Tuesday, September 16, 2014

Spectrum reform is more than closing community television

Reports of Communications Minister Malcolm Turnbull’s Radcomms2014 address focussed on the plan to force community television off the airwaves and onto the internet.

As with all spectrum discussions attention soon turned to the revenue that might be raised by a sale of the spectrum that would be cleared.

But the speech was far more informative of the Government’s plans for spectrum reform, and not all of it is consistent.

Currently there are four different licence types for spectrum use. Television and radio broadcasters are licenced to use spectrum through the transmitter licences issued under the Broadcasting Services Act. All other spectrum is licenced as either apparatus, spectrum or class licences.

Class licences are sometimes referred to as unlicensed - they include the 2.4 and 5.8 GHz bands used for WiFi. As class licences, so long as a use meets licencing parameters, there is no need to separately register the devices.

Spectrum licences revolve around the allocation of a specific “spectrum lot”, meaning a frequency range and geographic area to one user. However, there are still significant restrictions placed on the user to avoid interference.

Finally apparatus licences are very specific about the location of transmitters and receivers and specific frequencies. In practice most of these licences are issued from a template of licences designed for various bands.

The Minister identified three areas of potential reform. The first, under the guise of creating “a clearer and simplified policy framework to ensure transparency and accountability in decision-making” foreshadows a new regime where the Minister will be more empowered to give policy directions to the ACMA in the administration of spectrum.

The second is “to explore with industry is moving to a single licensing framework where there is flexibility available on licence parameters.” In particular this move is designed to improve flexibility.

One difficulty with the existing regime, for example, is that once a spectrum licence has been issued the ACMA is unable to licence another use, even if that use will have no impact on the spectrum owner’s application.

These moves together spell the end from the very simplified version of spectrum licences that have been pursued over recent years. The history of spectrum licencing stretches back to a paper by Ronald Coase in 1959 on The Federal Communications Commission.

Coase claimed that the FCC regulation of the content of broadcasting was not warranted by the “scarce” resource argument, as all resources are scarce butt are allocated using price based mechanisms.

(A similar argument had actually been made by Leo Herzel in 1951, though his focus had been on the standards for colour television rather than the content regulation).

The idea of spectrum licencing as an alternative to regulation gained more widespread interest in the property status of airwaves followed a 1964 article by Ayn Rand.

While the focus of all three had been on the licencing of spectrum used for broadcasting, the use of spectrum licencing in Australia has largely been restricted to the auctioning of bands that have previously been designated by the ITU as bands for mobile telecommunications.

The major exception was the actioning of MDS (2.3 GHz) licences that were used for subscription television. Interestingly those frequencies are now used by NBN Co and Optus for a mobile technology - TD-LTE.

It is unclear exactly how the Minister envisions a simplified regime working, but it presumably will look like a hybrid of spectrum and apparatus licencing. It is also unclear how the class licences fit in such a regime.

It was the third proposal relating to broadcast spectrum that drew the most attention. As well as announcing that community television would be moved off the sixth digital channel in 2015, the Minister confirmed he is staying with the Labor policy of not licencing a fourth commercial channel.

The Minister was encouraging the networks to move to MPEG4 encoding, but seems to miss the point that the networks have no incentive to do so. FTAs had to dragged kicking and screaming to complete the digital switchover.

A move to MPEG4 is potentially just as traumatic, because to be successful every receiver needs to be capable and not all would be.

Announcing that networks can have flexibility in channel use (except for subscription) is presumably designed to encourage the use of MPEG4. But it was actually the reverse, the ability of government to mandate use, that made conversion to digital possible.

The size of the prize could be considerable however. The current frequency restack plan divides the country into five zones, each with six adjacent 7 MHz channels for television. Two additional channels are reserved for digital radio. The decision to not use the sixth channel frees up 5*7 Mhz, while a move to MPEG4 with shared transponders could clear another 2 channels in each zone.

Thats a total of 21 channels. To fully utilise it would require a further restack. That is more than the 18 channels cleared for the Digital Dividend. So not rushing to release the spectrum for the sixth channel and managing an MPEG4 upgrade can pay off well.

