tommorris.org

Discussing software, the web, politics, sexuality and the unending supply of human stupidity.


copyright


No, BBC News, Getty aren't "lifting copyright"

On the BBC News site today, I saw this headline in the five most shared stories:

“Getty lifts copyright on 35m photos”

That’s pretty exciting.

Except they aren’t lifting the copyright on 35 million photos.

They are making it so you can add a blob of JavaScript to your website to show the images so long as you conform to a stringent set of terms and conditions.

No wonder lots of people misunderstand copyright law: journalists—who should fucking know about copyright law—conflate a revokable license to use an image in some very limited circumstances with the copyright on that image. Don’t bother explaining the gratis vs. libre distinction because that’s hard work.

If journalists can’t write and think about these issues clearly, that will saturate into the consciousness of lawmakers and judges who will make stupid and uninformed decisions about copyright. Which they aren’t doing now, obviously.


ActivePlaces, UK government data set, is probably a copyright violation

This weekend there is a hackathon in London called hack4health. One of the data providers is called Spogo, a lottery-funded “sport and fitness finder”. It tells you where to find things like sports clubs, gyms, swimming pools etc.

They provide data on their site in CSVs. The data is provided from two sources: ActivePlaces, which is basically Sport England, and a layer on top of that which Spogo own themselves. The ActivePlaces data is available under the Open Government License, and Spogo’s data is available under a very restrictive license.

Now, I thought that the ActivePlaces data would be quite useful: we could just merge it in bulk into OpenStreetMap, so OSM would have more useful data about sports venues.

But there’s an issue. On the Spogo site, they make available as a PDF a document from ActivePlaces, under the section on data quality, there is this statement:

Data is collected via Google Earth in latitude and longitude and then reconverted back into X and Y co-ordinates which are held in the database, to enable Sport England to use with OS products. We have geocoded via maps to allow for a greater degree of accuracy than compared to postcodes, which is a particular issue in rural areas where postcode boundaries tend to have larger areas. We have also had to do this because many grass pitches do not have postcodes.

What they are saying here is that the ActivePlaces data is derived from Google Earth. That is, it is derived from copyright data. ActivePlaces is basically a giant violation by Sport England of the copyright in Google Earth. This means that technically the ActivePlaces data is a derived work from Google’s copyrighted material.

There’s a reason why OpenStreetMap doesn’t do this: because derived works inherit the copyright restrictions of the work they derive from. If it was just one or two places, it might be arguable under the legal principle of de minimis, but there’s 64,493 records…

Incidentally, if they had used Bing’s satellite imagery, there wouldn’t be an issue as Bing has licensed their satellite data for reuse and it is used by the OpenStreetMap community as one in a number of tools for producing OpenStreetMap. See the OSM wiki page.


A simple but opinionated guide to open source licensing

Open source licensing is hard. You don’t want to have to deal with all that. The guide below should make it pretty easy.

Simple cases

There are two basic license families to choose from: copyleft and non-copyleft licenses. Copyleft license are those licenses that require you to share modifications you make. The Creative Commons ‘Share Alike’ licenses are copyleft licenses because derivative works you produce have to be shared under the same license as the original.

The best example of a copyleft license is the GNU General Public License. There are two versions in wide use: either the current version, version 3, or the older version 2. The two are incompatible, and are different. Richard Stallman, the author of the GPL and the founder of the free software movement, has a list of the differences and the reasons you might want to choose v3 over v2.

If you want a non-copyleft license, that is, you want a license where modifications do not have to be released as with the GPL, the absolute simplest you can possibly use is the MIT License. I encourage you to do so because simple licenses are generally preferable to more complex licenses.

For most cases, the choice of license will boil down to copyleft or non-copyleft, and either one of the two GPL licenses or the MIT license.

Special cases

But it’s not that simple. There’s sometimes cases for using other licenses.

If you are in a language community where there is a widely used license of choice, you may wish to just follow that use. If in doubt, ask that community. For instance, if you are building a plugin to work with some existing software, you probably should release it under the same license as used by other plugins.

For web applications, there is a version of the GPL called the Affero GPL which ensures that you have to abide by the spirit of the copyleft/’share alike’ principle by releasing modifications you make to versions you are offering as a web service.

It is possible to release code under multiple licenses.

You are also able to release code into the public domain (although you may wish to use CC Zero for this, as in some countries it is not possible to release material into the public domain).

Documentation and other stuff

If you have separate documentation (that is, say, manuals, handbooks, design documents, a wiki etc.) it is best to release that as either public domain/CC Zero or under either the Creative Commons Attribution or Attribution-ShareAlike license. These are maximally compatible with other freely released text.

