Anyone who has ever run a website — or has merely used one from time to time — knows what one of the key jobs is. It’s not getting a lot of pretty pictures up. Nor is it signing up as many banner ads as possible to pay for the site. Nor is it packing as much information as possible on the site’s home page. No, one of the most important jobs is getting a decent search engine. One that winnows down the original request and takes the user where he or she wants to go. It’s an apparently simple idea: but, as we all know, it’s one that many websites do poorly. It is clearly something that many companies find quite difficult. Okay. Now imagine you’ve got to effectively co-ordinate a search engine not just for a company, but for the entire public sector — central as well as local government. Suddenly the phrase “whole of government”, which has become something of a cliché in the mouths of some politicians, takes on a whole new and rather daunting meaning. As a briefing paper to ministers earlier this year notes — with an almost audible sigh — “The business of government is complicated, and that is why e-government is not a simple, straightforward task”. However, if the e-government project is to work, such a tool is essential. New Zealand’s e-government portal is formally being launched this month, and sitting behind that gateway to this country’s public sector is a couple of years’ work on getting the “metadata” issues right. The portal itself is vital to e-government. “The existence of an authoritative government portal, capable of supporting activities at all four stages of e-government [that is, from the provision of information, through online interactions, transaction-based services and the transformation of the design and structures of government itself] is the most critical part of the e-government infrastructure,” ministers were told in a Cabinet paper last year. The overall aim is to enable citizens to perform as many interactions with government agencies as possible online. A few key ideas sit behind this strategy, one of the most important being that the average citizen should not have to know which particular agency they need to go to — be that agency in central or local government. A further driving force is to reduce the compliance costs for businesses in their dealings with government agencies. The same Cabinet paper quote above uses an Australian study to show that, as a conservative figure, the average New Zealand business could save anywhere between $1000 to $5000 a year in more streamlined dealings with public agencies. Even if only half the companies in New Zealand use the portal, that is a saving to the economy of about $50 million a year. While Treasury has suggested those figures could do with a little more analysis, there is — at this point — no comparable New Zealand study. For the private citizen, the Australian study estimated the cost savings would be in the region of $NZ20 per user per portal transaction. The main benefits for the first five years will be in the wider economy rather than in cost savings to the government itself. At the centre of the development of the portal’s ability to search for government services is the NZGLS metadata standard — a system for describing government information and services. The development of this standard — “not very sexy, but vital” as one official puts it — has been carried out over the past two years between the e-government unit in Wellington and various other government agencies. The standard is based closely on two well-established offshore metadata templates — the Dublin Core Metadata Element Set and the Australian Government Locator Service. “We didn’t’ create this from scratch,” says Brendan Kelly, a programme architect in the State Services Commission’s e-government unit. Where the New Zealand standard goes further than those two, however, is in setting out a metadata template for services. “That is an extension on the others — they have only looked at provision of information,” says Kelly. The Dublin Core Group which devised and now maintains that particular standard is looking at incorporating the New Zealand approach. The other New Zealand innovation is the development of “subject” and “functions” theasauri. The range of information sources that have to be tied together in this standard can be gauged by the fact that there are 65 government agencies, and the e-government unit had, by the end of May, identified 1100 service descriptions and 2500 information resources. “That’s not a complete set, by a country mile,” says Kelly. “But it’s a core, it’s a start.” The development of the metadata matches noun and verb — for example, “registering a car” but it also builds in preferred terms and non-preferred terms for each activity, says the e-government unit’s relationship manager, Andrea Gray. That means a user who keys in a non-preferred word can be steered towards what they might actually be asking for. A further component has been the involvement of local government — often the Cinderellas of the public sector — in the project (see accompanying story). “We definitely wanted local government there,” says Kelly. A core group of councils held a preliminary series of meetings and discovered that about 90% of the local government services provided are the same, whether the council is in Rodney or Roxburgh. “The few differences might be in things like fire permits — different rules in different areas depending on climate and geography.” It was felt quite strongly that a government internet portal without local government would be a two-legged stool. Many citizens do not know whether the service they want is performed by central or local government. A core idea behind the portal — indeed, behind the whole government project — is that the citizen does not need to know whether the service he or she is after is carried out by their local council, the Inland Revenue Department, ACC or whatever. The development does provide some opportunity for revenue down the track. “The metaloge that has been developed — we own the code for of that,” says Kelly. “It is a unique resource and we will be leveraging that.” But the development partner was Gen-i. “We don’t want to be a vendor. That’s not what our job is, our job is to provide a government service.” Further private sector involvement was considered by the government but rejected by Cabinet. Other options included developing the portal as a public-private partnership — something ministers have talked about in general terms in a number of areas but have thus far not embarked upon. A further option was to use the metadata as a public resource but to provide it to various approved providers to encourage private sector development of New Zealand portals that include access to various government services. Yet another option was just to provide the metadata as a public good to all comers. These two appear to have been ruled out fairly quickly. “Neither … would ensure the development and implementation of a new government portal that would support the goals of the e-government strategy,” one Cabinet paper argues. “Nor would they ensure the existence of an ‘authoritative’ government portal that would not be associated with unrelated commercial activity.” And perhaps surprisingly, given that public-private partnerships have been touted as a way of reducing the cost of public sector infrastructure projects to the taxpayer, that latter option was seen as being more expensive than the government doing its own design and build. Ministers also considered charging for access to the portal, either using fees per transaction, or a subscription model. The transaction model would have involved charging other government agencies, rather than end users, and would have discouraged them from going online — not the approach the government wishes to take right now. The previous online access to government, the NZGO site, experimented with a subscription model but suffered a high number of users defaulting on payments. The cost of collecting subscriptions was also counter-productive. Another option was to upgrade the level of service provided on NZGO — which has been in existence since 1996. That, however, would have cost about half a million dollars, with a further half a million annually on operational costs. Lesser agency drives standards The public sector agencies have a kind of informal — but for Wellingtonians highly recognisable — pecking order. Treasury tends to be ranked at the top, with the mega-agencies of Social Development (formerly WINZ), Inland Revenue and Education next. Well down the bottom in any pecking order are the government agencies, known as “Crown entities”, without full departmental or ministry status. Yet one of those, the Land Transport Safety Authority, has played a fairly major role in development of the metadata that sits behind the e-government portal.“E-government is still only mandated for departments,” says LTSA chief information officer Tony West. “It’s kind of strongly recommended for other agencies, while local government is only encouraged to take part in the projects.”
Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.