Computerworld reports on an Auckland High Court injunction requiring Codeshed, a software developer, to hand over the source code of its software to a customer. The injunction is an interim measure in a wider dispute between Codeshed and its customer, centering on the ownership of the intellectual property in the code. The customer obtained the injunction on a without-notice basis (previously known as an ex parte application) which meant the customer has to give an “undertaking as to damages” – i.e. once the final dispute is heard, if it turns out the customer is wrong, the customer must compensate Codeshed for having the injunction served on it. An ex parte injunction requiring the handing over of valuable IP is a relatively extreme remedy (it can include lawyers turning up on your doorstep with a court order in hand!) but the judge decided it was justified in this case (pending final resolution of the dispute).

According to the article, the matter is now being arbitrated, which usually means it will not be publicly commented on (by the parties) any further.

No iMonopoly for Apple

In Australia, Apple has been told that it does not have a monopoly on the letter “i” for technology products. The Trade Marks ruling said:

“[Apple] has not therefore demonstrated to my satisfaction that the person of ordinary intelligence and memory would be caused to wonder, or be left in doubt, about whether the Goods come from [Apple] merely because the Trade Mark terminates in the letter “i”, however that letter may be presented. Nor do I think the fact that the Trade Mark is made up of the letters of [Apple's] IPOD trade mark in reverse order would cause a significant or substantial number of relevant consumers to wonder whether the Goods were those of [Apple]“.

Amazon’s 1-click patent confirmed – with a Kiwi connection

Amazon’s infamous 1-click patent has been confirmed by the US Patent & Trademarks Office. To refresh, this patent is “A method and system for placing an order to purchase an item via the Internet” and, on the face of it, covers much of e-commerce. It has been rejected in the EU, and was not applied for in New Zealand. It has been widely criticised (rightly so) since first being granted. In 2005 New Zealand actor Peter Calveley filed a formal challenge to the patent with the USPTO. After some wrangling in the re-examination, Amazon was forced to make some very minor amendments to the patent in 2007. The amended patent has now been re-examined and affirmed by the USPTO.

While this highlights the need for reform of the patent system, a major development may occur next month when the US Supreme Court is expected to rule on the Bilksi case, involving the patentability of business methods.

Peer-to-Patent Australia opens part of the patent examination process and invites the public to review patent applications volunteered for the trial… The process is designed to tap into the expertise of the public to help assess whether a particular application is eligible for a patent.

The system is currently being run as a 12 month pilot programme, limited to business method patent applications.

The original Peer-to-Patent initiative completed a 2 year US pilot in June 2009. While the US Patent Office has yet to report on its results, the first year report from the New York Law School was positive (some criticisms are listed at Wikipedia). The US pilot was not limited to business method patents, although interestingly the patent application which received the most user interaction was such a patent – “Method, apparatus and computer program product for providing status of a process” (hmm, maybe they could also give it a unique name like ps or top?).

It is a good idea for the Australian pilot to be limited to business method patents, which covers most manifestations of software patents. It appears that IP Australia is giving good support to it. Let’s hope that IPONZ watches this closely – IPONZ (like the MED) has good IT systems and has been quite progressive with IT. If the pilot works, it will be a major improvement to a currently in-need-of-improvement system.

Of course, as with any community-driven system, its quality will only be as good as the community supports it. For open source, could it be a case of “many eyes make bad software patent applications unsuccessful”?

For the record:

• Copyright protects original works, not ideas.
• Trade marks protect origin, not ideas (or content).
• Patents protect novel, non-obvious, documented inventions, not ideas (or configurations).

Despite what the IPONZ homepage implies, there is no way to prevent someone “having your  idea”. A patent comes close, but protects the invention – not simply “an idea” as such.

It is certainly a good thing to be able to use intellectual property law to protect the improper taking of your intellectual property, but making sure someone else doesn’t “have the same idea”? Not a good idea.

Credit: good spotting by Jonathan Hunt, HuntDesign

There has been so much written on this subject (though I have yet to read anything much in favour of them) that I will only add a few brief comments to the debate.

Patents have never been considered inherent rights of inventors. They must be applied for and granted by the state subject to specific terms. They are limited in scope, duration and availability. As was once taught in Form 5 History, the origin of patents in our legal system was the “monopolies” granted by the Kings and Queens of England. After various abuses and reforms (some by way of the English Civil War), the modern system of patents emerged. The economic rationale of granting limited patents was to encourage innovation by protecting the investment made in creating those innovations. By and large, this system worked well over several centuries and could, in fact, be shown to have encouraged many key innovations. In other words, the system worked.

Enter software patents. These can be shown to have the opposite effect (or at least have the likelihood of that) – discouraging innovation, or in some cases attempting to shut down innovation altogether. Software patents operate to limit the possible uses of an infinitely configurable device – the computer. Virtually all computer programs, except the most basic, low-level electrical systems, rely on implementing processes and functions to manipulate and configure a computer to produce a desired result. With software, there are no physical constraints as to how the functions and processes could be used, merged, integrated, or otherwise hacked. The result is an unfettered ability to innovate. This can include, where permitted, freely adapting or integrating someone else’s code to create an entirely new program (the basis of open source software).

Should this ability to innovate be blocked – possibly at a fundamental level – by the fact that someone else has patented the manipulation or configuration of a computer in a particular way? To do so is contrary to the current purpose and rationale of patent law. Software patents have a clear tendency to limit the innovations which may be derived from computers, for economic purposes. Patent law is not intended to protect commercially valuable intellectual property (although that is a valid economic effect). It is intended to encourage innovation. When the opposite result is occurring, it is time to either change the law to correct its operation (by banning software patents), or acknowledge the problem and redraft the law in light of a changed purpose of protecting commercially valuable intellectual property. The stated purpose of the Government’s review is:

“to ensure that [the New Zealand patent regime] continues to provide an appropriate balance between providing adequate incentives for innovation and technology transfer while ensuring that the interests of the public and the interests of Maori in their traditional knowledge are protected.”

Software patents are an international issue. Successive Governments here and overseas make endless statements about “embracing the digital age” to acheive a “high value economy”. Whether the software patent problem is fixed – and soon – will be an early test of their commitment to that cause.

Two recent events concerning trade marks, one in New Zealand and one in the US, highlight the same underlying issue, albeit in different ways.

The first incident: local blog Editing the Herald received a cease-and-desist letter from APN, publisher of the NZ Herald, on the basis that EtH was using a modified version of the Herald masthead, a trade mark of APN.

The second incident: the author of a book on Ubuntu, whose website used the Ubuntu logo without authorisation from Canonical (owner of the US Ubuntu trade mark) sparked much comment by claiming that “trademarking is almost totally incompatible with the essential freedom offered by open source. Trademarking is a way of severely limiting all activity on a particular product to that which you approve of.”

Both of these incidents highlight that trade marks are not about “ownership” in the way that copyright is, but are about identity and origin. As the Intellectual Property Office of New Zealand says:

“A trade mark enables businesses to distinguish their products or services from similar products or services offered by competitors.”

If you are the registered owner of a trade mark, then you have the exclusive right to use that trade mark in connection with the products and services that it covers. If anyone else uses your trade mark (or something similar to your trade mark) without your permission in the course of trade, you are entitled to take action against them. Use of a trade mark that is not in the use of trade is not infringement, however the meaning of “in trade” is wide.

It is important to note that a trade mark will usually only apply to the goods and services that it is registered for. That is, it does not amount to a blanket monopoly on a word or phrase. For example, the word Kiwi is registered as a trade mark by a number of companies for different products such as shoe polish, bacon, beer, milk and packaging. If you owned the Kiwi trade mark for beer, you could not prevent someone using the Kiwi name to sell clothing, for example.

The trade mark (logo) for “New Zealand Herald” is registered for the following goods and services:

“printed matter; publications; paper and cardboard; stationery; posters and photographs; instructional and teaching materials (except apparatus)”

and

“publishing; electronic publishing; education, entertainment and cultural activities in this class including (without limitation) the organisation and running of competitions for education and/or entertainment; information services (including, without limitation, on-line information services) relating to education or entertainment”

The terms “publications” and “publishing” are very broad, and would clearly apply to someone publishing a blog. So using the New Zealand Herald logo (or an imitation of it) on a blog without authorisation would infringe the trade mark, if the use is “in trade”. But is a non-commercial blog “in trade”? EtH sums it up well: “I would have at least an arguable case in court but it would cost both money and time, only one of which I have in appreciable amounts”.

Should there be “parody rights” for trade marks? Probably not – trade marks are used to designate origin, not protect content. Copyright, on the other hand, should expressly protect parody as fair use.

As for the trade mark and open source issue, this is far from the first time it has arisen. Indeed, early on in the history of Linux, one opportunistic William R. Della Croce, Jr registered the US trade mark Linux and attempted to extract royalties for its use. Fortunately the matter was soon resolved, and the US trade mark is now owned by Linus Torvalds.

The Ubuntu author’s statement that trade marks are “totally incompatible with the essential freedom offered by open source”, is off the mark. The “essential freedom” of (fully) open source software is that you are free to copy, modify and redistribute the code. If someone releases software under their trade mark, you are free to take their code and re-release it under your own name or trade mark (subject, of course, to any terms and conditions of the specific licence such as attribution). Of course, the trade mark owner may allow you to release it under their name. But that is their choice. The unfettered ability to release software under any given name (and therefore possibly mislead as to its origin) would certainly be a “freedom” but is not the essential freedom.

