Electronic Frontiers Australia

After the 'Facebook Files', the social media giant must be more transparent

Most people on Facebook have probably seen something they wish they hadn’t, whether it be violent pictures or racist comments.

How the social media giant decides what is and isn’t acceptable is often a mystery. Internal content guidelines, recently published in The Guardian, offer new insight into the mechanics of Facebook content moderation.

The slides show the rules can be arbitrary, but that shouldn’t be surprising. Social media platforms like Facebook and Twitter have been around for less than two decades, and there is little regulatory guidance from government regarding how they should police what people post.

This article is by Nicolas Suzor from Queensland University of Technology and was originally published on The Conversation. See the original article.

In fact, the company faces a significant challenge in trying to keep up with the volume of posted content and often conflicting demands from users, advertisers and civil society organisations.

It’s certainly cathartic to blame Facebook for its decisions, but the true challenge is to work out how we want our online social spaces to be governed.

Before we can have that conversation, we need to know much more about how platforms like Facebook make decisions in practice.

The secret work of policing the internet

Apparently weighing in at thousands of slides, the newly published guidelines give some more detail to the vague community standards Facebook shares with its users.

Most of the documents are training material for Facebook’s army of content moderators who are responsible for deciding what content should go.

Some of the distinctions seem odd, and some are downright offensive. According to the documents, direct threats of violence against Donald Trump will be removed (“someone shoot Trump”), but misogynistic instructions for harming women may not be (“to snap a bitch’s neck, make sure to apply all your pressure to the middle of her throat”).

The Guardian’s Facebook Files explainer video.

The rules appear to reflect the scars of legal and public relations battles Facebook and other social media platforms have fought over the last decade.

The blanket rule against images of nude children had to be changed after Facebook controversially banned the famous image of Kim Phuc fleeing napalm bombing during the Vietnam War. After years of controversy, a specific procedure now exists so people can request the removal of intimate images posted without their consent.

Because these rules develop over time, their complexity is not surprising. But this points to a bigger problem: without good data about how Facebook makes such decisions, we can’t have informed conversations about what type of content we’re comfortable with as a society.

The need for transparency

The core problem is that social media platforms like Facebook make most decisions about what constitutes acceptable speech behind closed doors. This makes it hard to have a genuine public debate about what people believe should be allowable to post online.

As the United Nations’ cultural organisation UNESCO has pointed out, there are real threats to freedom of expression when companies like Facebook have to play this role.

When governments make decisions about what content is allowed in the public domain, there are often court processes and avenues of appeal. When a social media platform makes such decisions, users are often left in the dark about why their content has been removed (or why their complaint has been ignored).

Challenging these decisions is often extremely difficult. Facebook allows users to appeal if their profile or page is removed, but it’s hard to appeal the moderation of a particular post.

OnlineCensorship.org provides guidance to users about how to appeal content moderation decisions.

To tackle the issue of offensive and violent content on the platform, Facebook says it will add 3,000 people to its community operations team, on top of its current 4,500.

“Keeping people on Facebook safe is the most important thing we do,” Monika Bickert, head of global policy management at Facebook, said in a statement. “We work hard to make Facebook as safe as possible while enabling free speech. This requires a lot of thought into detailed and often difficult questions, and getting it right is something we take very seriously”.

But without good data, there is no way to understand how well Facebook’s system is working overall – it is impossible to test its error rates or potential biases.

Civil society groups and projects including Ranking Digital Rights, Article 19 and the Electronic Frontier Foundation’s OnlineCensorship.org have been advocating for more transparency in these systems.

Facebook and other social media companies must start listening, and give the public real insight and input into how decisions are made.


Related Items:

Copyright for Australia that makes sense. That’s fair.

Volunteer Wikipedians in Australia are highlighting the need for the introduction of fair use in Australia. You can visit FairCopyrightOz to learn about how Wikipedia does, and Australia could, benefit from it.

Imagine a land in which everything was outlawed,
except for the things that were specifically allowed.
Our laws are based on principles rather than prescriptions.
Except for copyright. – Peter Martin

Have you ever done one or more of these?

  • Shared photos you didn’t take on social media?
  • Re-posted or create memes?
  • Backed-up your DVDs?
  • Forwarded an email?
  • Photographed graffiti or a mural?
  • Quoted from an article or book on your blog?

All these actions copy other people’s copyright material. In Australia, none of these common practices are allowed under copyright law without permission.

With this in mind, volunteer Wikipedians in Australia are highlighting the need for the introduction of fair use in Australia through a banner on the English Wikipedia. In doing so, they add their voices to six government reports since 1998 which have recommended introducing fair use to bring balance to copyright rules. You can visit FairCopyrightOz to learn about how Wikipedia does, and Australia could, benefit from fair use.

In Australia all copying requires permission unless you are only using an insubstantial part of a copyrighted work (which is typically very hard to judge), or the Copyright Act provides a specific exception. The most important exceptions, the fair dealing exceptions, cover research, study, criticism, review, parody, satire, reporting the news, and professional advice as long as the use is “fair”. Any use not for one of these purposes will be illegal, no matter how fair or reasonable it is, unless the government introduces a specific exception for it. This means Australian copyright law cannot keep pace with change, as every new technology or activity requires its own new exception. This takes time and a lot of advocacy. Using a VCR at home to tape television programs was illegal until the legislation was amended in 2006, over 30 years after their invention.

Fair use would fix this. The United States’ fair use law judges each instance on whether it is fair, guided by four fairness factors:

  • purpose and character of the use;
  • nature of the copyright material;
  • amount and substantiality of the part used; and
  • effect upon the potential market for, or value of, the copyright material.

In concentrating on what is fair, it adds flexibility to the law, allowing it to keep up with changes in technology and society. Any Wikipedian who has ever uploaded a fair use file will be familiar with the “fairness” test and the thorough analysis it requires.

Without fair use, Australian copyright law will always lag behind common practice. The lack of flexibility to allow socially beneficial uses—like non-commercial private uses or incidental and technical uses—greatly limits people’s ability to interact with their own culture. The difficulty, or often impossibility, of getting permission means that groups like schools, libraries, archives and technology companies are limited in what they can do, even when their activities aren’t harming copyright owners. It also means Australian schools end up paying millions of dollars each year to use publicly accessible online content on websites that you use at home for free. No one is asking to be paid for using these websites, and the money rarely makes it to the copyright owner. Just as importantly, the use is transformative and socially beneficial. But because the Act doesn’t say such uses are allowed, payment still has to be made.

