Copyright Amendment Bill, Analysis and commentary

Copyright (New Technologies and Performers' Rights) Amendment Bill

This bill makes a number of overdue modifications to copyright law, notably in the area of reaffirming the rights regarding decompilation, and introducing the idea of "transient copies" to prevent copying of audio data via memory or system bus from prompting copyright infringement.

However, there are a number of serious questions raised by other sections, most controversially the introduction of a penalty for bypassing, or providing the means to bypass a TPM (Technological Protection Measure) on a copyrighted work.

TPMs

This is in many ways similar to the concept introduced by the US DMCA. In essence, the problem is one of ease of copying, and expertise:

1. Digital works are trivial to duplicate, and so the normal barriers to duplication on a physical work no longer exist, and therefore (presumably) people will illegally duplicate copyright works more.
2. While technical measures can be put in place to prevent this, there are experts who are capable of creating a digital work that can bypass these measures. Due to (1), these "TPM spoiling devices" will rapidly be adopted by the original works market, thus defeating the barrier and leaving us at square one.
   

The DMCA-style solution to this, is to make the distribution of a "TPM spoiling device" illegal. The penalty for distributing one is then said to be equivalent to having actually performed the copyright infringement yourself.

The simple solution is, however, put in serious doubt by a number of questions:

1. While the Act permits the copying of a work under a number of reasonable circumstances, a TPM must prevent or restrict copying regardless of its legality, since it is not possible to determine the legality of the act of copying itself in software.
2. Significant questions are raised as what precisely constitutes a TPM. While in a legislation-free environment a TPM must be effective in order to be useful, the addition of legislation permits a copyright holder to do little more than the bare minimum to qualify as a "protected work", and then use the legal system to make infringement claims against citizens who stumble upon the trivial methods to bypass the TPM and talk about it. This is a step *below* what we have now, where at least the only person subject to legal proceedings is the actual infringer.
3. While the Amendment provides two remedies of appeal to the copyright holder, and appeal to a "qualified" party to remove TPM on a work for the purposes of fulfilling a right given by the Act, there is no motivation for the copyright holder to actually fulfill this obligation - they rarely have any better means of determining the legality of the action than the TPM itself does - and no direct provision for an organisation dedicated to the removal of TPMs so that citizens can exercise their rights is made, instead the responsibility is placed on the shoulders of libraries and educational establishments, neither of whom will see this as their core responsibility.
   

The predictable outcome is the least desirable environment. Works will be released protected by the same TPM used in every other legal environment, maximally restrictive. Remedies to permit the exercise of the rights to record for the purposes of format-shifting, time shifting, adding subtitles or making complaints will be difficult, time consuming and often unavailable.

Single Copies

A oft-used phrase within the Amendment (and probably the Act itself), is the idea of "1 copy". You are permitted, for example, to make "1 copy" of a recording for every playback device you own:

Clause 44, "New section 81A (clause 44) allows copying of a sound recording, subject to a number of conditions. If those conditions are satisfied, the copying of the sound recording does not infringe copyright in the sound recording or in a literary or musical work contained in it. The conditions are set out in new section 81A(1)(a) to (g) and include the conditions that the copy is made by the owner of the sound recording, that it is made for that person's private and domestic use, and that no more than 1 copy is made for each type of device for playing sound recordings that is owned by the owner of the sound recording."

However, this concept of "1 copy" takes no account of the true nature of information as used in computer systems.

Digital data is regularly backed up. It is certain that, were I to legitimately receive a sound file to my gmail email address, there would now be at least 3 redundant copies of the data. Should I fail to delete my email subsequent to downloading the attached file to my computer, there would now be 4. If I back up my computer once, 5, if I perform monthly full backups, one additional copy per month.

A similar clause requiring timely deletion under certain circumstances needs further consideration, some methods of backup (to write-only devices) do not permit the removal of individual pieces of data after an elapsed time. Nor may the owner be in control of the process that performs this backup, corporate and educational environments often have centrally managed backup programs, and some operating systems maintain backups of data automatically without the users knowledge.

Worse, the right to maintain "1 copy" is under a "sunset" clause:

There is a sunset clause attached to new section 81A(1)---under new section 81A(3) it expires after 2 years after the date on which it comes into force, unless renewed by the Governor-General by Order in Council

No explanation as to why this might be there is offered, nor is it clear why, with the overall vision of "creating certainty" about digital copying, such a significant change in rights may happen in such a short time. Should this clause silently expire after two years, I may find myself suddenly in violation despite previously having been perfectly legal - not a situation to inspire certainty or confidence.

Conclusion

While the idea of helping copyright holders maintain their rights in the face of a new form of information that makes duplication so trivial is well intentioned, the fundamental disconnect between our notion of "property" and the true nature of digitally transmitted information requires much more robust approach than has so far been proposed.

The DMCA has demonstrated how dangerous it can be to attempt to legislate physical property laws onto digital copies without careful consideration. New Zealand can learn from these mistakes.

It is absolutely clear that if copyright holds rights are to be maintained by supporting TPMs, that citizens rights must be similarly upheld by creating legal obligations on the part of copyright holders to provide unprotected copies of digital content to an organisation tasked with aiding New Zealand citizens in exercising their rights under the Act. This cannot be a half measure, funding must be provided, and penalties must be able to be exercised against copyright holders who fail to provide the unprotected copies in a timely fashion.

There is no question that digital works will become a large component of our cultural history in the future, it must not be the case that these works become slowly inaccessible to us over time due to TPMs operating to prevent us from exercising our rights.

