Thursday, February 18, 2010

What does the White Pages ruling mean for Australian government data?

There is a trend towards greater openness in the licensing of Australian government data. Queensland's government a few years ago put in place a framework (GILF) for Creative Commons licensing and Victoria's government recently committing to using Creative Commons as its default copyright licensing system.

Some steps have taken place at a federal level, with both the ABS and Geosciences (see their footer) moving in the same direction.

However the recent court case where Telstra sued the publishers of Local Directories over the republishing of Yellow and White pages information - and lost - marks a further step in the process.

In the case, Telstra Corporation Limited v Phone Directories Company Pty Ltd [2010] FCA 44 (8 February 2010), the judge found that Yellow and White Pages listings were not covered under copyright law as they were not original and that (requoting from the article Telstra loses copyright case over Yellow Pages and White Pages,
"None of the people said to be authors of the Works exercised 'independent intellectual effort' or 'sufficient effort of a literary nature' in creating the (directories)."

"Further, if necessary, the creation of the Works did not involve some 'creative spark' or the exercise of the requisite 'skill and judgment'."
This case follows a related decision in the IceTV case in 2007, where Channel Nine claimed that its TV Guide was a literary work and IceTV could not create a copy of it through independent effort.

So what does this mean for similar forms of government information released under Crown Copyright such as transport timetables, budget accounts, lists of elected officials, statements of interests and other lists and statistics which did not require 'creative spark', 'independent intellectual effort' or 'sufficient effort of a literary nature'?

I am not a lawyer and don't trawl all the legal cases reported online on a regular basis, however to my knowledge no Australian state or federal government department has recently gone to court against individuals or corporations replicating and reusing statistical data of these types. So there is no actual ruling I am aware of to test whether this government data remains legally protectable under Crown Copyright.

In at lease one case, involving NSW RailCorp in early 2009, cease and desist letters were sent by RailCorp's lawyers (to three iPhone application developers). This didn't end up in court as the resulting publicity brought the situation to the attention of the then NSW Premier who ordered RailCorp to negotiate arrangements to share timetable data with less stringent copyright provisions.

I believe that a reasonable supposition at this time is that where publicly released government data does not meet the required tests in the copyright case, it would be difficult to prove why it should be protected under Crown Copyright.

This would make copyright over lists of names and figures very hard to justify.

I do appreciate that government departments have concerns over information being used in inaccurate or misleading ways, or that people may rely on out-of-date information through third party sources (a particular concern for transport networks). However Crown Copyright may not be the most appropriate tool to mitigate these risks anymore.

Maybe we need to look at other approaches, such as making it easier for third parties to use data in the way intended - such as providing data feeds at consistent URLs for reuse (which means third-party applications will be as accurate as the government figures), ensuring that data labels are human readable and clear (to reduce misinterpretations) and including date stamps in data so it is clear when it is current from and to.

In cases where data is used inappropriately, government still has recourse through Creative Commons type licensing and other aspects of Australia's legal system to restrain this usage while supporting appropriate use.

Further comments and legal views by lawyers and interested parties are heartily welcome!


  1. Agree this case has wider implications. Importantly, Sensis pursued this case with the aim of helping ensure differentiation in the market thus encouraging fair play and reduced confusion for advertisers and consumers.

    We believe allowing others to mimic the content, format and branding of our products creates confusion among advertisers and consumers. Further, we think it inappropriate for others to springboard off the hard work, thought and effort our people put into the White Pages® and Yellow Pages® directories.

    Other directories businesses can apply to access the Integrated Public Number Database (IPND), which was created under the Telecommunications Act (1997) to foster competition in the directories business and includes raw data of all residential, business and government listings in Australia. To meet privacy concerns, the IPND database (like the White Pages® and Yellow® Pages directories) cannot be used for marketing purposes, only for the creation of directories.

    Sensis will be appealing the Federal Court’s decision.

  2. Indeed the Telsta case provides a great precident that clarifies the law even more. I used this case as justification for publishing the raw data.

  3. Thanks for the comment Karina.

    Separating the IP in the design and format of the Yellow and White Pages from the raw data itself (names, addresses and numbers) will be an interesting challenge under Australian copyright laws.

    It would generate a very interesting precedent if successful.

  4. Glad to hear Sensis pursued this case because they wanted to "help".

    So what is the special contribution to this raw data?

    Did Sensis invent alphabetical order?

    Or did they invent a special structure, such as for addresses: street number, then street name and type, followed by suburb? I this is rather derivative.

    Does Sensis own my business name and address just because I list my business with them? I've given them use of a fact ie paid them to put it in their list and give others access to it.

    Now Sensis claim some form of ownership over the fact of my address or business name as well? Did I sign something handing over these rights of IP to Sensis? If you want to get technical, it may not be mine to give either. The local government 'created' the address when they assigned it to my property. So should Sensis pay royalties to the Councils everytime they use it?

    The address (if it's a 'quality' address) was hopefully assigned in compliance with AS/NZS 4819 Street Addressing Standard. That Standard was developed through the hard work of a group including representatives of state and national survey & mapping agencies and Australia Post, among others (no intellectual contribution from Sensis).

    So, technically, Sensis is leveraging off the hard work, thought and effort of governmnet emplyees who assignment addresses, and manage the roads and road names etc? After all, these are all value-adding processes, without which the address is a rather useless piece of information.

    Taken to logical conclusions many IP arguments turn silly very fast.

    The wider implication that Sensis need to take a fresh look at their business model.

  5. PS sorry for all the typos in previous comment, I was having some refresh problems with the page.

  6. IANAL.

    I'm happy with the decision, but there's some important things that should be read in the details of the ruling.

    The judge effectively said there was no intellectual / creative effort in it; but he clarified it:
    * Everything was done within a piece of software
    * The software set the rules
    * The software did a large amount of work automatically
    * Importantly, when users were needed to help improve, users could only choose options controlled by that bit of software (as defined by the rules) - basically, choose "A" or "B".

    This means that things like the factual data (statistics, ratings, raw numbers) would be fine, but textual descriptions of the schools are not fine to infringe on - a human has put in some creative effort there to describe it.

    Since the statistics and descriptions are intertwined on the HTML rendering of, you might be in a bit of deep water if you utilised anything from there.

    However, it does give you a nice bit of leverage for a whole new category of sites.

    They flat out refused to grant access to the raw data, but now I'm more than within the bounds of the ruling to scrape the content.

  7. Karina

    I understand your concerns about Sensis being misrepresented or Sensis brand being compromised in any way and that is a valid objection. However, that is a different situation from the use of the data for other purposes that have nothing to do with Sensis. To give you an example. Let us assume I have a social networking website that has a group of people who wish to communicate by telephone with each other. Rather than people having to put up their telephone number they could be looked up automatically and phoned directly. This brings Telstra business and is easier for the social networkers. As you do not charge people to look up their name in your directory it would seem difficult for you to justify stopping screen scraping as the only loss would be the extra cost from these extra hits on the search pages.

    The IPND data base is interesting. Perhaps you should be arguing that it be made available directly on line for people to look up others. That could take away any extra load on your websites?