Wikipedia is one place where Australians regularly notice the benefits of fair use. Around 10% of Wikipedia pages in English have some form of fair use content—that’s over 500,000 articles quoting from a book or an article; showing a company or sports team logo that contains an artistic work; including an audio-sample or album cover; or referencing a book or film title image.

Imagine reading the Wikipedia article on Australian Markus Zusak’s classic The Book Thief, without seeing its front cover, or reading about the classic song Land Down Under by Men At Work without hearing a short clip from it. That’s what Australians would have if Wikipedia’s users could not upload files under the principles of fair use. On the English Wikipedia, copyright rules are based on US fair use guidelines that support the values of the free culture movement. Australian users in Australia should have the benefit of the same principles.

Starting this week, banners will appear to Australians accessing English language Wikipedia articles over the next few weeks. It is rare for Wikipedia editors to place banners across articles. It is even rarer to draw attention to a legislative issue. Wikipedia prides itself on its neutral point of view, after all. However, in a discussion among Australian editors on whether to take action in support of the recent Productivity Commission report, two things became abundantly clear.

  1. Australian Wikipedians strongly felt that it was important to our mission of public education—that the general public should know that we, as volunteers, are already benefitting every day from fair use in Wikipedia articles. Consequently, Wikipedia’s readers do too.
  2. There are misconceptions about what fair use means in practice which we are in a position to dispel. Some Australian Wikipedians commented that they thought Fair Use already is Australian law, which goes to show just how far common practice differs from the law.

Allowing fair use images in Wikipedia is a matter of editorial policy determined by each language community. Wikipedia editors take great care to ensure that all content is free for other people to use in as many circumstances as possible. We want other people to improve and share Wikipedia’s educational resources far and wide. The inclusion of fair use material in Wikipedia reduces the ability for it to be re-used by those who live in countries without this exception. However, Wikipedians for the English-language Wikipedia have determined that the benefits of having these materials available outweigh the concerns that fair use might not be open enough.

The Australian Digital Alliance (ADA) and Electronic Frontiers Australia (EFA) have long championed fair use for Australia and were happy to help support Wikipedians in raising awareness of this issue. Parallel to the Wikipedia banner campaign, on Monday they launched faircopyright.org.au where citizens can learn more, and take action by writing to their member of Parliament—encouraging the government to accept the fair use recommendation made by the recent Productivity Commission Report and other enquiries. As Peter Martin has written in his article about this fair use campaign, this is a first for the Australian Wikipedia community. Wikipedians have also written a new article on the History of fair use proposals in Australia to help increase the level of verifiable and neutral information available to the public on this matter of public policy.

To learn more, visit FairCopyrightOz on Meta-Wiki, or visit the campaign site set up by the ADA/EFA: faircopyright.org.au.

This article is by Liam Wyatt, Wikimedia community member and Stephen LaPorte, Senior Legal Counsel, Wikimedia Foundation. It was originally published on Wikimedia's blog. See the original article.


Related Items:

Data retention has always been about a lot more than 'national security'

Attorney-General George Brandis told us in ­November 2014 the data retention regime ­“applies only to the most ­serious crime, to terrorism, to international and transnational organised crime, to paedophilia, where the use of metadata has been particularly useful as an investigative tool, … only to crime and only to the highest levels of crime”.

The mandatory data retention legislation was duly rushed through the parliament in March 2015 at a time of “heightened national security concern”.

Remember all those flags?

Image: Alex Ellinghausen

But the claim that it was all about national security and child predators, was, of course, far from the complete truth. Telecommunications data — more commonly, metadata — is also extremely useful in identifying the source of government leaks and for tracking down whistleblowers.

It’s been regularly accessed for this purpose, including by ASIO, and it was just such an attempt to identify the source of a leak to the press that got the Australian Federal Police into such dramatic trouble last week.

So, let’s be clear, data retention is, and has always been, about a whole lot more than identifying jihadists and child predators.

It was only last-minute pressure from the journalists’ union — the MEAA — and a number of news organisations that resulted in the addition of a requirement for any access to the retained data of a journalist to require a warrant.

At the time much was said about the critical role of journalists in the democratic purpose, including from then-communications minister (and one-time journo himself) Malcolm Turnbull, who explained, “the work journalists do is just as important in our democracy as the work that we do as legislators”.

Yet, curiously, politicians were not given the same level of protection (it’s pretty clear most of them are yet to realise this). Nor were lawyers, doctors or counsellors. Or, for that matter, the rest of us mere citizens and residents of Australia.

As Friday’s mea culpa from the AFP demonstrates, the practical ­reality is that it’s impossible to provide special protection to any subset of the population in the context of an indiscriminate data retention scheme.

In defending his decision not to take any action against the officers who illegally accessed the phone records of an as-yet-unidentified journalist, AFP Commissioner Andrew Colvin said they “did not realise they were required to obtain a warrant to access the journalist’s metadata … There was no ill will or malice or bad intent by the officers involved who breached the Act. But simply it was a mistake”.

This suggestion the “investigating officers” were unaware of the requirement for a warrant is ludicrous. First, this requirement has existed for 18 months (since October 2015). Second, these are “Professional Standards” ­officers (or what is known as Internal Affairs in TV land). They’re the cops who investigate the cops. Presumably, they’re supposed to know the rules better than anyone. And we know the AFP conducted training because Freedom of Information requests by The Guardian have produced the slides.

All the officers had to ­remember were the words “journalist” and “warrant”.

But even if it were inadvertent it is a clear demonstration of just how ineffective it is having an extra level of protection for one subset of society.

And it is a very long-established principle in criminal law that ignorance of the law is no defence.

It’s also worth noting there are a number of criminal sanctions defined in the data retention legislation that carry a two-year jail term, including the disclosure of “the existence or non-existence of such a warrant”.

I see no reason why the same penalty shouldn’t also apply to unauthorised access to this data.

