Local IT leader highlights redaction problem

Dane Wright, IT strategy manager for LB Brent, said the government's demand that public bodies begin to release details of all spending over £500 from the beginning of next year will place stiff demands on the redaction mechanisms of their IT systems. Speaking at the Socitm 2010 conference in Brighton, he described this as the most problematic issue facing organisations in complying with the requirement. "For example, the information in invoices is about payments to individuals, which is essentially personal data which is subject to the Data Protection Act," he said. "It has to be identified and the names and addresses removed. "Work needs to be done mainly on identifying payments at the outset to remove personal information." Wright pointed out that the requirement is not yet enshrined in legislation, but said organisations will be under pressure to comply and it could well become law in the future. Other issues include the fact that it does not as yet apply to Scotland or Wales – although they could well follow suit – and whether organisations will show a strong commitment to the principle of open data or do the bare minimum. He added that organisations should not necessarily stop at the level defined by the government, and should make most of their datasets available. The requirement is part of the government's transparency agenda, which also deals with making public sector information more easily available for re-use. Last month it published the new Open Government Licence to replace the widely used Click-Use license model. It covers information including that under Crown copyright, databases and source codes. "Licensing is crucial in all of this," Wright said. "There's no point in it if all the datasets are subject to different terms of re-use. "There has been Click-Use, Creative Commons and others, and now we have the Open Government Licence. We need that one standard licence that people can re-use; until then there's little reason for us to start to use it." Wright also emphasised the importance of making it possible to link different streams of data in order to make it more useful within the semantic web, which he forecast will be the main thrust of development over the next few years. "Linking web pages for readability will be one of the key things," he said. "It's so everything gets defined and has a unique identifier. But when you get there you want some form of standard format so there can be links to other unique identifiers." He also said that, while most datasets are currently published in CSV (comma separated values) format, it is likely that over the next few years they will move to RDF (resource description framework).