Of course there is even more if the Minister thinks linerar television is truly dead and everything moves to the Internet.

Reforming spectrum laws is never easy. Ultimately the Government has to retain sufficient control to coordinate different industry and consumer participants to achieve efficiencies that the market alone will not deliver.

That’s a balancing act that extends well beyond the decision on community television.

Monday, September 8, 2014

When all you have is a hammer, everything looks like a nail

In advance of tonight's Online Copyright Infringement Forum I thought I would indulge myself with one more, slightly differently argued, point about online copyright.

But let me start by noting how predictable submissions on the discussion paper have been. Rights holders uniformly back the proposals, ISPs and others suggest the "problem" is the lack of availability of legal content and the need to "follow the money" to close down advertising supported sites.

What is only slightly less predictable is the commonality of global positions, with, for example, a submission from Assotelecomunicazioni-Asstel which is the business association of Italian telecom operators.

But the title of the blog post reflects the fact that the proposed solutions in the discussion paper are to amend copyright law to change the definition of authorisation, and then change the safe harbour provisions. This is a dangerous way to proceed for three reasons:

  • Any change to the law that failed in the High Court just brings us back to the beginning of a series of requests being complied with or not and potentially leading to further litigation;
  • Relying on the tightening the authorisation and broadening the safe harbour protection involves a lot more than the ISPs, especially groups like Universities who make no direct revenue from the infringing behaviour, and
  • It continues to place the focus on the "innocent" party - the ISP - rather than the infringing party - the end-user.
The issue should be dealt with less as a matter of trying to extend one definition of offence and then limiting its application and more as a matter of straight-forward law enforcement.

I regularly refer to the rules that apply to pawn brokers and dealers in second hand goods. Each state has its own legislation similar to that for NSW, which requires dealers to be licensed. A condition of the licence is to do certain things to assist law enforcement. This includes production of records to authorised officers without warrants (s17).

Telecommunications service providers already have obligations under (s313(1) to "do the carrier's best or the provider's best to prevent telecommunications networks and facilities from being used in, or in relation to, the commission of offences against the laws of the Commonwealth or of the States and Territories." 

This obligation is typically included in the provider's contract with its customers - for example, clause 3.7 of Telstra's general consumer terms states "You must not use a service to commit an offence or allow anybody else to do so." 

As is now well known s313(3) requires service providers to "give officers and authorities of the Commonwealth and of the States and Territories such help as is reasonably necessary " for (amongst other things) "enforcing the criminal law and laws imposing pecuniary penalties." 

In my understanding ordinary copyright infringements do not incur "pecuniary penalties", only damages. However, under s132AC of the Copyright Act 1968 the infringement on a commercial scale potentially incurs pecuniary penalties and imprisonment.  

Hence, on that reading the legislative powers necessary to require telecommunications providers to assist in prevention of copyright infringement already exist.

This is particularly clear in the case of blocking the commercial websites. While both Google and Communications Alliance have raised the spectre of the misapplication of s313(3) by ASIC, the reality is that in the other two cases where it has been used (one by ASIO as reported in The Australian in July 2011 and the blocking of the INTERPOL 'worst of' list) have been proportionate. 

No one has questioned the validity of ASIC in seeking to block access to the website it was targeting. The issue was that ASIC was stupid enough to block the IP address of the hosting provider rather than the domain name, and that it took three days to identify where the blocking request came from.

These issues can be simply rectified by amending s313 to include a provision that the Minister can make regulations about the way that assistance is to be provided. One regulation should address website blocking and introduce three requirements - that only a domain name can be blocked (not a page within a domain nor an IP address), that the block must redirect to a landing page explaining the site has been blocked, and that both the agency requesting and the recipient must keep a record of the request and notify their relevant regulator (ACMA for provider, AGD for agency).

While it might seem extreme to create a subordinate regulatory power to the Minister in this way, the regulation would only restrict the requests that can already be made, the instrument would be transparent, and the instrument would be disallowable.

In brief, there is no need for any law to be changed to implement a website block for sites that are infringing copyright on a commercial scale. 