The telcos/ISPs that receive these data access requests have no way of knowing which of their customers are journalists. So they’ll dutifully process them as they would the other 99 per cent of requests that don’t require a warrant.

The only way to achieve meaningful protection from malicious or inadvertent access to this data is to introduce a warrant requirement for ALL requests for access to data.

But critics will argue that “would make law enforcement grind to a halt”. Would it? Really? Most EU countries already require some form of independent authorisation for access to telecommunications data.

It’s clear that law enforcement in this country has grown used to using telecommunications data in almost all investigations for the simple reason that it’s been available to them without any meaningful controls, ­regardless of whether it’s useful or not. It’s time that we introduced some meaningful controls.

The only effective protection for anyone is protection for everyone.

The government must therefore legislate to extend the warrant ­requirement for access to retained telecommunications data to the ­entire population without delay.

This opinion piece by EFA' Executive Officer Jon Lawrence was originally published as "National security? Data laws misused to spy" in The Daily Telegraph on Thursday 4th May 2017.


If you agree that warrants should be required for all access to retained telecommunications data ('metadata') then please sign our petition.
Related Items:

Data retention: universal warrant requirement is only effective protection

Last week’s revelation by the Australian Federal Police that they illegally accessed the retained telecommunications data (‘metadata’) of a journalist without first obtaining a warrant demonstrates the complete lack of effective protection provided by the current legislation.

AFP Commissioner Andrew Colvin asserted that ‘the police officers investigating the leak did not realise they were required to obtain a warrant to access the journalist's metadata.’ The requirement to obtain a warrant to access the data of journalists came into effect on 13th October 2015. It is therefore inexcusable for the officers involved to be unaware of this requirement. In addition, of course, ignorance of the law is no defence.

As journalist Paul Farrell told ABC’s The Drum, this incident is “a systemic, structural failure of the AFP’s internal policies and the law”.

EFA has warned in the past that a warrant requirement limited to one group provides no effective protection whatsoever in the context of an indiscriminate, society-wide mandatory data retention scheme.

EFA Executive Officer Jon Lawrence said, “A whole range of relationships are no less deserving of independent protection than are journalist’s communications with their sources, including lawyers and their clients, doctors and their patients, and any other relationship where privacy is critical. The only effective means to achieve such protection is to have a universal warrant requirement for access to retained telecommunications data.

“Without a universal warrant requirement we will continue to see instances of unauthorised access to data, regardless of whether such access is inadvertent or malicious.

“A majority of European Union member states have some form of independent, judicial authorisation required for access to telecommunications data. Such arrangements are therefore clearly workable and Australians are no less deserving of the same protection.”

EFA therefore calls for the immediate introduction of a universal warrant requirement for all access to retained telecommunications data.

Support this campaign, by:


Related Items:

Government says NO to expanding data retention to civil cases

Just before Easter, the government announced that it will not be expanding access to telecommunications data ('metadata') to civil litigants.

This is an important victory.

Had the government allowed even a limited expansion of access, it would almost certainly have been just the first of a number of such expansions.

It's also heartening to see that, despite running over the Christmas-New Year period, the government's consultation received 262 submissions, including 217 from individuals.

All but "a small number" of these submissions were opposed to any expansion in access.

This is a significant number of submissions to a consultation that was clearly intended to slip under the radar. We hope that the guidance we published helped some of those people with their submissions.

You can read here the report of this review that was tabled in parliament.

But, the fight against warrantless mass surveillance is far from over.

There is nothing to stop this or any future government deciding to give civil litigants access to data that is only retained for the purposes of the data retention scheme in the future.

Any such expansion could open up whole new troves of data for all sorts of civil actions including copyright enforcement cases, divorce/property settlements and employment disputes.

We're lobbying federal MPs and Senators to bring forward the review of the data retention legislation - it's currently due to be commenced in April 2019. As part of that, we're also pushing for:

  • a universal warrant requirement for access to data (currently warrants are only required when a journalist's data is requested);
  • no expansion in the restricted list of 22 agencies that are currently able to request data; and,
  • a reduction in the retention period for data from the current two years to no more than six months.
Like our work? We need your support

We rely on donations and membership subscriptions to continue our work. Effective lobbying requires travel and Canberra gets pretty expensive to get to and stay in when parliament is sitting.

Your support will help us continue our work on data retention and other digital rights issues.

If you can, please contribute today with:

Or, get actively involved:


Related Items:

Get a VPN today!

From today, 13th April 2017, all Australian telecommunication providers are now required to collect a whole range of your telecommunications data ('metadata') and retain it for two full years, so that it can be requested by government agencies.

This data includes information about your phone usage (including texts and your location) and about your Internet connection. This information allows very detailed conclusions to be made about many aspects of your life and there are almost no protections against investigative "fishing expeditions" or systemic abuse of power.

With the exception of journalists' data, no warrants are required for access to this data, and there is little effective oversight. The data retention scheme therefore represents a genuine threat to the privacy of all Australians.

That’s why we’re supporting today as a national day of action – we’re calling on Australians to educate themselves about the scale of this surveillance and take appropriate precautions.

So, we're declaring today, Thursday 13 April as 'National Get A VPN Day'.
1. What is a VPN and why do I need one?

A Virtual Private Network (VPN) is an online service that creates an encrypted 'tunnel' from your computer to a remote Internet gateway, which will often be in a different country. The encryption means that your Internet Service Provider (ISP) will not know which sites you are visiting - they will only see that you are communicating with a single address, that of your VPN.

Image: RapidVPN

Let's say you're active with an environmental group that the government is interested in, and the government has obtained access to the list of addresses that have visited that group's website. If you're using a VPN, they will not be able to identify you as having visited that site as they'll only have the address of the external gateway of your VPN.

Simply put, using a VPN breaks the identifying links between your computer and the websites you visit, thereby protecting you from government surveillance.

Because they encrypt your traffic, VPNs also provide protection from eavesdropping. If your traffic is ever directly intercepted, the encryption means it will be unreadable. This is particularly important if you're using a public wi-fi service.

For more information, here are good overviews from LifeHacker and from Wired.

2: Which VPN should I choose?

Different VPN services vary significantly in terms of quality, and particularly in terms of how much privacy protection they include.