Images and other material should be also licensed likewise.

If you release your documentation and imagery under either public domain/CC0 or Creative Commons Attribution or Attribution-ShareAlike, it becomes compatible with Wikimedia wikis like Wikibooks and Wikiversity, and other projects which may wish to host and translate material.

Things to avoid

Do not use free content licenses (that is, licenses intended for use on text, images, audio and video recordings, like the Creative Commons licenses) for code: use the equivalent software license.

Don’t make your own custom license unless absolutely necessary. (It probably isn’t.)

Don’t be a dick.


Non-pathological copyright laws now!

David Allen Green has a post on the New Statesman blog about copyright and whether a world could exist without copyright.

I’m actually very ambivalent about whether we should have a world without copyright. I can see a case for a radically reformed copyright law, much more like the one the U.S. originally had (maximum term: 28 years).

The problem with copyright isn’t that it exists. It’s that the implementation is completely fucking insane.

The simple idea of copyright is to grant a temporary monopoly on reproduction of works in order to give the author of that work the chance to profit from it. It is a right granted by society through the state. Intellectual and artistic work is ultimately something we all own in common because they are by nature non-rivalrous, non-excludable goods. That is to say that if you come to know the argument in, say, Plato’s Euthyphro, it doesn’t harm anyone who already know it. It’s not like a banana where only one person can eat the banana.

Concepts, ideas and facts may be non-rivalrous, non-excludable goods and exist in common for all. But what motivates people to contribute to that common? Why should I spend my life finding things out and sharing them with people if I’m not going to get rewarded for it? The U.S. constitution describes the motivation perfectly for a limited copyright system: “to promote the Useful Arts and Sciences”. Copyright is a property right granted to further the public good. It has a private benefit: you can make money from it. But that is just a means to an end, and that end is the benefit of the public.

But our culture through law and ideology has made people believe that the granting of intellectual property rights is still ultimately a short-term, temporary granting of a privilege. It’s seen as more like an ultimate human right like the right to a fair trial or the freedom of religion. People now treat copyright as being primarily about private benefit. If you were to design a system solely for private benefit, you could do that.

Here’s a slightly absurd example: make it so that in Britain, only the Royal Shakespeare Company can perform Shakespeare. They would be granted exclusive rights in perpetuity to perform Shakespeare. They would do fantastically out of it. They could charge through the nose, and make bucket loads, because there would be no other way of seeing Shakespeare being performed.

There is no public good being served here except in a secondary fashion (the company would be taxed and those taxes could be used to provide public services etc.). Indeed, a great deal of public harm is being done because a culture where only one theatre company are allowed to perform Shakespeare is a much worse off culture. What if someone wants to come along and do a radically different interpretation of the same play? Like, oh, set Romeo and Juliet in New York City and replace the houses of Montague and Capulet with two teenage street gangs roughly based on Italian Catholics and Jews. Or perhaps replace them with the Israelis and Palestinians (West Bank Story). Or perhaps some new interesting technology like cinema or radio happens and you want to adapt it to the new technology.

Everything has a line of derivative works behind it. It’s just sometimes they are well-hidden. As scholars like Lawrence Lessig have repeatedly pointed out, all cultural and intellectual works build on previous work. Open any scholarly non-fiction book and you’ll find reams and reams of footnotes (basically the old-fashioned version of hyperlinks). But even for fictional work, no work happens in a vacuum. Try and imagine Harry Potter without C.S. Lewis’ Narnia or J.R.R. Tolkien’s Hobbit and Lord of the Rings, or indeed the many boarding-school novels–Tom Brown’s Schooldays, Talbot Baines Reed and even more recently George Orwell’s magnificent essay “Such, Such Were The Joys” or characters like Nigel Molesworth or the St Trinian’s girls. Or indeed a whole rich history of literature on education and the British class system.

As a result of the significant erring in how we think about copyright, countries have implemented absolutely ridiculous copyright laws. This isn’t some kind of Stallmanite hippie “freetardism”. This is just bad policy. I’ll discuss a few insane things which are undeniably wrong and stupid about most current copyright law.

Panoramafreiheit

First of all, some countries do not implement panoramafreiheit, the “freedom of panorama”. The practical upshot of this is that it if you take a photo of the Mayor of London’s office (that “wonky egg” building on the South Bank), you, the photographer, own the copyright on that work. In Britain, as in Germany and a lot of other countries, you do have panoramafreiheit. But if you take a photograph of the European Parliament building in Brussels, you do not own the copyright on that photograph—the designers of the building do. That’s because Belgium does not grant panoramafreiheit.