In the case of individual users there is a case for creating a new enforcement agency (or task for an existing agency) for copyright infringement. The extent to which said agency should be required to provide assistance should be contingent upon the extent to which the rights holder has acted to protect themselves - in particular by making legal means to obtain the material at a fair price available to Australians.

This requires the conversation to be conducted by the Communications Minister exclusively, not the Attorney-General. With the simple amendment to s313 the Minister could create a system of assistance to be afforded rights holders. This approach has the added benefit of recognising that the Act at s314 already provides that the terms of providing assistance are agreed between the person assisting and the agency seeking assistance, and that in doing so the person neither profits nor bears the cost of providing the assistance.

I will be otherwise engaged tonight, but I'm prepared to bet the conversation will go absolutely nowhere productive, but I do intend to endure the recording. Both industries are experts in arguing from positions rather than principles. 


Monday, September 1, 2014

Should assistance to rights holders be limited

Ed Husic in the Federation Chamber today made the point that if we are going to look at mechanisms to aid the enforcement of copyright then we should also look at the issue of price.

In doing so he was picking up on one aspect of a common argument mounted by copyright infringers - that they only do it because it is so hard/impossible to get the content legally. This criticism has some validity, but it is by no means a "slam dunk."

The Husic argument is the flip-side of the rights holders' argument that the Australian Government has an obligation to assist in enforcing copyright law, for which the response should be that this obligation only exists to the extent that rights holders aren't ripping off Australian consumers.

Before people get too carried away by the rights holders' arguments just remember two things. The first is that for each individual firm the "piracy is killing us" line is a great diversion for CEO's to spin to shareholders - nothing quite like an external factor beyond your control to blame for any under-performance. The second is that the industry associations that make the point so forcefully are just doing their job, they aren't representing a "cause."

The fundamental current issue is what role should ISP's play in helping enforce copyright.

According to the AG's discussion paper:

Australia is obliged under its free trade agreements with the United States, Singapore and Korea (not yet ratified) to provide a legal incentive to ISPs to cooperate with rights holders to prevent infringement on their systems and networks.

That is a very wide description. It is an obligation to provide an incentive, and it is only to co-operate. It doesn't seem that this construction actually runs to an obligation on ISPs to take reasonable steps to prevent infringement. This was where iiNet parted company with everyone else. They were happy to assist rights holders get details of infringing parties so that rights holders could take action, they just drew the line at taking action themselves based only on an assertion. The other question is at what cost are the steps "reasonable."

There are two parallels. The first is the example of traders in second hand goods that I frequently mention. Such people are required to get and record identification information from sellers. This works as an effective mechanism to restrict them from being used as a "fence." There seems to be nothing wrong with the requirement that an ISP be required to identify and record which account was used when a specific infringing activity took place. There is equally no reason why the said ISP should not notify the account holder that the allegation of infringement has been made. But there is no basis on which the ISP should be required to take any action whatsoever to degrade or cancel a service on the basis of the mere assertion of another party, nor that the onus of resorting to a court to determine the facts of the matter should rest with the account holder (who might claim their service was unjustly restricted) rather than the rights holder.

But the Husic point is whether such assistance should be afforded to a rights holder who is merely trying to enforce a pricing regime that charges Australians more, rather than a globally equitable pricing regime and all that is sought is protection of copyright.

A similar issue arose with respect to protectionism in the earliest days of Federation. To ensure that the benefits of protectionism were enjoyed by workers, and not just pocketed by the owners of capital, an excise tariff was imposed on manufacturers that would be waived if workers were paid "fair and reasonable wages." The determination of what was fair and reasonable was actually what the Harvester judgement was about - it wasn't a "basic wage" case as we know them today.

This is our second parallel - but in reverse. Should we provide assistance to all rights holders, or only to rights holders who make their content available on fair and reasonable terms? The determination of fair and reasonable terms might take some doing - but it is a worthwhile endeavour.

There is, hopefully, a lot of constructive dialogue yet to be had about these copyright proposals. Hopefully this can be added to the mix.