For a better understanding of how VPNs can (and sometimes can’t) be trusted to protect your anonymity, see this article from Brian Krebs.

Some things to think about include:

  • What data does the VPN record? Is the VPN retaining web logs? Does the VPN know your IP address and the times that you connect to their servers? Also, what kind of advertising data does the VPN service store and does it hand that data over to third parties?
  • How long does the VPN store data? Nearly all VPNs will store some data in order to troubleshoot network issues. However, the duration of that storage plays a key role in terms of the privacy protection afforded to users. After all, if the data has been deleted, then it cannot be accessed by a third party. Ideally, a VPN should be wiping user data within hours of it being recorded. If a VPN is storing data for anything more than a few days then beware.
  • Read the privacy policy carefully. If you don't find the answers to your questions in their privacy policy then ask them directly, or steer clear.
  • What country are they based in? For example, you may want to avoid services based in Australia, UK, US, New Zealand or Canada (the so-called 'Five Eyes' countries, which have comprehensive intelligence-sharing arrangements in place). You may also want to avoid services based in countries with authoritarian governments.
  • What payment methods do they support? Using BitCoin &or other digital currencies will provide you with an extra layer of anonymity

Here are some good reviews and guides that will help you find the right VPN provider for you:

Or, you can, if you're technically-minded, roll-your-own. Here’s a handy guide for creating your own VPN service from Crypto Australia.

3: Help spread the word - tell your friends to #GetaVPN

Once you've got yourself sorted, don't forget about your friends, family and work colleagues.

  • Send them a link to this page
  • Retweet our link on Twitter, using the #GetaVPN hashtag
  • Share our Facebook post
  • Write to your local newspaper - letters to the editor can be an effective way to highlight an issue. See the contact section on your chosen media outlet. Keep it short and to the point.
4: Tell your MP and Senators what you think of mandatory data retention

We've been lobbying MPs and Senators over the last few years about the dangers of mandatory data retention, but adding your voice will help us to achieve the review of this legislation that we're seeking.

See our guidance on lobbying parliamentarians for ideas on how to be most effective, and for links to find your local MP and Senators from your state.

You may want to mention the following points when you contact them:

  • All access to this data should require a warrant - not just for journalists' data. A majority of European Union countries require some form of independent, judicial authorisation for access to this sort of data, so there's no reason why Australians shouldn't enjoy the same protection.
  • It's important that additional agencies aren't added to the list that are allowed access to this data. The one good part of the data retention legislation is that it reduced the number of agencies able to access this data from literally hundreds to less than two dozen (Police and anti-corruption bodies mainly).
  • The two year retention period is unjustifiably long and must be reduced to at most six months.

You can see which MPs and Senators voted for and against mandatory data retention on the excellent They Vote For You site.

You can also see which MPs and Senators voted for and against a universal warrant requirement for access to this data.

Related Items:

Bytes & Rights 2017 - Perth

The Bytes and Rights 2017 conference is being held as part of the Festival of the Web, an 8-day series of international conferences in early April in Perth.

See the full program of Festival of the Web events.

Bytes and Rights is a conference focused on the many issues around how society responds to changing technology, especially technology related to the Internet.

We will discuss legal, policy, regulatory and social responses to issues such as human rights, intellectual property enforcement, security and harassment. We aim to bring together experts including academics, lawyers, commercial practitioners, technologists, and civil society groups in a open dialog that cuts across specific disciplines.


Monday 3rd & Tuesday 4th April 2017


A variety of passes are available for just Bytes & Rights, or the entire Festival of the Web.

Please visit our registration page for more details and to register.

Related Items:

What is Fair? Karen Chester from the Productivity Commission

The Blue Diamond Gallery. CC-BY-SA

Good morning. Thank you Professor Giblin and congratulations on the launch of yours and Professor Weatherall’s book – What If We Could Reimagine Copyright? An anthology of ten thoughtful essays that collectively make for a forward‑looking treatise on copyright. And my thanks to the Australian Digital Alliance for inviting me to speak today, and to the broader church of the Australian Digital Alliance membership. Many of whom contributed time and effort in making submissions to the Productivity Commission’s inquiry into Australia’s intellectual property arrangements. Thank you.

This post is a speech given by Karen Chester, Deputy Chair of the Productivity Commission, to the Australian Digital Alliance Forum at the National Library of Australia on 24th February 2017. It is licensed under a Creative Commons Attribution 3.0 Australia (CC-BY 3.0 AU) licence.

With my words today I hope to do three things. First, share the lens through which the Commission reviewed and analysed Australia’s intellectual property settings, especially in matters of copyright. Second, do some much needed myth busting — to address claims made about copyright that on any objective examination are more fiction than fact. And third, and most importantly, convey what matters most in getting the policy settings right here.

At the get go of this Inquiry, we envisaged our task would be about how policy could grapple with the cocktail mix of technology, adaptability, creative endeavour, innovation and competition. And it did so to a large part. But at the end of the day — all roads led us to one simple truth; to ask and answer what is fair.

And when we use the term fair we’re not limiting this to fair use. Albeit copyright exception is the policy that matters most for getting the innovation and equity equation right. Because it’s not just about the creators vs the tech giants. And it’s not a zero sum game between rights holders and content users as some would have us believe.

It is about school kids, uni students, less tech savvy older people, less tech savvy younger people, documentary film makers, 55 year old redundant workers, universities and TAFEs trying to teach in a more accessible way, and the cost for anyone down under consuming the creative or innovative endeavour of others. For at the end of the day, out of kilter IP settings have and will continue to create a largely silent and growing class of ‘have-nots’.

So today I hope to connect the dots to the many everyday Australians that stand to benefit from the policy changes we have recommended to Government. For there is a compelling policy narrative to be had here — one of innovation and agility. But perhaps more importantly it is also one of equity that we can relate to everyday Australians. For when we relate the benefits of change to many Australians we know what is fair.

To understand our recommendations, it’s important to start at the beginning and think about IP more broadly. And in doing so our simple truth of what is fair should not be unsurprising. For today IP is embedded in all aspects of modern daily life. It is akin to love in the immortal words of The Troggs’ 1967 classic — Love is all around us. Because IP affects everything and everyone. And it is for this very reason that IP is a policy exemplar — it puts the public into public policy. And perhaps this is why changes to IP are so contentious — because their change affects everyone.

