This site was developed to support the Government 2.0 Taskforce, which operated from June to December 2009. The Government responded to the Government 2.0 Taskforce's report on 3 May 2010. As such, comments are now closed but you are encouraged to continue the conversation at

A League ladder of PSI openness?

2009 September 19
by Nicholas Gruen

I attended a function at Parliament House on Wednesday night which was designed to showcase a number of things Google have on the boil, not least the usefulness of their product offerings to governments. Such things as Google Apps and Google’s general capacity to deliver the cloud not just to retail users but also to corporates. (Alan Noble attended today’s TF with a Google bumper sticker on the cover of his laptop which read “My other computer is a data centre”.)

I also got a better look at Wave. It looks more intriguing the more I see of it.

Google is making good progress getting hold of data to make its products – particularly Google Maps – even more useful, but it’s also hard for them not to be frustrated by the silly things which mean that data that you and I have already paid for governments to collect, data collected with the sole purpose of generating public benefits, is not simply, easily, quickly released into a serendipitous world in which we find out (so often to our own surprise) how useful it can become. Google can’t get good data on toilet maps despite the Federal Government’s having it (we earmarked this as an early win when our Taskforce began but at the halfway mark, I guess its status as an early win is running into ontological trouble.  Now we’re fairly confident of a ‘better late than never’ win).  And Google can’t get good data on the precise location of bicycle paths.

Then I had an idea. Since I conveyed it to Google, it seems only fair that I convey it to you. Why doesn’t Google report on governments’ preparedness to release data. It could produce a methodology and apply it consistently. Since Google Maps is an Australian originated product it would make sense to develop the methodology here where it could be applied in ‘beta’ form to Australia’s state governments.

One thing I’ve observed is that State Premiers like to claim that their state is the best or one of the best at something. State Oppositions also spend their time drawing attention to the ways in which the government they are opposing is sending their state to the dogs, choosing whatever comparative stats demonstrate their government’s relative under-performance. And of course there’s no reason to stop at state governments. National governments could also be compared.

The political obstacles to releasing most public sector information are not ‘hard’ ones. By that I mean that in almost all cases, releasing the kind of information that Google is after is not like raising a new tax or closing a school. The reason that the information has not been released is just that it’s a lot easier for a lot of people for it not to be released. Releasing it may involve legal advice and changes to copyright policy. It may involve some cost or inconvenience. And who knows what the information might be used for? Can officials be sure that the information can’t be used to embarrass them or the government? Usually they can’t be sure, and so they decide they’d better be on the safe side. So the reason the data is not being released is not because there are any clear political roadblocks to its release, but because any decision to release it must run the gamut of the that dark dank place which I call the Hall of a Thousand Cobwebs where so many worthy proposals die slowly, quietly and anonymously.

When there’s no political ’story’, no transgression by the government in not releasing the data it’s just so easy not to – even if it’s no real life or death issue for the government. But given the ’softness’ of the obstacles to release, the counterweight provided by a league ladder of openness of public sector information could often provide sufficient visibility for the issue  to make a substantial difference.

I expect that Google would probably like someone else to run such a league ladder – like Transparency International.  But in the spirit of releasing early and releasing often, it seems to me that Google (or anyone else who wants to) could get this ball rolling fairly quickly with a view to handing it over to others once they were ready to carry the baton.

Your thoughts?

21 Responses
  1. 2009 September 19

    Mash the State comes to mind, but that is a goal orientated effort rather than a league table. With the diversity of government in Australia and around the world could a league table get complicated? Perhaps as a starting point a simple 5 step Open PSI maturity model might work? Then we just need to agree on the broader patterns of good practice and policy. And of course find someone to evangelise it.

  2. 2009 September 20

    Surely there’s a way to think about this that doesn’t require rushing into the soft embrace of Google?

    Google can easily get good geographic data such as toilets and bike paths, but it has chosen not to because it views open licenses as a poison pill. OpenStreetMap (.org) is the de facto locus of open geographic data activity on the web today, and its license – currently CC-BY-SA, ODBL soon – makes clear that information provided to OSM will be and remain useful to a wider community in perpetuity. Google has chosen to ignore OSM in favor of its own homegrown MapMaker initiative, which borrows techniques and terminology from OSM while ignoring the complementary commitment to data openness.

    Encouraging projects like OpenStreetMap in preference to companies like Google would be a long-term win here, potentially requiring more up front time and organizational energy but ultimately resulting in superior public benefit. One way to encourage grassroots development of this sort would be to address the perceived trustworthiness of projects like OSM with an explicit way for governments to verify and sign geodata in open databases.

