" /> net.wars: November 2006 Archives

« October 2006 | Main | December 2006 »

November 24, 2006

The Great Firewall of Britain

We may joke about the "Great Firewall of China", but by the end of 2007 content blocking will be a fact of Internet life in the UK. In June, Vernon Coaker, Parliamentary Under-Secretary for the Home Department told Parliament, "I have recently set the UK Internet industry a target to ensure that by the end of 2007 all Internet service providers offering broadband Internet connectivity to the UK public prevent their customers from accesssing those Web sites." By "those", he means Web sites carrying pornographic images of children.

Coaker went on to say that by the end of 2006 he expects 90 percent of ISPs to have blocked "access to sites abroad", and that, "We believe that working with the industry offers us the best way forward, but we will keep that under review if it looks likely that the targets will not be met."

The two logical next questions: How? And How much?

Like a lot of places, the UK has two major kinds of broadband access: cable and DSL. DSL is predominantly provided by BT, either retail directly to customers or wholesale to smaller ISPs. Since 2004, BT's retail service is filtered by its Cleanfeed system, which last February the company reported was blocking about 35,000 attempts to access child pornography sites per day. The list of sites to block comes from the Internet Watch Foundation, and is compiled from reports submitted by the public. ISPs pay IWF £5,000 a year to be supplied with the list – insignificant to a company like BT but not necessarily to a smaller one. But the raw cost of the IWF list is insignificant compared to the cost of reengineering a network to do content blocking.

How much will it cost for the entire industry?

Malcolm Hutty, head of public affairs at Linx, says he can't even begin to come up with a number. BT, he thinks, spent something like £1 million in creating and deploying Cleanfeed – half on original research and development, half on deployment. Most of the first half of that would not now be necessary for an ISP trying to decide how to proceed, since a lot more is known now than back in 2003.

Although it might seem logical that Cleanfeed would be available to any DSL provider reselling BT's wholesale product, that's not the case.

"You can be buying all sorts of different products to be able to provide DSL service," he says. A DSL provider might simply rebrand BT's own service – or it might only be paying BT to use the line from your home to the exchange. "You have to be pretty close to the first extreme before BT Cleanfeed can work for you." So adopting Cleanfeed might mean reengineering your entire product.

In the cable business, things are a bit different. There, an operator like ntl or Telewest owns the entire network, including the fibre to each home. If you're a cable company that implemented proxy caching in the days when bandwidth was expensive and caching was fashionable, the technology you built then will make it cheap to do content blocking. According to Hutty, ntl is in this category – but its Telewest and DSL businesses are not.

So the expense to a particular operator varies for all sorts of reasons: the complexity of the network, how it was built, what technologies it's built on. This mandate, therefore, has no information behind it as to how much it might cost, or the impact it might have on an industry that other sectors of government regard as vital for Britain's economic future.

The How question is just as complicated.

Cleanfeed itself is insecure (PDF), as Cambridge researcher Richard Clayton has recently discovered. Cleanfeed was intended to improve on previous blocking technologies by being both accurate and inexpensive. However, Clayton has found that not only can the system be circumvented but it also can be used as an "oracle to efficiently locate illegal websites".

Content blocking is going to be like every other security system: it must be constantly monitored and updated as new information and attacks becomes known or are developed. You cannot, as Clayton says, "fit and forget".

The other problem in all this is the role of the IWF. It was set up in 1996 as a way for the industry to regulate itself; the meeting where it was proposed came after threats of external regulation. If all ISPs are required to implement content blocking, and all content blocking is based on the IWF's list, the IWF will have considerable power to decide what content should be blocked. So far, the IWF has done a respectable job of sticking to clearly illegal pornography involving children. But its ten years have been marked by occasional suggestions that it should broaden its remit to include hate speech and even copyright infringement. Proposals are circulating now that the organisation should become an independent regulator rather than an industry-owned self-regulator. If IWF is not accountable to the industry it regulates; if it's not governed by Parliamentary legislation; if it's not elected….then we will have handed control of the British Internet over to a small group of people with no accountability and no transparency. That sounds almost Chinese, doesn't it?

Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, at her personal blog, or by email to netwars@skeptic.demon.co.uk (but please turn off HTML).

November 17, 2006

Waiting for Gowers

So here it is November, and we are drumming our heels impatiently (to the annoyance of the new downstairs neighbours) still waiting for the results of the Gowers Review of Intellectual Property, which was supposed to report to the Chancellor, the Secretary of State for Trade and Industry, and the Secretary of State for Culture, Media, and Sport "in Autumn 2006". I'm not sure when "Autumn" officially begins or ends, but I'd go with August Bank Holiday/Labor Day to the Sunday when the clocks change.