But perhaps it’s also reflective of the plethora of reviews and studies into IP policy over the past two decades or so — work that reflects thousands of hours of professional endeavour, angst and millions of dollars.

This figure depicts the number of reviews into the IP weeds, and often these have been very tightly focused on a particular sort of IP right rather than considering a suite of IP rights. And we know that an array of rights is almost always used by firms and creators to protect expressions of ideas in the modern age.

And it is the siloed nature of these previous IP reviews that has rendered them less effective. Where the concentrated costs of change are readily accounted for fully, while the diffuse and at times unquantifiable future benefits to the community are considered partially at best. And it’s hard, if not impossible, to make good public policy when you’re only thinking about some of the public.

Only three reviews have taken a whole-of-IP approach in the last two decades: the 2000 Ergas Committee on Intellectual Property and Competition Policy Review, the 2008 Cutler Review of Australia’s innovation system and then the Productivity Commission most recently.

The Harper Review of Competition Policy explicitly recognised this when it considered IP matters. It’s why they recommended the Productivity Commission analyse the IP system from a broad perspective. And the Government not only endorsed that recommendation but sent us a very broad terms of reference — the ultimate public policy circuit breaker.

And it’s in the Commission’s DNA to take such an approach. Indeed, our Act requires us to take a community-wide approach: to look at the IP forest rather than particular trees. The only shackle on the inquiry was the requirement to be bound by existing international agreements, but not to the extent that prevented us from making recommendations about how to improve such agreements in the future. And we certainly accepted the invitation to do so.

And with a community-wide view in our DNA, we invest much in community consultation and transparency. It is very much a ‘you tell us’ approach to public policymaking. Where all we ask of inquiry participants is to show us the evidence … and to be honest. And to harvest this evidence, we held 6 public hearings (hearing from just over 120 inquiry participants), we held 69 meetings with creators, consumers and experts, we conducted four round tables involving around 50 participants, we examined and consulted across seven different jurisdictions to get an idea about what was specific to Australia and what was not. And this is before considering the 620 plus public submissions made to the inquiry — every one of them read and studied. Along with our own original analysis. This is how we establish our evidence base.

And because we take an evidence based approach, we even (heaven forbid) change our minds when presented with compelling evidence. This can be seen in our final report where evidence in hearings and the second (post draft report) round of submissions did change our minds (from draft to final report). As can be seen in the areas of business method and software patents, and in plant breeders’ rights, and even in the form of fair use that we recommended in the final report. We conceded, and rightly so, that the smart folk at the Australian Law Reform Commission got the framing of fair use exceptions right and we strayed.

So it does beggar belief that some folk have suggested our report ignores the evidence. For those folk, it is the very breadth of our evidence that helps us to assess what some claim to be evidence but what on closer examination proved to be groundless and (at times) self-serving assertion.

Now balance matters in the high wire act of getting IP policy settings right. Crafting an incentive for creators and innovators to bring ideas to market, while making sure those incentives don’t cruel welfare of the broader community is no mean feat.

And the community had a lot to tell us about the balance of IP — and that the balance was out of kilter for some rights. For some, the balance was fundamentally broken – even if it still represents the finest legal thinking of the 19th century. So the inquiry’s immediate goal was to work out how to fix the balance, but also recognising that mechanisms needed to be put in place to keep the appropriate balance for future generations.


Perhaps what I mean is best shown by our examination of patents. For here we found that too many are granted to low-value innovation. And many are used for less honourable motives. We heard evidence of patents being used strategically to prevent follow-on innovation and stymy competitive forces, to delay the introduction of cheaper generic drugs (at an annual cost of a quarter of a billion dollars).

So we made seven recommendations to fix these problems. And then to make sure it stays fixed, we recommended an ‘objects clause’ — a legislated roadmap for future courts on when and how patents should be granted.

Now no one will argue with the principle that patents are supposed to reward socially valuable innovation and inventions (except perhaps some patent attorneys). For the new ideas and ways to implement them are ultimately what drive wellbeing in society. But in practice, the Commission found a large proportion of patents are granted to ‘low value’ ideas. Think a pharmaceutical of identical formulation to a predecessor, but just with a different dosage. Think a pizza box that folds out into a bib.

This isn’t a new problem, but it’s one that other jurisdictions (like Europe) seem to have had greater resolve to fix relative to Australia. And while there have been local efforts to ‘raise the bar’ — to make getting a patent harder — we had to assess the assertion that we had raised the bar enough down under by examining the outcomes. And in doing so we discovered that assessing patent eligibility had seen very modest change.

We examined the patents (for the same innovations) that had been granted here and in Europe since we raised the bar — and it looked more like raising the limbo bar at a toddlers’ birthday party — no one lost out.

For the Commission getting patents policy right is akin to the John West business model. It’s the fish that John West rejects that makes John West the best. And our original analysis revealed that Australia despite purporting to raise the bar continues to grant a lot of patents to innovations that the EU rejects on the grounds of not being good enough. So there is a long way to go before we are the ‘John West’ of patent policy.


Then there is fairness in enforcement. We heard from participants about the high cost of enforcing IP rights, particularly when a court is involved. And indeed it was a concern of authors with any change to the copyright exception provisions. One participant described the situation:

"… we have a Rolls Royce system called the Federal Court. You go there. The starting price will be $200,000 minimum… Take it from there. $400,000, and then you might have the costs of the other side".

A lot of IP disputes don’t need the Rolls Royce; they can make do with an agile, speedy Vespa. To alleviate these costs, we drew on the experiences of the UK’s Intellectual Property Enterprise Court. My fellow Commissioner Jonathan Coppel and I met with Justice Richard Hacon (the head judge of the UK IPEC) — a terrific meeting where Justice Hacon took us beyond the research and conveyed why the UK model has worked where others had floundered. By capping costs, trial times and damages, dispute resolution costs are reduced and firms have greater certainty. But most importantly the separate list had allowed the discipline of low cost DNA. And it is for this reason we recommended that the Government should introduce a specialist IP list in the Federal Circuit Court, encompassing features similar to those of the IPEC, including limiting trials to two days, caps on costs and damages, and a small claims procedure. For such a low cost, DNA appears to be alive and well in our Federal Circuit Court. And contemporary research from the UK shows the IPEC model is delivering access to justice to a large number of creators that would never have defended or challenged rights in the past.