    Check out this related comment thread between Google’s Ed Parsons and OSM’s Steve Coast and Richard Fairhurst (and others): – they’re arguing about aerial imagery licensing and derivative data rights. Google is being accused of a large degree of mushiness in its licensing terminology – refusing to specify what constitutes a “mass” derivation despite using loaded terms like “liberation”.

  3. 2009 September 21


    If you want someone to set-up a league table of Australian government openness, why not make it a funded Taskforce project?

    The funds would kick-start a not-for-profit or group of individuals who already wanted to do something like the league table, but didn’t have the time or resources to spend on it alongside other commitments.

    Unfortunately in Australia right now there’s not much in the way of funds available for people creating these types of online initiatives.

    That’s possibly why we don’t have our own version of the Sunlight Foundation yet – which has an impressive list of benefactors. I don’t know if many Australians would invest money in such an endeavour – or maybe they’ve just not been asked yet.

    • 2009 September 22
      Alan Noble permalink

      I agree that an independently funded non-profit makes a lot of sense.

      Any takers?

      • 2009 September 25

        Alan, I am currently working on setting up a Charity Foundation to cover this. More details will be available to you most likely sometime on Monday or Tuesday.



  4. 2009 September 21
    simonfj permalink

    Michal M hits this on the head.

    Most of the Published Recommendations which came from one GLAMwiki event can be said to apply to all publicly funded sites. The main two being:
    * Use a “free-culture” Creative Commons license (either CC-by or CC-by-SA) for content on and websites.
    * Publish stable and clean URLs for individual item records in collections, incorporating persistent identifiers.(so links don’t rot)

    The only way to handle this easily is to accept that governance is about making rules to which exceptions can be handled, and we have to choose between an open culture or a (semi) closed one. It’s been interesting to see institutions like the Smithsonian, when a new director walked in and said, “get everything up”, go through the flip. The screaming and wailing about “it’s not ready”, “it’s not complete” followed a little later by “yes, that was my idea”. So let’s just have the open policy, please. The Minister can say it is his idea now rather then later.

    Nic, you gotta ask yourself, if you really believe that “Google Maps is an Australian originated product“. (It’s not of course, It’s a product originated by an Australian working for a US multinational). Why is it that Australians always have to go OS to prove their ideas are world leading?

    I think you’ll find it’s because they are trusted by citizens and small business to provide them with solutions. But only a Multinational has the time and money to snow our poorly educated reps and bureaucrats, who seem to think they should politik rather than govern. You’re suggesting they might as well ask Google to do it. (Hmm, not a bad idea after all)

  5. 2009 September 21
    Brad Peterson permalink

    I wouldn’t be so hasty about rushing towards Google as the be all and end all of everything. Michal and Simon make a good points. We don’t need to rush overseas into a cloud (pun intended) of fuzzy licences when we can do it at home and keep it truly open.

    Another issue is that Google is welded onto its text search algorithm which supports, let’s be honest, “dumb” searches and completely ignores semweb which allows almost natural language search queries. We are talking about linking people and data. We need to be RDF aware. Google isn’t.

  6. 2009 September 22

    I see the fact that Govt sees fit not to publish [online] its spatial data as an opportunity for the private sector to step up and invest in novel applications. Equally, I am not comfortable with that always seeming to default to Google. Who wants annoying popup adverts on their maps in the future when Google so decrees? However, Google are getting us comfortable with webservices which is a great contribution.

    We at NuMaps ( publish ABS Census data via OGC webservices so that people can get this data in a convenient way. We also build analysis applkications targeted at users who do not have spatial skills. In some cases that’s back into Govt because they seem to prefer to inefficiently replicate the data [at least this data set] across many stovepipe divisions – the duplication is horrendous!

    We are only a startup but I think this is a great example of the public sector operating as a wholesaler (data manufacturer and maintainer) and the private sector as a value added and applications provider. Conroy would be pleased!

    • 2009 September 22
      simonfj permalink

      Thanks Brad,

      I was hoping an Aussie startup would give us some idea of “the model” in action. OK, so we can see that it’s providing the raw materials and privates doing the value add. If you have any other sustainable model I’d love to hear, cause I can’t find another.

      But you gotta give our intrepid leader credit. He’s a terrier, amongst the stovepipes and closed culture of Canberra. At least we can understand the real reason for them.

      Can officials be sure that the information can’t be used to embarrass them or the government? Usually they can’t be sure, and so they decide they’d better be on the safe side.

      Can you give us and example of how the horrendous duplication is blocking some idea you have? particularly if the stovepipe is across three levels of government? And use the agency names so we can look at their websites.

      • 2009 September 23
        Kevin Cox permalink

        I can give an example but I am not going to do so publicly because that has a risk of delaying the release further.

