One server for open govt data mooted
Subscribe now for $100 (23 issues) and save more than 37% off the cover price!
Get the latest news from Computerworld delivered via email.
Sign up now
A single server for all government “open-data” requests is the goal of a group of government agencies led by Land Information NZ (LINZ).
Last year, the Declaration on Open and Transparent Government formally encouraged government agencies to release non-personal, potentially high-value data to the public.
However it’s one thing to release data and another to make it conveniently accessible and usable, says Matt Amos, senior technical advisor at LINZ. “What’s needed is to have a coherent view of all related datasets that are available and have those kept up-to-date as frequently and as promptly as possible. We should have a one-stop shop, or if that’s not possible, at least have a series of interoperable platforms and tools,” he told the Digital Earth Summit in Wellington last week.
“In terms of a common platform, a single source would be ideal,” Amos says, but if that can’t be done a set of standards should be in place so as many separate services as possible interact with a central platform.
Tools should also be provided “to enable the organisation that provides that data to get it onto the platform in a useful way, as simply as possible”, he adds. “That’s really important.”
The Canterbury earthquakes were a wake-up call to establish a coordinated, accessible and usable data source, he says. “There was a great response to a call for data in the wake of the earthquakes. However it posed challenges for data quality, accessibility and availability. Such failings are felt most keenly in times of crisis.”
A key goal of the proposed service is to share the investment across government agencies. “Producing services to release government data is really expensive,” Amos says, “and if each agency did it themselves it’d be a waste of money.
“We want to simplify the process of releasing data and also make the data more easily discoverable,” he says. “If it’s all in one place, it’s a lot easier for people to find, rather than negotiating with the bureaucracy of government agencies that are restructuring all the time.”
Data formats should be as open as possible, he says, and the metadata that assists in discovering and assessing the relevance of the data should also observe open standards.
“We’re planning to build on the experience of the Linz data service – a service Linz developed a year ago; it pre-empted the Cabinet [open data] directive and has been fairly well taken up by the public.”
The ICT Council – a group of CIOs from government agencies – has supported the group in this work and has gained their provisional approval, Amos says.
“We’re going to do some consultation [with potential providers and consumers], prepare a business-case and present it back to the ICT Council by the end of this year for final approval early next year.”
The working group is made up of representatives of 10 agencies, including a local-government representative “and potentially someone from the social development sector”, because social-sector data is particularly important to government decision-making, he says.
The group anticipates that the server will provide data either for download into a consumer’s own computer system or for access through a web service “so you get the advantage of the data being continuously available and updated as often as possible”.
A central server should be reliable enough to be the only place the data is stored, removing pressure on agencies’ individual storage resources, Amos says.
“We’re hoping to save money by building one slightly larger system a whole lot more cheaply than 50 smaller ones.”
A Notice of Information has been placed on the government’s electronic tenders site (GETS) asking potential suppliers of technology to register their interest.
The notice calls the project Open Data Service (ODS). It “seeks information from prospective suppliers to help the ODS project to: Identify options for government ODS solutions; understand market capabilities and opportunities; provide inputs that will help identify whether or not there is a viable business case for government ODS solutions; and inform the construction of a business case, if one is viable.”
Posted by Andy at 9:31:14 on September 12, 2012
Posted by Chris Faherty at 11:27:26 on September 11, 2012
Posted by Anonymous at 10:16:50 on September 11, 2012
Posted by Anonymous at 9:25:28 on September 11, 2012
There would be a cluster of servers for the service, and we would expect it to be running off the AoG IaaS platform. It would contain the DR and redundancy necessary for the grade of service required. It is likely to be one of a small number of Open Data Services and rather than one single one for all government data, this one would be focussed on non-private location based information. However, that alone prevents duplication of potentially dozens of services being stood up just for that data alone.
Posted by Matt Amos, LINZ at 9:22:45 on September 11, 2012
I never thought that the agencies concerned would house the function in a single hardware "box"; but I appreciate my use of the word "server" could give that impression. "Service" would have been a better way to put it.
Posted by Stephen Bell at 5:50:34 on September 12, 2012
Posted by Anonymous at 12:44:04 on September 11, 2012
Posted by Matt Amos at 9:24:32 on September 11, 2012
... Or perhaps maybe the the govt is stupid enough to put all their eggs in one basket?
Posted by Anonymous at 8:31:17 on September 11, 2012
Posted by Anonymous at 9:12:28 on September 11, 2012