Now enforcement might sound tedious, but it is at the end of the day an enduring element of what is fair and what is good public policy. So again we return to equity — access to enforcement is access to justice alike for authors on copyright and firms, especially SMEs or new entrants, for patents, design rights and trade marks. It is also a way of future proofing IP policy so it remains fair, balanced and in the interests of the community today as well as tomorrow.

Now at this point, a few of you may be quietly thinking ‘I thought I was attending a fair use conference, but now I’m being lectured on patents and enforcement’. So let’s talk copyright.


Our patent recommendations were largely about addressing what is perhaps best thought of as unfinished business — a material residual imbalance. In contrast, we found our starting point for copyright policy was arguably about trying to find any semblance of balance. Term (at life plus 70 years) and scope (with our current exceptions) are not balanced, and are firewalled from change by international agreements. But we looked and found some areas where meaningful reform can and should be made.

Thanks to geoblocking, Australians pay more for digital content (around 67 per cent more for music) or get less or latter access (like the diminished library of titles available on Netflix in Australia relative to the US). You know something is amiss when the haves and have nots are delineated by who has a teenager in the home capable of circumventing the geoblock. We heard from many participants there is legal uncertainty about the ability for consumers to access legitimate overseas content. And this is the only fair — and indeed workable — weapon to counter online piracy. Creating fair access and eroding the unfair geographic price discrimination that is geoblocking. So we recommended that consumer rights be clarified (and this also applies to ensuring that rights holders can’t contract around copyright exceptions, or rely on technological protection measures to prevent legitimate uses).

Turning to copyright collecting societies. They play an important role for rights holders and they can make a meaningful difference in lowering transaction costs for authors, creators and content consumers. But they can also wield market power. This lifts the governance high bar for what we need to see from a transparency and accountability perspective from these agencies. There have been questions in this inquiry about the effectiveness of the Code of Conduct for Collecting Societies. And we learned in meetings with UK and European experts, and even their collecting societies, that they had lifted the governance code bar in a substantive way and in their view well above the down under code of conduct. So we recommended that the ACCC review arrangements for collecting societies with a view to strengthening governance and transparency, ensuring that the current code represents contemporary best practice (in substance and form), balances the interests of societies and licensees, and whether the code should be made mandatory. For at the end of the day, and as a de minimis, you need to be able to follow the money. And we couldn’t and nor could rights holders or rights users.

Busting the myths: Parallel Import Restrictions

Turning now to the myth busting part of our inquiry — and here it seemed like a monumental sand dune of argument and assertion to be traversed. Three steps up and then two back. And this was especially the case when it came to any mooted change to copyright, and especially parallel import restrictions on books and fair use.

The inquiry was told definitively by publishers that parallel import restrictions do not raise book prices and was provided with some purported evidence to that effect. But on closer examination this just didn’t stack up. So the Commission purchased data on book prices, compared more than a thousand like-for-like titles in Australia, the UK and the US, and found that books were indeed more expensive — by around 20 per cent on average — than in those other jurisdictions like the UK. Myth busted.

The inquiry was then told by publishers and authors that parallel import restrictions are crucial for local markets and to support local authors. But, alas, this stumbled in considering the workings of the market — for PIRs don’t just apply to books by Australian authors. Hilary Mantel’s books get the same protection as Hannah Kent’s, with the benefits largely going to offshore authors and publishers. So PIRs are effectively a tax on readers in Australia, and the publishers the revenue collection agency. And the higher costs of books are borne by all Australians from the bibliophiles, to the students as they (or their parents) are forced to pay more for Harry Potter, Diary of a Wombat and the dreaded text books.

And we know from our previous analysis that from the annual $25 million book tax (from PIRs) around $15 million flows offshore. So it’s hard not to view PIRs as anything but the least effective way to support local authors and perversely at the expense of local readers. We thought about limiting PIRs (and their tax impost) to only the books of local authors – so at least the support is targeted at local authors (although we’re still not quite sure how much of this they see and I’ll come back to this later). But alas the shackles of our international agreements have relegated that option unavailable. So direct government support becomes the policy no brainer if the goal is to cost effectively support local writers and creators, without harming their readers and with the added bonus of cutting out both the middleman and offshore authors. And we explored this angle more in our final report — including establishing that the Government (and ultimately taxpayers) provide around $40 million of direct support to local authors today.

And on the middleman — we did listen to the case made by locally based publishers that the additional money made from PIRs delivering them higher prices is then used to cross subsidise local authors. So we requested this evidence — show us the money and what you do differently to your counterparts in the US and Europe. But we were met with the sound of deafening silence. So again we could not follow the money. Myth busted.

The inquiry was then told that removing PIRs destroyed the New Zealand publishing sector and decimated New Zealand authors. Indeed, based on some of the submissions and commentary made to the Commission, one might expect that literacy had all but vanished in Middle Earth.

But when the inquiry looked closely at these claims, the timeline didn’t stack up — a gap of more than a decade between PIRs being removed in 1998 and the global restructure of the publishing sector which unsurprisingly reached New Zealand given its market size and locale. Moreover, the removal of parallel import restrictions in New Zealand does not appear to have had significant negative effects on domestic creative effort in the books sector. Analysis by Deloitte Access Economics in 2012 (some 14 years after PIRS removal in New Zealand) found that the number of new NZ book titles that published annually has remained fairly steady. Data on the number of authors shows that, following the reform, the share of authors in overall employment has increased in New Zealand. So rumours of the demise of Kiwi authors are just that – rumours and not evidence. Myth busted.

The inquiry was then told that removing PIRs would lead to the dumping of cheap books printed overseas into Australia.

Again we asked for the US based evidence from the publishers that they purported in our public hearings. But again all we heard was the sound of silence. It’s a hard task to check something that hasn’t happened, but the Commission examined the claim by looking at who actually publishes what in different markets. Using more than a thousand like-for-like titles across the Australian and UK markets, we found that about 95 per cent of books were published in both markets by the same publisher or subsidiary. So the threat that you’ll materially erode your own profit margins if you don’t get your way is not the most compelling business case nor corollary public policy argument. So myth busted.