        Here is the reason. If you ask a bureaucrat “can I do something” they are almost obliged to say no – no matter what it is and once they have said no then it tends to stay no until they are forced to change as finally saying yes is an admission of giving the wrong answer. It is no fault of any public servant but if you ask permission to do something then there must be some doubt right? Public Servants have to be conservative and have to avoid risk. I know I have picked on public servants but it is the same with all bureaucracies. (Have you asked a University to open up some data that it has collected that the public has a right to know?) That is the nature of bureaucracies and while it might frustrate some of the time I think it is a good rule because people do “test” the system.

        We could however advertise the positives and point out the areas where there is access. That is the inventory of sources will help public servants in their decisions. Once a government department in say SA releases information then this is a precedent and the question that can then be asked of the public service is – “I notice that SA is releasing toilet information – can NSW do the same?” This now requires NSW to give a reason and if they say no it also means they are criticising SA and perhaps showing their inadequacies – so the answer may well be “can you help us do it”.

        The Task Force however can ask without being given an automatic no and delaying the process because that is its bureaucratic function. The answer to the Task Force will be – “we would like to but we do not have the resources”

        A public league table of “bad departments”is unlikely to help while a public league table of “good departments” has a greater chance of success.

      • 2009 September 23


        By way of example, I am talking to one state govt department, no names here :-) , about providing my DemographicDrapes (thematic map overlays that can be overlayed in GIS and mashups) webservice. They tell me that many divisions within this dept typically get (under their own separate stovepipe budgets) the ABS Census data, load it into their individual environments for access by their own internal users – so straight away the first level of duplication although this probably only gets done every 5 years – each Census cycle.

        However, these individual groups then go about building the many different filters that allow them to deploy thematic maps – the graphic representation of demographics. They all do this independently so its quite possible they all build a filter for say basic population densities and many others – another level of duplication. In addition, one group may build a very useful filter that others could also need and use but they never get shared around for a lot of stovepipe culture reasons – not replication but all the same not smart. Then add to that any data series update which can happen during a release cycle and some upgrade whilst other’s don’t – who’s right and how much duplication there? And this example is probably only the tip of the iceberg as its covers just one Govt dataset in one Govt dept in one State!

        So as I said, I see this as an niche opportunity to take the ABS Census data and provide a published resource that is accessed via the industry standard API (OGC) for spatial data – and that is not GoogleMaps API by the way although a GoogleMaps app can ingest OGC services. The Census data is one very comprehensive and excellent dataset that is in my mind under utilised and this is largely based on the fact that it really is targeted for the elite demographer market but could be commoditised for a broader market place.

        Now why doesn’t ABS publish the raw spatial data and its attributes? I think they may one day and maybe I can shift my value added service over to their infrastructure but in the interim I want to have a go at it as I’m not yet convinced Govt can service open Internet services as the private sector can. Maybe I’m wrong?

    • 2009 September 23

      What do you mean not publish it’s spacial data, the spacial data is published but is Copyright to the Government. This in no way stops you from negotiating with them on using it.



      • 2009 September 23


        What I mean by ‘publishing’ spatial (that’s with a “t” by the way) data is when a Govt Agency exposes its mapping data on a web server somewhere in an open way which is accessible to the world or to subscribed users. Once ‘published’ this way it can be used by the broader public and other Govt agencies in a shared manor without having to replicate the data throughout the community.

        Some govt data is being published this way. Notably the Bureau of Meteorology although behind account security and Landgate in WA. This is all about the establishment of a National Spatial Data Infrastructure where Govt Agencies (and private sector firms) ‘publish’ this way and the public connects and uses the data in infinite applications. Right now, that nirvana is a pipe dream even though the technology to do it is well and truly here.

        Of course I can acquire a copy (free or licensed) of spatial data from Agencies but that is my copy and its potentially replicated thousands of times. I’m not saying that isn’t useful just that as we increasingly move to a network culture – BTW I think that will get here before the paperless toilet does ;-) – we will move towards the online publishing of spatial data. Its just my opinion though!

        • 2009 September 24

          Fair enough, re spelling mistake always a sign of tiredness with me, pulling really long days for weeks on end isn’t exactly healthy for the brain or the body :-/

  7. 2009 September 23
    simonfj permalink

    Thanks Kevin,

    I appreciate what you are saying about carrots being better than sticks. It’s why ‘good’ case studies are the only education we normally get. (the ‘bad’ get buried of course). But that wasn’t my intent. I’m a great believer that PSers are just people doing a job; some progressive, some conservative. The places where they live have more to say about the culture of the bureaucracies they find themselves in. E.g. Canberra’s is so very different than Brisbane’s is so very different than ….. If I wanted to politic I would try the divide and progress approach, as you’ve suggested. I’ve never been a John Howard fan.