What does this mean in reality? Plenty of public buildings cannot have photographs of them included on Wikipedia. That is dodged by the fact that under US law, one can use copyrighted images under the terms of “fair use”. But why should a site like Wikipedia have to rely on US law to use a copyrighted image of the European Parliament building. Isn’t that quite stupid?

Sure, if governments feel the need to allow copyrights for the design of buildings, there may be a good public policy reason for that: the designer of a building probably has a good reason to be pissed off—indeed, financially harmed—if someone were to make a copy of a building they’ve made. If suddenly a building of exactly the same design as the European Parliament building were to pop up somewhere else in the world without the permission of the designers of the European Parliament building, for example, that might be something copyright law might want to legislate. But why, to prevent that, do we have to make it so that the architects have complete ownership of every depiction of every representation of that building?

Let me put this in stark terms: I have, on my computer, a number of images that I took in 2008 of the outside of the Pompidou Centre in Paris. France does not have panoramafreiheit, so under French law, I, a British citizen, will not own the copyright in images I took with my own camera in 2008 until… god knows when. Probably 2047 at the earliest. I’ll be in my sixties when that happens!

What if I want to publish a scholarly monograph on modern architecture? Why should I have to negotiate with the people who designed a building back in 1977 to use a photograph I took? Isn’t that completely insane?

And where does the restrictions on panoramafreiheit end? What if instead of a photograph, I sat outside the Pompidou Centre and did a pencil sketch of the building? I wouldn’t own the copyright on that either. How about if I wrote a detailed prose description of the building? Would I own that? Who knows? What about if I wrote a poem which alluded to aspects of the building’s architecture? The only way to truly know would be an expensive legal case in France.

Uruguay Round Agreement Act

Here’s another story of how insane copyright laws affect ordinary people (i.e. me).

I’ve got a book on my desk that was published in 1963 called Honest to God. It’s written by John A. T. Robinson, who used to be the Bishop of Woolwich. It’s interesting as a piece of intellectual history: it’s a theological book where Robinson describes a radically different idea of God than that of traditional Christian doctrine. And it was quite influential. It was one part of what has come to be known as death of God theology. At it’s most radical, it’s the influence of Nietzschean philosophy into liberal Christianity.

Yes, yes, unless you are a philosophy or theology nerd, that’s not very interesting. Fortunately, I am, so it is.

Robinson died in 1983. Under British law, this book will not enter the public domain until 2053, 90 years after initial publication. That’s pretty nuts.

But, there may be some hope for legally releasing it in the public domain under US copyright law. Under US law, all works before 1923 are public domain. That’s easy enough to understand. But for works published between 1923 and 1963, there are different rules.

As far as I can tell, this book was published in England in 1963 but not published in the U.S. I cannot find any evidence that it was ever registered for a U.S. copyright: I’ve looked in the relevant databases to see if a renewal was ever filed with the U.S. copyright office.

Not all the U.S. copyright databases are online, and the only way to really know is to go to the Library of Congress and search the paper files–you can pay someone to do it, but it costs a few hundred dollars to do so.

The situation is ridiculously complex. But then in 1996 along comes the Uruguay Round Agreements Act which grants retroactive copyright to works that had lost their U.S. copyright due to not satisfying U.S. copyright formalities over registration.

I’m now pretty much resigned to the fact that this work of cultural and historical interest will probably not enter the public domain until 2053 (that is, unless another insane term extension gets rammed through by the music industry lobby—because Paul McCartney being able to continue making money is far more important than allowing the public domain to grow).

One other avenue I’m looking into is whether or not works published by serving Church of England bishops count as works by the government. It may not help in this situation, but theological and religious texts are kind of a strange matter. The Vatican assert copyright over statements made by the Pope. Under U.S. law, if the Vatican really is an independent state, as they claim to be but some (like Geoffrey Robinson) dispute, that would make statements by the Pope basically government publications, which are considered public domain works by U.S. copyright law.

Seeing how far that could stretch is an interesting matter: obviously, under U.S. law, something like Hansard is considered public domain as it is a direct publication of the government. Would publications from the Church of England be considered publications of the government, as it is the established church? What about publicly funded universities? How far does this rule stretch? Because… if people wanted to make public data available and hackable and all that stuff, they could use that data under U.S. public domain laws by basically starting a non-profit in the U.S. and running servers there. Things like MySociety could be done legitimately in the U.S. rather than having to do dodgy things like infringe on Crown Copyright in Britain.