Busting the myths: fair use

The inquiry then turned its attention to fair use, where the same underlying issue of imbalance persists but it is a faster growing divide.

In a nutshell, the existing fair dealing provisions provide prescriptive exceptions to use copyright material, whereas fair use is a more principles-based approach to dealing with copyright exceptions. The biggest difference between the two in operation — prescriptive exceptions are glacial at best to respond to change, where principles-based exceptions can adapt and respond more readily. The glacial adaptive experience with fair dealing is best captured in legislative refresh around recording shows on VHS and time shifting using PVRs. The family VHS VCR was mothballed down under by the time our copyright act recognised its form of copying.

So the question is one of whether prescription or principles is most appropriate in a modern economy of today and tomorrow. And it is here there’s a paramount point of distinction between PIRs and fair use. We know with parallel import restrictions that technology, the digital age and new business models have proved a great equaliser. Digital books, real time publishing (as we are seeing in countries like France) will continue to discipline the price premium local publishers will extract with PIRs. So perhaps where we find ourselves today, with PIRs costing Australian readers around $25 million each year, is about as bad as it will get.

And while technology and the digital age reduce or constrain the costs of PIRs, the same cannot be said for our system of copyright exceptions. And here’s the policy rub and where the greatest policy imperative looms largest for government. For the inequities and costs of fair dealing are growing and will continue to do so with technological and digital advances.

So it’s critical to put fair use very closely under the magnifying glass.

It also required the heavy glass frames of the myth busters. One claim was that fair use would lead to increased court costs and uncertainty. The question about courts and uncertainty is a complicated one.

The Commission consulted widely on this issue, and the community-wide response was far more negative about the existing regime than one of fair use.

The Commission heard stories about librarians being unable to provide material to the community due to uncertainties around fair dealing. The Commission heard about the gains that could be made by making greater use of grey literature, to which fair dealing did not always extend. The Commission heard how fair dealing was constraining and costing our local documentary film makers. The Commission heard directly from Universities Australia about how institutions were reluctant to use material for Massive Open Online Courses — MOOCs — because fair dealing might not extend to them. The Commission heard about how the status quo meant that millions of dollars of public funds are spent each year to pay license fees for freely available internet materials and even thumbnail images of book covers so that they can be used on school intranet sites. Written evidence from Council of Australian Governments that Australian schools are paying the Copyright Agency over $9 million each and every year for material that is freely available on the internet. And we know there is a further $11 million each year that the Agency collects and cannot redistribute. So it goes into a pool to be distributed to members who were not the creative originator. On listening to the full spectrum of consumers, creators and curators, the story that emerged was one where the status quo was uncertain and inefficient, and in spite of the name, anything but a fair deal.

The real question then is: could fair use be worse? Having already addressed some concerns about court costs and access to justice separately, it is really a question about whether it’s appropriate for rights holders or for content users and ultimately an impartial third-party, like a judge, to determine when an exception to copyright should apply and when is it fair.

Under the model proposed by the Australian Law Reform Commission (ALRC) and endorsed by our inquiry, fair use in Australia would use four fairness factors — purpose, nature, substantiality and the market effect. Now courts are well-versed in applying principles-based laws in many areas such as consumer and employment law. And we also met with folk from the US who showed us practical guidance materials on how teachers, libraries, businesses use such guidance to confidently apply such factors in their day-to-day lives — and these guidance notes which abound in the US, could be readily applicable to Australia.

And in the Commission’s view, there’s ample evidence, both at home and abroad, that with such guidance the community can be trusted to employ fair use fairly. Myth busted … and with a modicum of certainty.

Another simpler myth to bust is the claim that fair use is really free use.

This is simply an oxymoron – it cannot hold as an assertion because of the 4th principle — market effect. The market effect on rights holders is a key component of the fairness factors and what’s allowable. So a use that erodes the market potential for a creator is simply not allowable under this principle. So we asked the publishers and the authors to give us examples of what they are being paid for today under fair dealing that they would not be paid for tomorrow under fair use. And we either heard stony silence or we heard of two US examples — Google books and the case of the transformative rapper.

They argued that Google’s “open slather” digitisation of US library books is tantamount to free use. But the US courts did not agree with this portrayal. They instead found that Google’s Library Project did not provide the books in their entirety as a substitute for original works, and instead only provided very small snippets. And most importantly in assessing the fourth fairness factor — the effect of the use upon the potential market for or value of the copyrighted work — the courts found the snippets did not fall foul. Where the snippet view provides a researcher or student with all the information they need to know and they did not then buy the book in its entirety, the Courts examined the evidence and found that this type of information was most likely to be factual in nature and therefore not even subject to any copyright. Moreover, if you step back for a moment and think about what the Google Library Project represents — it is no more than the 21st century equivalent of browsing in a bookstore. So Google is supporting book markets and thereby authors.

Myth busted.

It was claimed that fair use destroys publishing industries and has done so in Canada, and particularly their educational resource sector. That claim did not stand up to even modest scrutiny: the experience in Canada has been grossly misrepresented and ignores specific market factors there. To begin with, Canada doesn’t even have a system of fair use — they have fair dealing. And our Canadian cousins also jettisoned its educational licensing regime — we have not.

But that didn’t get in the way of some trying to shoe-horn unrelated factors in Canada into a story of potential Armageddon in Australia. And to sell a story of Armageddon you need a big number. The number that’s been oft cited by some local luminaries is that fair use would cost the Australian economy $1.3 billion. The number is based on work by PwC and commissioned by rights holders, and curiously contains the following disclaimer:

This Report was prepared for APRA AMCOS, PPCA, Copyright Agency│Viscopy, Foxtel, News Corp Australia and Screenrights. In preparing this Report we have only considered the requirements of these organisations. Our Report is not appropriate for use by persons other than these organisations, and we do not accept or assume responsibility to anyone other than these organisations in respect of our Report.

As the CEO of an economics consulting firm in a previous life, this is a revealing disclaimer. So we read on. We read the entire PwC report cover to cover. And we found it to be an accurate disclaimer.

