Thanks to the Great Emancipator we have seen a US view of “Government-wide Information Sharing for Democratic Accountability“. The author, J.H. Snider, is suggesting using semantic web technology to monitor the connections between powerful politicians, public officials and lobbyists. He points out that these techniques are used on the weaker members of society, but the powerful will cite privacy and cost for not applying the same methods on their activities. He recommends that the President’s transactions are monitored in detail as a showcase for monitoring Congress and executive relationships. What are the chances of linking up data of the Prime Minister, Ministers, senior civil servants, leaders of councils and their suppliers of hospitality?
Well, there is the work going on in the Cabinet Office led cross-government information architecture. They are attempting to use standards to provide interoperability between different departments and agencies. Amongst these are the building of an “Upper Ontology for Operational Service Delivery” – perhaps this has the same intention as Snider’s “Who-What-When-Where ontology”. The USA has a more accessible name for what may be the same thing.
It will take a long time, if ever, to develop a lobbying information system. Even though some MPs are calling for the establishment of a register it would eventually have to be linked to lots more sources of reliable data.
Where Quarkside differs from Snider is in the use of Global Unique Idenifiers (GUIDs). They may think they have them in the USA, but it is not credible or politically acceptable in the UK. Reflecting back on a post from last year,
- A person does not need a Unique Identifier (UID).
- The Law does not demand a UID.
- Use just sufficient data to identify a person.
Openness in personal relationships can only built from an understanding of federated identity, multiple identities, not by demanding a UID. The likes of Experian can do it – so could UK plc. Maybe Liam Maxwell could assist here. Maybe we could also publish information about political lobbying that would improve democratic accountability.
Last week, Quarkside made the case for assertion based identity matching, avoiding reliance on imperfect UIDs. How come I am such a strong supporter of the Systems Interoperability Framework (SIF), which does rely on GUIDs (Globally Unique Identifiers)?
- You do not need a Unique Identifier (UID).
- The Law does not demand it.
- Alternatives are better.
This has developed into the ambitions of No2UID. Paradoxically, this does not mean that there is no place for UIDs. Take the example of the (SIF). The SIF specification requires every thing to have a Global Unique Identifier (GUID) – essentially a UID. The reason is simply that systems that use UIDs are here now. They do work and are simple to understand by school administrators using systems interoperating via SIF.
- If a child changes any personal information, such as subjects studied or contact telephone number. Then the same information is automatically available to any other authorised system. No data has to be re-entered into another system AND the data is exactly the same. It is more efficient and errors are reduced. Error correction is currently one of the most time consuming processes.
- The concept of one system providing a master reference record makes sense. The master record must be correct and school administrators are blissfully unaware of the background processing, linking via GUIDs.
- SIF has a hub at the centre of many systems, known as a Zone Integration Server, or a ZIS. A ZIS does not store any data; it only checks the data quality. It only orchestrates the transfer of valid records.
So I don’t discount UIDs. It’s all part of evolution. But I won’t say that UIDs are essential. No2UID solutions are not off the shelf and it will take several years to develop the toolkits.
Identity Documents Bill 2010-11. Thanks to Mike Ballard for the link and highlighting:
“A small number of provisions in the 2006 Act – unrelated to ID cards – reappear in the Bill. These cover offences relating to the possession and manufacture of false identity documents such as passports and driving licences. The Bill also re-enacts data-sharing provisions in the 2006 Act designed to verify information provided in connection with passport applications. “
Does that mean that the bathwater has gone and the baby remains – or vice versa? Every provision in the bill should be both challenged and supported – by different groups, of course.
No2ID has plenty to say. Far less ranting than I was expecting – measured and reasonable. “The Bill is a good start. With some bad features. Were clause 10 deleted and the re- enactments modified along the lines we suggest then we would support it wholeheartedly. But it is not sufficient to remove the threat of a National Identity Scheme from Britain.”
Clause 10 “… is a huge enhancement of the database state and mass surveillance. No case has been made for it. The most plausibly effective change in Home Office issue of passports, a check for a ‘social footprint’ via credit reference agencies, was introduced several years ago without it. The power is unnecessary and undesirable in itself, but in a broader administrative context it would facilitate the reconstruction of an ID scheme in a slightly different form, based on the passports database. This has actually already been proposed by several advocates of ID cards, including David Blunkett.”
So, the debate centres around the purpose and use of the Passport database. Re-use of data is an obvious way of reducing frustration in citizens who are repeatedly asked for the same information – and often suffer the consequences of inaccurate recording. An all-singing all-dancing data base is not an acceptable UK solution. However, it is perfectly reasonable to cross-check, and extract data from, high quality data sources such as held by the IPS. You don’t need a UID (or root identity) but identification will benefit from triangulation methods based on combinations of facts and assertions.