The inverse makes the point. It would be unfortunate if a someone could simply take any well-known open source project, modify it into a buggy version, and then promote it under the original name. Admittedly, in some cases the issue is slightly less clear cut, as in the Firefox / Iceweasel tussle. In that situation, a more light-handed trade mark policy may have solved the problem.

Ultimately, it is the “essential freedom” that wins the day. If a project becomes too oppressive on a matter such as trade marks, it can be forked and rebranded, the prospect alone of which may help common sense to prevail.

In a case that will possibly influence New Zealand law, the High Court of Australia (their highest court) last week handed down a major decision that makes it harder for companies to claim copyright over compilations and databases such as television schedule listings.

After a hearing involving no less than six QC’s (or Senior Counsels as they are now known in New Zealand and most Australian states), the High Court unanimously ruled in IceTV Pty Limited v Nine Network Australia Pty Limited that IceTV, which provides electronic programme guides, did not infringe the Nine Network’s copyright by partially reproducing its television schedule listings.

TV networks have long regarded their schedule listings as copyright material. In fact, it is common for networks to generate revenue by selling the rights to publish the listings in newspapers, magazines, etc, as seen here. As the network owns copyright in their TV listings, no one else is allowed to copy or display them without the network’s permission. There have been a number of cases on this issue over the years, and similar issues such as copying parts of telephone directories, with varying decisions (e.g. the well-known US Feist Publications case).

IceTV uses the time and tile information from the schedule listings of Australian TV networks (via publications such as the TV Guide) and combines them with its own research to produce its electronic programme guides. This seems innocuous enough – after all, why would the TV networks complain about a service that makes it easier for viewers to watch their television shows?

The first answer is that it is standard commercial practice for corporations to protect their intellectual property. The second and more cutting answer is: because the IceTV guides can be loaded into Personal Video Recorders which can be programmed to skip the advertisements. Some networks do provide “official” online schedules that can be used with PVRs, but these can only be used with “approved” devices which prevent ad-skipping, whereas IceTV’s schedules will work with “unofficial”, third-party devices that enable ad-skipping.

So Nine Network sought to prevent IceTV from using its TV listings by claiming that IceTV infringed copyright in those listings. If Nine Network was successful, then IceTV would have needed to obtain a licence from the network in order to incorporate the material into its personal guides, which (it was said) would effectively have shut down IceTV’s business.

It is important to note that Nine Network does own the copyright in its TV listings – that was conceded by IceTV, and it has long been recognised by the law that copyright does exist in such compilations of facts (in this case, the times and titles of upcoming TV shows). If IceTV copied Nine’s listing in their entirety – including the time, title, episode summary, rating, etc – then that would certainly breach copyright.

However, a key fact in the case was that IceTV did not copy Nine’s listings in their entirety. Instead they used only the title and time information to correct and fill in gaps in schedules that they largely produced themselves by watching TV and “reverse engineering” a basic schedule. Nine argued that this still amounted to the copying of a substantial part of their listings, and therefore infringed their copyright.

Under New Zealand and Australian copyright law, copyright is only infringed when a work is copied (or otherwise improperly dealt with) in its entirety or in “substantial part” – section 29 of the Copyright Act 1994. In other words, copying a “non-substantial” part of the work does not infringe copyright. Of course, the trick is determining what “substantial” means. As this case and others have shown, different judges can come to different conclusions on the same facts.

In the IceTV case, the central question was essentially whether the time and title information that was copied from Nine’s listings was a “substantial part” of the work (the listings as a whole).

The High Court said no, overturning the opposite finding of a full bench of the court below. The High Court cited the following principle:

“[T]he more simple or lacking in substantial originality the copyright work, the greater the degree of taking will be needed before the substantial part test is satisfied.”

While acknowledging that the programming schedule was developed through a creative process, the court found there was very little “originality” in the actual TV listings stating the titles and times of upcoming programs. The ideas behind the programming schedule may have been highly original, but the expression of the schedule in the form of the published TV listings was not – as the court said, there is very little choice in how to express a time and title for a TV show. Therefore, the originality of the TV listings was very low, and the amount of copying that would amount to a “substantial part” was high. On that basis, applying the principle stated above the court decided that the amount of copying by IceTV was not substantial and therefore did not infringe copyright.

Importantly, the court downplayed arguments relating to the need to protect Nine’s “investment of skill and labour” which had found favour in the court below. The High Court said that while the skill and labour that has gone into producing a work is still a relevant issue, that issue should not distract from the main inquiry, which is whether a substantial part of the actual work (and not the ideas behind the work) was copied.