But there was a modicum of economics in the report. In particular, the following in relation to the effect of Canada’s introduction of a broader fair dealing provision for educational material:

These impacts, while significant for the industry, represent transfers (i.e. from creators to users) rather than economic costs. (That is, if secondary derivative works are not truly transformative, then fair use would merely represent a transfer of supply and demand between various groups within society and would not represent ‘net new’ economic growth.)

So even if we are to accept at face value our local luminaries oft cited cost of $1.3 billion if Australia were to adopt fair use, this would represent a transfer to Australian readers and consumers of the copyright material. The libraries, the new business entrants, the students, the MOOC makers and the local MOOC recipients. So their big number actually represents a big benefit to many Australians.

So whilst we spent some time carefully unpacking the assertions and claims in the PwC report (and a read of our box on page 197 of our report provides the highlights and a sobering read), late in the day, the inquiry also had access to another resource: a cost-benefit analysis undertaken by Ernst and Young for the Department of Communications. This report specifically analysed the winners and losers from moving from fair dealing to other arrangements, including fair use. It was a refreshing read — a considered albeit conservative analysis of what might happen today if fair use came to Australia. It was a here and now analysis not forward looking. It revealed that there was no immediate Armageddon from fair use, rather there would be immediate net economic benefits. And that’s before taking into account how the shortcomings of the status quo affect matters into the future.

So allow me to share some forward looking thoughts. Because it reveals that moving from fair dealing to fair use is not a zero sum game as many portray.

Think, no access to data for data mining means no incentive to the workforce to develop those skills — skills which other jurisdictions are developing in spades.

Think, hampering access to cloud computing means that Australian firms and families are left to use inefficient, antiquated systems in comparison to other markets and countries that can make use of the latest technology.

Think, schools and universities not paying $9 million each year for material that is freely available.

And as flagged earlier, think, providing universities and educators fair and certain access to material for MOOCs will enable a new way to skill and reskill our workforce. And this is perhaps one of the most compelling equity issues hidden away in the fair use free for all. For it’s not just about the millions of lost export dollars of our universities being constrained and unable to develop and export MOOCs.

It’s about what’s needed to re–equip our workforce to remain relevant. Research reveals that the nature of work is changing such that education needs to be continuous and there is the need to make adult learning routinely available for all. A university student today will have 17 different jobs and what they learn at school or university in no way represents the conclusion of their formal learning if they are to remain productive and more importantly employed. And if you think of the structural changes in today’s labour market, mature age workers facing or avoiding redundancy, today’s workers will need to readily tap into new ways of learning. MOOCs will play a vital role in doing so; and fair use in Australia will play a vital role in making sure they can.

Think Israel — in introducing fair use did so with a mind to what would be of future benefit to creators, innovators and educators. So the world’s cultural and innovation pin up country “gets it”. Indeed it cast its policy narrative in this very way. And if we are to be a truly agile economy, this is a policy change that lends itself to an incredibly positive policy narrative. And the narrative flip side is that fair use is a policy lever to avoid the looming education divide of haves and have not’s. Nor do we need to reflect for too long to see what political and policy outcomes await if we allow that to happen. So you can see why there is no single big number of benefit. Because at the end of the day doing such an analysis is complex and simply doesn’t lend itself to a single number. But what we do know is there is no Armageddon. And there are benefits to be had and they reside where the interests of innovation and equity co-exist.

And we know that these benefits can only grow in the future as technology evolves. But perhaps more importantly, so will the costs of the policy failure if we do not jettison fair dealing to the moth ball smelling attic alongside the VCRs.

So I hope today’s tales of myth busting reveal less a case of the Commission’s ideology (as some have suggested) and more an open mind that we try to bring to bear when considering public policy change. We asked, we listened, we evaluated — using work in the public domain, international experience, work commissioned by the Australian Government, work commissioned by others and our own analysis — in order to determine which arguments are wolves in sheep’s clothing and which we can rely on to frame policy that makes a positive difference for all Australians. And our resulting recommendations to remove parallel import restrictions and introduce an exceptions regime of fair use are based on evidence and with only the interests of all Australians, not just a few in mind.

So this inquiry’s story ends with a suite of policy change across all forms of intellectual property (some 25 recommendations) that we have made to the Government. And taken in their entirety, they represent an opportunity to deliver tangible benefits to most Australians and not just a few.

Consumers and content using businesses would benefit much — from fair, certain and (for books) cheaper access to content and creative endeavour. Government and ultimately taxpayers would benefit from a substantial reduction in health costs (at least $250 million each year) by constraining the costly and strategic use of the PBS with pharmaceutical patents.

Rather than hindering innovation and creativity as claimed by some participants, IP reform would also invigorate innovation and competitive forces. Australian firms will be able to take full advantage of opportunities in cloud computing solutions. Medical and scientific researchers will be able to better utilise text and data mining. Universities and TAFEs will have the flexibility to offer MOOCs. The education sector will avoid paying millions of dollars each year to use materials that are freely available online. University students will pay less for text books and have more MOOC access. Workers needing to remain skill relevant whether due to age or structural change will also have access to skill adaptive MOOCs. Innovative SMEs will be able to innovate without fear of infringing frivolous or strategic patents and be better able to enforce legitimate rights through low-cost dispute resolution mechanisms.

All in all — there is a compelling policy narrative to be had here.

So on a concluding note, let me share a snippet I chanced upon recently. But not from Google. In the very first essay of the copyright treatise: What If We Could Reimagine Copyright?, and authored by the book’s editors Professors Giblin and Weatherall, there is a heading. And it is — ‘The ‘public interest’ (please don’t stop reading)’. It made me pause and reflect on our Inquiry report. And of the words of Theodore Roosevelt’s Man in the Arena — who spends himself in a worthy cause but if in doing so he fails, he does so by daring greatly. For in the arena of copyright policy, perhaps our final chapter should have been entitled ‘the public interest — please don’t stop believing’.

The challenge for policymakers is to focus on the near-silent majority of users, of adapters, of educators and creators that will need fair use to bring about the next wave of innovation, jobs and equitable prosperity. For its absence will simply foster a society of less haves and more have nots.

So for the Commission, fair use has become not a nice to have, or even a good to have, but a policy must have. At the end of the day we asked and answered a simple question – what is fair?

Thank you.




Related Items: