View geo2gov

Contest Winner
Highly Commendable Mashup and Transformation Challenge
Created by
Adam Kennedy and Jeffery Candiloro datasets used
Australian Federal Election Boundaries
Other datasets used
Australian Bureau of Statistics Product 1259.0.30.001, Australian Standard Geographical Classification, Australian Bureau of Statistics Product 2093.0.30.001, Census of Population and Housing: Census, Geographic Areas Digital Boundaries, Australian Electoral Commission Proposed NSW Redistribution Geographic Boundaries, Australian Electoral Commission Proposed QLD Redistribution Geographical Boundaries, Australian Electoral Commission 2007 Election Geographical Boundaries, Australian Electoral Commission 2007 Polling Place Database, New South Wales Election Commission State Division Geographical Boundaries, Victoria Election Commission State Division Geographical Boundaries, Victoria Election Commission State Region Geographical Boundaries, Electoral Maps (KML) - Various, Google Maps Geocoder API, GeoPlugin IP2Geo API, New South Wales Department of Local Government Listing of Mayors, New South Wales Election Commission Local Government Election Ward Boundary Maps

The geo2gov API is an online service that will take a location description in a wide range of formats, and map that location to the government.

It can determine jurisdictions, electorates, representatives, where your location is defined demographically at each layer within the Census, and provides identifier maps so you can link into other websites such as the Federal Parlimentary website, OpenAustralia, MyMP and more.

geo2gov can be easily incorporated directly into other mashups as a search or address box, and can be deployed as a virtual applicance locally in addition to the regular online Amazon EC2 cloud-hosted service.

To make the Census results from even more useful, it also provides pre-compiled single file database packs of the Census for PostgreSQL and SQLite.

The HTML version of the website is provided for debugging and development purposes, and demonstrates a fairly typical usage of the API.

Judges' Comments: geo2gov serves as an excellent example of what can be possible with open government data. This entry provided an online service that will take a location description in a wide range of formats, and map that location to the government. The testament to its utility was demonstrated by the fact that several other entries used geo2gov. Contest judge Mark Pesce said that this app that was such an impressive prototype of what was possible with government data that it made his geeky pants wet.

17 Responses to “geo2gov”

  1. Adam Kennedy says:


    This is an announcement from the geo2gov team. We have been observing with great dissapointment the rampant vote-stacking and vote-poisoning occuring with the MashupAustralia competition website

    At this point, the voting system has been abused so heavily in pursuit of a $2,000 prize that we believe it no longer holds any relevance.

    To ensure that these activities do not result in any offender taking the prize, we intend to manipulate the voting system so obviously that it will be clear to all and sundry that the voting is of no use.

    We hereby declare ourselves ineligable for the People’s Choice prize. If awarded this prize, we will refuse it.

    Our intent it to place ourself at the top of the voting ranks, thereby denying first place to any and all who thing they can vote-manipulate their way into victory.

    This is a community of IT professionals and democracy nuts. You should know better than to do this kind of thing. Shame.

  2. Carlos says:

    Mate, your arrogance is not inspiring. Your application is not inspiring either. Shame.

    • Adam Kennedy says:

      You are correct.

      The goal with the voting stuff is not to inspire, simply to ensure the problem is not overlooked.

      I’ve tried to compensate by providing reviews and comments on as many mashups as possible. I hope you will also.

  3. john says:

    i’m sure he’s getting more visits to his mashup by being a martyr than by just standing by doing nothing (which I guess a lot of us are doing). hell i checked the web logs this morning, and my mashup had a total of 8 visits from mashupaustralia, half of which didn’t even check out the mashup. so yeah, you have to scam to compete (or just to be seen) unfortunately, so standing around doing nothing is doing nothing. i agree adam could have been a little more humble about it, but i guess these things need to be done.

  4. Eddie says:


  5. Raul says:

    I also noticed the issue and ask the organisers using the contact form. This is the reply I got:

    “The selection process will take other factors and criteria into account and the current voting issue will also be taken into account prior to Judges making selections”

    Not sure what it means exactly in practical terms but I thought I’d let you all know


  6. Hopefully they take into account the fact that other entries have suffered by batch lower votes.

  7. Ben says:

    We experienced lots of negative votes Right after being posted. Seemed weird as people didn’t even go to our site to see the mashup.