The precise nature of the organisational methods to do this I'll leave for another entry, but it has to be done. The Amendment as it stands should not go ahead until these issues are addressed.

• Link to the Amendment
• Hon Judith Tizard, Sponsor of the Bill
• Boing Boing post on the subject
  

Thanks to Andrew Garrett for the heads-up.

Phi and Phoenix vs Australia

Short entry for this one. Myself and Phoenix (Also NZ it turns out) took on two Australians on Fields of Isis. Phoenix econ'd, while I operated on a medium pressure strategy right out the door, grabbing the choke points then maintaining a regular tempo of assaults for the duration of the game.

Poor commander placement by red initially resulted in him losing his commander early, and while blue attempted to use artillery to push me out of the chokes, a failure to follow up meant it was relatively futile.

Main items of interest were a heavy drop by me into Red, which eventually got squished by a pile of siege bots from blue but not before some solid damage was done, and more importantly the psychological pressure ratcheted up even further.

My partner managed to give me a large number of siege bots, presumably by selecting them then clicking "give units". This was an interesting twist I hadn't seen work before.

In the end, red was primarily ineffective, possibly slightly out of his league, and blue managed a respectable economy but lacked the tactical ability to make it count in engagements. A number of poor attack decisions, including one mind-blowing one where his smaller but more cohesive force sat and waiting on the edge of the range while my forces coalesced rather than attacking immediately to give himself an advantage in weight-of-fire. This resulted in my shield generators arriving by the time we engaged and he got well and truly pounded.

He died eventually in a suicidal commander rush into a small forward force of siege bots.

Phi vs Sicarius

A 1v1 ranked game on Sentry Point (3 player)

Final result, Victory for me at 27 minutes.

The game was plagued with a fairly high but consistent lag, so it seemed to take much longer than it did, giving me a constant feeling of being "behind", in the sense that my economy and tech didn't feel as advanced as they should be for any given moment.

General Strategy

I opted for a ground-centric approach. I also determined right from the start that I wanted to take the 3rd starting point if I could. Initially I conceived of it as a resource base, however it became the linchpin of my final assault late game.

Sicarius had a more balanced approach, with both ground and air, and managed to maintain an economy that matched mine despite the lack of the 3rd starting place. Most interestingly this seemed to be achieved by the production of a small number of t1 mass fabs as a substitute for the extraction points he couldn't reach. Since I never upgraded most of mine past t1 it worked fine.

My tactics

My primary failure was my very light-handed approach to anti-air. I came too close several times to having insufficient AA to deal with what was eventually a t2 gunship threat, although never in large numbers.

I did have particular success in my use of terrain however, with a significant part of the game spent south of my base in a forward position that had a number of compelling qualities.

The first, and major one was that it was the main choke between the two bases. While there was a possibility of a strike via the western entrance forces could be easily diverted.

Other benefits were that it was in easy range for solid T1 radar coverage, the civilian outpost provided additional fire over part of the access way, and a unique twist in the terrain meant that I could place artillery units "below" the hill, allowing them to fire without receiving fire.

My dual-sided approach paid off, with both bases finally acting in concert to deliver a pincer on his only base. Sicarius showed a limited ability to identify weak points and continued to hammer on my strongest fortification for the majority of the game.

Sicarius

Of interest was his build order (1-2-1 Land which I haven't seen before), and his light but appropriate use of mass fabs to keep his economy even (or in some cases, ahead).

His most significant practical issue was that he failed to deploy any kind of radar. While his visual intel was superior to mine, the lack of radar put his units at a serious firing disadvantage and often caused him to make poor judgements about my range.

At a more strategic level, he simply failed to adapt to the position I created. The forward position was not close enough to provide a significant threat (with the exception of artillery), and my (numerous) weak points went utterly unexploited until late in the game, at which point it was too late.

He also had a tendency to send him small, unformed batches of units (possibly through a lack of familiarity with the formation commands) where a larger, more disciplined force with radar might have been sufficient to push me off my "hill".

Overall

The game was remarkably fast paced for all the lag, constant attacks meant that construction at the "front" was slow and units needed replacement regularly. The need to keep Sicarius focused on the front position and the imaginary "threat" it represented meant I had to push forces forward several times into uncomfortable positions to force him to react.

Watching the replay, it was clear from reasonably early on that I was going to win, but it certainly didn't feel like it until near the end.

Replay

Review: Resplendent

Resplendent is the 4th in the Xeelee sequence by Stephen Baxter. Unlike the previous book in the series, this one really succeeds in taking the reader into the future in a way that seriously expands your perception of what it means to think "long term".

Consisting of what is essentially a set of short stories strung loosely together over a million years of human future, the book hops and skips across a species that moves from captor to conqueror, where wars go on for longer than the human race has existed so far.

The real beauty of the book is the way it slowly stretches your impression of time until, when you finally pull yourself out of the book, you can barely imagine that we believe 50 years could be considered long term planning, and the idea that life will remain static and comfortable is almost impossible to believe.

Baxter is not a great storyteller, but his ideas, as always (when he isn't obsessing about "Michael Poole") are enormous in scope and fantastic in detail. Definitely recommended for the hard sci-fi fan.

Unlikely tactics that worked

In what is likely to be some kind of rare series, "Unlikely tactics that actually worked"

This is my opponents base, from his view.



This is my view.



Needless to say, things went poorly for him from this point on.

Replay