A large part of the argument in the court below was whether or not IceTV had “misappropriated” Nine’s work. The court below found that it had, and this was a major reason for that court ruling that IceTV had infringed Nine’s copyright. The High Court disagreed with the relevance of this argument and noted:

“A finding that one party has “appropriated” skill and labour, of itself, is not determinative of the issue of infringement of a copyright work. The [Copyright Act] does not provide for any general doctrine of “misappropriation” and does not afford protection to skill and labour alone… [The] statutory requirement that the part of a work taken must be substantial assumes there may be some measure of legitimate appropriation of that investment.”

The court also engaged in a limited economic analysis that will be of great interest to those who believe that copyright law has become too restrictive in modern society:

“Rewarding skill and labour in respect of compilations without any real consideration of the productive effort directed to coming up with a particular form of expression of information can lead to error. The error is of a kind which might enable copyright law to be employed to achieve anti-competitive behaviour of a sort not contemplated by the balance struck in the [Copyright Act] between the rights of authors and the entitlements of the reading public.”

While the carefully worded and nuanced judgment is bound to be the subject of much debate, it does open the door for the increased legitimate (i.e. “fair dealing”) use of copyright material from compilations and databases such as TV listings, for personal and commercial purposes.

Note: the Government is proposing to repeal of this rule. As of April 2009, the amending Bill (carried over from the previous Labour-led Government)  sits at number 18 on the Government’s Order Paper (right after the Dog Control Amendment Bill), so the rule may not be repealed for some time.

The commissioning rule – who owns 1^st copyright in code?

If a software developer is hired (i.e. “commissioned”) by a customer to develop some code, an important question is who owns the copyright in the resulting code – the customer or the developer? The rule that governs who owns commissioned works is known as the commissioning rule, which in New Zealand is contained in section 21 of the Copyright Act 1994. Under that rule, the answer will typically be one of the following scenarios:

1. If the software developer is a person (as opposed to a company) who was hired as an employee (part time or full time) by the customer, then the customer (i.e. employer) will own the copyright in any original code produced by the developer.
2. If the software developer (either a person or a company) is contracted by the customer to produce some code, and is to be paid for it, then the “default rule” is that, again, the customer will own the copyright in any original code produced by the developer. Importantly, the customer will have ownership of the software whether or not they have actually paid the developer for it.
3. If the software developer (either a person or a company) is contracted by the customer to produce some code, and both agree that the developer will own the copyright, then the developer will own the copyright in the code they develop for that customer. The agreement would usually be in writing (though it need not be a formal, or even signed, contract). A verbal agreement will suffice, though without any written evidence, it may be difficult to prove what was agreed.

The simple rule is that if the software developer and the customer agree in advance as to who will own the copyright, then that agreement will prevail.

However, in reality as many will know, the contracting/paperwork side of software development can be a bit of a mess. Projects are started without contracts being place, undocuments variations and “sub-projects” may emerge, “interim” off-contract solutions are implemented, and final agreements may never be signed.

In particular, with the increasing use of agile development methodologies, the emphasis is even more on delivering working code than negotiating contractual terms – though as always, getting a contract in place before starting is highly recommended.

When the parties have not agreed (usually in writing) who will own the copyright, then the “default rule” is that the customer will own the copyright (scenario 2 above).

Copyright ownership and code reuse

Copyright ownership is of particular importance to software developers who reuse code – which means virtually all developers. Simply, unless a developer owns copyright in some particular code (or obtains an appropriate licence), they are not entitled to reuse that code anywhere else. This applies even if the developer wrote the code themselves – no copyright ownership, no reuse. If the developer does not own the copyright, then the copyright owner’s approval (or their authorised licensee) must be obtained.

This is the basis on which open source software works. Under the GPL version 3, for example, a developer is licensed to reuse the code as they see fit (subject to some conditions). The developer does not actually own the copyright of the open source code they are reusing, but they are licensed by the copyright owner(s) to reuse the software in their projects.

For in-house code libraries, it is essential that the developer retains copyright in the relevant code they develop. Often, this code will not be of particular commercial value to any one else but the developer who wrote it, but it is very valuable to the developer to be able to reuse it in other projects. It should be noted that if a developer already owns copyright in code that is used in a customer project, then the developer will not be at risk of losing that existing copyright. But if the code library is modified, then the copyright in those modifications may be be deemed to be owned by the customer. It is easy to imagine how this could potentially result in complex, tangled situations.

Unfortunately, it is not possible to retrospectively “claw back” ownership of code that has been developed for and owned by another customer. In this situation, the developer would need to get the customer to assign (e.g. transfer ownership of) the copyright to them. If the customer wants to own the copyright of code they have paid for – which is perfectly reasonable – then the developer may want to either separate out and retain ownership of common, generic code that they intend to reuse on other projects, or license back such code.

In any case, it is much easier to deal with these issues at the outset of the project, in the form of a clear, easily understood contract.