful in this area, it would welcome more detailed work and clearer guidance on this issue to protect the public. Any proposed requirement, should take into consideration the resources available to undertake publication and the reasonable time that organizations may have to take to achieve this, to ensure that local priorities for front line services are not compromised.

The current time limits within the FOI Act provide a sufficient balance between the benefit to the public of receiving the requested information and the cost to the public purse of collating and providing that information. We do not believe that a higher cost limit for datasets is practical or proportionate. Our experience to date has been that in the few instances where datasets are requested as part of FOI requests, either we do not collect that particular set of information at all, or it does not hold it to the level of granularity requested. There should not be a mechanism that forces local authorities to create data sets by default as a result of individual questions. We do not believe that any additional cost burdens on public authorities are appropriate at this time.

It is essential that consideration of the cost burden of publishing a particular dataset is a fundamental part of the decision whether to publish, and that this decision should ultimately remain in the hands of locally elected officials. It does not believe that the proactive publication of datasets be made mandatory. It believes the decision of what to publish and when is best made locally, to meet the needs of the local electorate.

Additional costs of producing datasets being made by the requestor should not be allowed to distract a public service organization from publishing other, more useful data. Therefore there would need to be consideration of the best use of the resources available, and that decision should rest ultimately with the public service organization concerned. It is clear that the confrontational and sensationalist style of some of the media is in itself a barrier to the greater acceptance of open data publication. There needs to be a greater awareness of the real benefits, this means greater persuasion, greater awareness raising rather than the introduction of additional regulatory burdens and an imposition of a “blame culture”. Current systems used for collection and storage of data were not designed or implemented with publication of data in mind, and therefore considerable work may need to be commissioned not only to improve the reporting capabilities of the system but also changes to business practices to ensure that the right information is collected. Proper safeguards need to be in place to ensure that the appropriate balance is in place regarding the cost of collecting/publishing the data and the actual benefit to the public at large.