    You can imagine that govdex has been set up to see about creating communities of people in remote bureacracies who are doing similar things, so they can take a coordinated to doing something; in our case, releasing some data. But only people in Canberra would actually put up in red that the scope of their initiative precludes the public and public facing communities.This is a culture; one of exclusion. It needs to be recognized.

    Now I’m not having a go at anyone here. I’m just pointing out something which we all might consider as containing a hint of truth. When I talk to bureaucrats in the Amsterdam or London its quite different. They’re not as defensive, or would make it quite this obvious. My suggestion above aimed at eliciting some description of what might be done, if only ……..

    It’s been my experience that if you talk early about possibilities, are clear about what you are trying to achieve and gain a community of (more) knowedgeable peers to talk it through, it becomes like this place. We might not always agree with one another about, but we sure understand that we’re on the same side. The only thing missing is a way to prioritize things, which the taskforce in its wisdom has attempted with the ideascale. So we know that the culture wants to change, which it might when these kind of discussions happen in the govdex domain. Until then, are you one of ‘us’ or one of ‘them’?

    • 2009 September 25
      Kevin Cox permalink


      To give you a specific example of a non government organisation blocking the use of data and later a government organisation that I will not name.

      If you look up someone in the online white pages and you use that information for another service and sell the service then Sensis believe they can sue you for breach of copyright. How do I know – we have a letter from Sensis threatening us with legal action. How did Sensis know? We asked them if we could use the information for an unrelated service (that is it was not threat to the white pages business and the information was not published). Why did we ask because they say on the white pages website that you cannot use the information for any commercial purpose. The reason is that if you want to use the information for a commercial purpose then Sensis want to get paid. We now pay Sensis for the use of the data.

      I think we would not have to pay if we took Sensis to court but who is going to take Sensis on let alone the government. It is unlikely to stand up because of the recent IceTV case against Channel 9 where Channel 9 tried to stop the reuse of TV program information. They were successful in a lower court but it was dismissed in the high court

      To give you another case – which as far as I know has not been challenged. The Afl believes it “owns” the football draw. If you want to use the draw without paying the AFL money they will threaten to sue you. How do I know because a footy tipping competition that was a “for fun” site and which I used (no prizes or gambling) had to close down because the AFL wanted payment.

      On many websites you will find things like the following.

      2.2 What you cannot do:

      copy any of the material on this web site or otherwise incorporate into or store in any other web site, electronic retrieval system, publication or other work in any form; or
      provide a link to this web site from another web site without XXXX’s prior written consent.

      Even on they say

      Anyone may link to this website provided that they do not use solely for commercial practices

      That is, organisations everywhere (including government websites) try to limit what you do with information they publish and are particularly concerned if you make a profit from doing so.

      An organisation we know asked a government department could they use the data on their website for a commercial purpose (because the website said they should ask). The government department said no verbally but not in writing.

      We told the same department what we were doing with their data – as a matter of courtesy – and because we thought our use could help them and be of interest to them. They said, we do not want to know and we reserve the right to ask you to stop if we feel like it in the future. We of course shut up and we were pleased we did not ask permission.

      Hopefully you think I am still one of ‘us’ :) but this is the reality of trying to use government (and commercial) data that is already published.

      It is also why this task force efforts to give government departments some direction on licensing and access is so important. Departments do not know what they should do and so of course they will say no and block the use of data if asked.

      • 2009 September 25


        I agree that its extremely important to address this licensing issue. The notion of an Government infrastructure being established that encumbers commercial activity is just crazy in a capitalist society. That is akin to saying Govt will build a road network from your hard earned taxes but Logistics companies, or any company for that matter, cannot use it if they are going to run a profitable business on it – although it may fix up Sydney’s chronic traffic problems ;-) . If there is to be any license fee it must be nominal such that it does not become a showstopper to innovation.

      • 2009 September 28
        Brad Peterson permalink

        Anyone may link to this website provided that they do not use solely for commercial practices

        I can go one better. Every WA local council web site contains the following in its Terms and Conditions: “You may only link to the site with the express written permission of the Sire of . Any links, if allowed, must only link directly to the site’s homepage (and no other pages within the site) unless otherwise agreed by the Shire of in writing.”

        Where on earth did they get that idea? You do not need permission to create a text link to any other site/page including, for example, the 2009/2010 Adopted Budget for the Shire of Wagin (PDF).

  8. 2009 September 23
    Gary Nairn permalink

    While there is still lots of data that the Federal Government could release the real culprits are the States. They continue to charge substantial fees for data that the taxpayer has already paid for. The States with limited revenue raising powers and therefore remain reluctant to lose any form of income even though in many cases the cost of administering “cost recovery” is more than the income. The Federal Government made a substantial step forward in 2001 when they went away from a cost recovery model for most of their data (but obviously not all).

Comments are closed.