  8. Confused says:

    How do I find the Australian Federal Election Boundaries data from the dataset used in your mashup?

    – I typed in Karratha WA into your mashup but couldn’t find Durack.
    – Similarly typed in Esperance WA, couldnt find O’Connor.

    I thought the Federal Election Boundaries were redistributed to Durack and O’Connor around the end of 2008.

    • Jeffery Candiloro says:

      G’Day Confused,

      The data is based on the electoral boundaries at the 2007 election.

      Right now they are the ‘active’ divisions. Redistributions don’t become ‘active’ until the next general election.

      So we return data based on the 2007 election. There is a plan to return a “you are in X division now, you will be in Y division at the next general election” but we’re not there yet. :)

      • Confused says:

        Thanks for the update.

        From downloading the Australian Federal Election Boundaries dataset from, the data appears to refer to both Durack and O’Connor.

        Your entry states the use of just 1 dataset datasets used
        The Australian Federal Election Boundaries

        I understand this means your entry complies with competition eligibility rule #3.

        Have you actually used this dataset in your mashup?

        • Adam Kennedy says:

          The revision of the dataset provided by the AEC is essentially corrupt.

          It contains neither the set of electorates currently in force as per the 2007 election, and it doesn’t contain the set of electorates that will be used in 2010.

          It’s a broken release and any mashup that chose to use the specific revision linked to on would be broken essentially by design.

          This creates some uncertainty, because the Temporal Coverage of the dataset is undefined. It doesn’t say you have to use that specific release, but it doesn’t say you can’t use another release of the same dataset.

          We are using the more correct official 2007 version of the same dataset.

          That is, we are using COM_ELB_04032008.MID/MIF instead of COM_ELB_12112008.MID/MIF. It is the same dataset at a different point in time.

          The data is largely correlative between the two on the Easy coast (where distributions have only just been completed). So we are using about 2/3rd identical data to the current dataset.

          If the judgement is that using a different file name, containing the same data and of the same dataset, somehow doesn’t count as “using the dataset”, then I suppose we could do a centroid correlation between the two and make sure the actual (identical) bytes came from that file.

          We already import three releases of that file into our database (The 04032008, 12112008 and a new post-redistribution one), we’d just need to change a dozen lines of code in the schema compiler.

          But that would seem like pointless makework to me.

  9. Steve L says:

    I don’t see Adam’s comments as being arrogant – he’s highlighting a badly designed feature of the contest, and is being open about doing so.

  10. The Absurd says:

    This is actually pretty useful, which might not seem immediately obvious, but it’s important data, and it’s great that it’s downloadable (although the postgres census file wouldn’t import, but that might be user error). Looking forward to seeing where this might end up going.

  11. Anonymous says:

    Wow, someone saying that is manipulating the system should be disqualified immediately from the contest!

    Wrongdoing is still wrong, even for noble ends. You should have followed the correct channels such as contacting the organisers and making strong statements with the Gov 2.0 taskforce if necessary. This is Australia we live in a land of laws and rules. I don’t want to start a war here, your mashup is probably the most useful I’ve seen here, but there’s little civic value in breaking rules.

    I hope you reconsider your statements.

    • Jeffery Candiloro says:

      I don’t want to start a debate here on the merits or otherwise of our actions. However, if you want to discuss the nature of Australia and her democracy then you cannot ignore the fact that this is one of the few places in the world where it is safe to engage in civil disobedience, where anyone can question any and every law and where one is allowed to speak out. And here we are, disagreeing, in a public forum, on a site that is working to open up information.

      In fact, one of the core aims of our entries is to make it even EASIER for people to engage in their common representative democracy by knowing where they live and who represents them.

      As for disqualification, we have already withdrawn ourselves from that part of the competition that is impacted by the public voting.

      Thank you for your support of our application.

  12. Huh says:

    You state that “… the dataset provided by the AEC is essentially corrupt.”

    Is this statement based on your claim that, “…the Temporal Coverage of the dataset is undefined.”?

    If so, isn’t the Temporal Coverage for all datasets specified on the download page for each dataset?

    For example, the download page for Australian Federal Electoral Boundaries clearly states both the Date Published and Date Updated.