Ignorance of the law is…

…no excuse, right?

You have to abide by the law, even if you don’t know what the law is.

But what if it were impossible to find out what the law is? Then you’d be in Britain.

Under English law, there are a few ways that things can become law. The most obvious is to pass an Act of Parliament. You know the drill: Green Paper, White Paper, committees, three votes in the Commons, then up to the Lords, and it all passes, it gets Royal Assent and bam, you’ve got law. That’s easy enough to understand.

There’s so-called secondary legislation. You pass a law which includes a whole bunch of sections which say that someone in the government can activate a law at a particular time under certain conditions. Or you pass legislation which allows certain people to pass regulations that deal with certain things. A bit more complex, but still, you can work that one out. Devolution is a great example: the government passes a law which enables, say, the Scottish Parliament to pass their own legislation.

Then there’s European law. The relevant EU bodies pass a law and then the British government has to implement it. For programmers, this is best understood as being like Java interfaces. The EU pass a law defining a particular set of things the government has to do, and then the UK have to enact a law that implements the interface. The same is true with commitments made under the European Convention on Human Rights.

All of these sources of law can be discovered relatively easily. You can read legislation here, you can look up European law on EUROPA.

But then there’s interpretation of the law by the courts. And huge quantities of this is not viewable by the general public. Let me repeat that: you are bound by laws you cannot actually practically read.

Here’s how it works. Some important legal case comes to the Court of Appeal or the Supreme Court (formerly the House of Lords Judicial Committee). The judges go off in a huddle and decide how to interpret the law, and then come into a court and give the decision. These are long and complex documents. But because of the formalism of the court system, they are read to the court–not necessarily actually read to the court in person, but they may as well be. Then they are taken down in shorthand by court-approved shorthandwriters, who are basically owned by a commercial entity, WordWave International Ltd., which is a divison of a company called Merrill Legal Solutions.

Those are then put on to commercial legal databases like Westlaw and Lexis-Nexis which lawyers, law schools and others pay to get access to.

Let me repeat this. The legal obligations that you have as a citizen of the United Kingdom are the property of commercial entities. We pay for judges and courts to interpret the law. Then the only way to access the decisions we pay for is to buy access to commercial databases to read them? This situation is slowly getting better: through sites like BAILII, you can get access to read a lot more cases than ever before.

Call me slightly naïve and idealistic, but compare and contrast:

System A: 1. Judge writes decision. 2. Judge reads out decision. 3. Shorthand writer from writes down decision. 4. Law report publishes decision. 5. Commercial legal database operator adds decision to database. 6. Users purchase access to commercial legal database to read decision.

System B: 1. Judge writes decision. 2. Judge reads out decision. 3. Promptly thereafter, the court service publish the written copy of judge’s decision on the web, preferably without any copyright.

I am but a humble keyboard-tapper, but System B certainly seems a much saner solution here.

Why do we have System A? Because stupid bloody copyright laws enable the blessed and saintly “rightsholders” (as they are always referred to) to take even the laws we have to abide by and turn them into intellectual property.

The absurdity of this beggars belief. Thanks to technology, every piece of law could have some kind of, I dunno, unique resource identifier. You could put it in some useful, readable format like, oh, HyperText Markup Language. And then people could read the law, read the interpretation of the law as given by judges, and then link directly to the laws in question.

But we don’t have this because of… copyright.

The result

The practical result of these insanely complicated and broken laws is actually a disempowering of ordinary citizens from participation in their own shared intellectual and artistic culture. Why try and build things like Wikipedia or They Work For You or any of the other countless good things we’ve seen flourish online by basically ignoring, hacking or ‘copylefting’ when the government’s reaction is to tighten it up, all to keep certain established entertainment businesses happy? Basically, we’re doing the equivalent of preventing the Enlightenment from happening in order to keep the court jester in work.

People shouldn’t have to hack copyright law. If they are doing that, whether it is through coming up with ingenious copyleft licenses like the GPL, GFDL and Creative Commons, or through hopping around between jurisdictions to use obscure bits of the legal code, or just through plain disobedience to bad laws, it shows something is very wrong.

I object to certain aspects of intellectual property law on an academic and ideological level. But this isn’t either of those things. Things like panoramafreiheit and URAA are just a bloody nuisance. The concept of intellectual property is a ravenous beast, and to satisfy it’s ever more taxing demands, governments have passed laws that are broken in practice for everyone except the tiny elite who profits from the system being broken.

Sadly, I don’t think political engagement fixes them. All we can do is keep on building the alternative: whether that’s millions of Wikipedia articles, whether that’s 10 million educational photographs that you can freely reuse, or whether that’s open source software. Some laws make it much more difficult for us to build that alternative, and we work as hard as possible to work around them. But we need to keep pointing out how silly they are.

When the recording industry or whoever get all huffy about filesharers, when they spend millions of dollars closing the gate after the horse has already bolted (Napster, Grokster, LimeWire), remember this: if you take a photograph of the Eiffel Tower at night, the copyright in that image doesn’t belong to you, it belongs to the people who designed the lighting. So long as absurdity is more profitable for the elite than sanity, the laws will continue to be absurd.


Catholicism and copyright

One of the most amusing things about Scientology–beyond the fact that it was started by a science fiction author who frequently misrepresented his own biography–is the fact that the scriptures of the church are copyright and some are kept very secret. The business model is simple: you have to pay to read more. In fact, they’ve even said that if you haven’t had the preparation to read the advanced scriptures, your head will explode.

This is all good fun, right? The Bible isn’t copyright. The Qu’ran isn’t copyright. If you want to publish your own version of a huge range of religious texts, you can. Pop over to Wikisource and you can read copyright-free editions of the Bible, prayers, the Apocrypha and the Tao Te Ching among many thousands of other religious texts (and why not some atheist/humanist manifestos too?). This enables scholarship: theologians, historians and others can make their own commentaries building atop these scriptures. Critical scholarship of the sort Biblical commentators do is helped by not having the threat of a lawsuit hanging over one if one quotes a bit too much from the text.

What makes the Scientology situation so egregious is that no independent theological, philosophical or critical reflection can happen when the text is locked away. There seems to me to be a conflict here. If you believe you have access to a truth that has the ability to save people in the afterlife or to dramatically make their life better in this one, you have some kind of duty to share it. Or rather, if you are keeping your religious truths to yourself and not sharing them, people have very good reason to believe you might be a huckster. If the Four Freedoms of the sort the copyleft and Free Software movement believe in should apply to any work of human culture, religious works would be a prime example of exactly what it should apply to.

The fact that Scientology fails so dramatically to adhere to that is cause for concern and criticism.

But I found out today that Scientology is not alone in locking up their teachings behind the wall of copyright. The Catholic Church does too. All of the copyright in the papal writings of Pope Benedict XVI now belong to the Vatican publishing house, Libreria Editrice Vaticana.

The writings of the Pope will not go out of copyright until 70 years after his death. The earliest the Pope’s writings will come out of copyright is 2081 (if he dies this year).

What benefit is this to anyone? Did the lack of copyright protection for writings of Popes before the current copyright regime prevent the spread of Catholicism? If everything the Pope wrote was in public domain, would it prevent the development of the “useful Arts and Sciences”, as the U.S. Constitution puts it? The motivation of the Pope is really not the same as the motivation of the Walt Disney company. Without copyright protection, the Church will not fall to bits.

Indeed, one interesting question is what the copyright status of the Catholic Catechism is. This is the basic doctrine of the Catholic faith. I would presume it is copyright in much the same way. If we criticise Scientology for locking it’s scriptures up behind copyright, surely the same could be said for the Catechism?

For a body like the Catholic Church, it would seem totally reasonable and straight-forward to simply release all their materials completely as public domain.


I haven’t really kept up on the copyright wars much, but I’m amused to hear that there exists a company called Perfect 10 which has the dubious honour of having to pretend to be a pornography publisher in order to mask the fact that they are in fact a ‘copyright troll’. Apparently, producing pornography is now more respectable than the fucked up copyright laws.

There’s a joke about Oracle lurking somewhere in here, you know…


Question: CC license compatibility between versions?

Here is a nice dull little legal question.

I want to take a bunch of content that is currently licensed under the Creative Commons Attribution-ShareAlike 2.5 license (the so-called “cc-wiki” license) and produce a derivative work where I combine it with a work that is released under the Creative Commons Attribution-ShareAlike 3.0 license, and I would like to publish it on a site where all contributions have to be made under the BY-SA 3.0 license.

Can I do this?

A meta question to go with it: isn’t me needing to ask this kind of question before producing a piece of work exactly what Creative Commons came into existence to avoid? Shouldn’t this kind of thing be documented on the Creative Commons site? I reckon a website is required which contains absolutely everything you need to know about all forms of licensing: for software and for culture, guiding you through how to pick a license and what the full implications of a license is and so on.


Speaking of crap newspapers, The Independent reckons that copyright law doesn't apply to Flickr photographs. Which is an interesting take on "all rights reserved".