Perhaps the delay is due to global warming.

The Gowers Review is large and complicated. One change the recording industry is lobbying for is copyright term extension for sound recordings; although copyright in sound recordings is 95 years in the US, here you only get (weep for them) 50. The Open Rights Group, on whose advisory board I email, held an event to air the matter earlier this week. It is heartening to report that the event was full of people and passion: one reason copyright has kept getting extended is that no one outside the industry seemed to care.

There are a number of things that *aren't* included in the review. Government information, for example, which had its own review in 2000. Crown Copyright and Parliamentary Copyright (it may not make sense to an American that the text of national legislation is copyright, but so it is). The Patent Office is doing its own review of trademarks and the definition of a "technical step" that's required to make something patentable – this applies in a vital way to the question of patenting software programs. But things like digital rights management, orphan works, archives' right to make preservation copies, and the problem that of perpetual copyright in unpublished work are all being considered. (Yes. A 15th century, anonymous, unpublished poem cannot legally be published or copied.)

The problem is that so many deals can still be cut in smoke-filled back rooms. The reviews' original plan seems more interested in business IP use than in consumers' rights.

We say again: all intellectual property law is a balance between rewarding artists and creators and the rights of the public to access and use their own culture. Corporations that have bought up large numbers of copyrights won't care about this, but (as I also keep saying) every creator is a net consumer of intellectual property. Every writer reads more than he writes; every musician listens to more music than he learns or composes; every filmmaker, even Woody Allen, sees far more films than he will ever make. The more restrictive – or, in Pamela Samuelson's word for it, maximalist – copyright becomes overall the less people will be able to build on the past to produce new work. And no one, no matter how much of a genius, ever creates things that are entirely new with no reference to what has gone before.

So my hope is that what's taking the extra time is that there are lots of impassioned submissions and Gowers and his team are having to consider public interests they didn't expect. And not, instead, that what's happening is behind-the-scenes dickering to skew the report against the public interest.

It's only the future of copyright in the UK.

Still: the point isn't to rush to release the report. The point is to get the report right.

What *should* happen? The Skeptics, another subculture I inhabit, have a saying with reference to the paranormal that "Extraordinary claims require extraordinary proof." Copyright has been with us for centuries, but the relentless march to extend it has vastly accelerated since the mid-1970s. I think we should class the claim that further extension is necessary as extraordinary, and we should demand commensurate proof of its need from those who are lobbying for it. Especially since the industry's major players are the same in every country; in other legal areas we do not assume that the UK must have the same laws as the US. Why should that be true in copyright?

While we're waiting, I have long thought that we need to replace the term "intellectual property". It's a bad metaphor, and calling the intangible results of the creative process "property" stacks the deck against anyone in favor of public access, because as soon as you talk about limiting the term of property rights you sound like a thief. I've been trying to come up with a term that expresses something about products of the creative process ("croducts"?) or what John Perry Barlow talks about as creatures that form in the intellectual and emotional space between two people. I haven't had very much luck. (Could we talk of a "clever"? or borrow Vannevar Bush's term for his Web-like fantasy machine,a "memex"?) "Intellectual children" is my best analogy: like children, you create and murture the products of your mind, and at some point they leave you and have to find their own way in the world. You do not, ever, own them.

A free copy of one of my books to anyone who can come up with a really good answer to this. Meantime, I'm sure Punxsutawney Phil will be along any day now, looking for his shadow.

Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, at her personal blog, or by email to netwars@skeptic.demon.co.uk (but please turn off HTML).

November 10, 2006

ICANN dreams

This week ICANN announced its three new board members for 2006 to 2009: Persistent readers of this column will know that I put my name in for the job. I'm not one of the three. They are: Robert Gaetano, Steven Goldstein, and Rajasekhar Ramaraj, and I know about them approximately what's written by Kieren McCarthy, a journalist who has spent more time than anyone documenting ICANN.

ICANN had 90 applicants for the open jobs – three for the board of directors, and four for various subgroups. I'm told that in Asian countries it would be a terrible loss of face to be on the short list and then not get chosen, and that this could be the reason the names of those on the short list have never been made public. But no statistics have been released either, so we don't even know how many made it that far. Nominating committee members are unlikely ever to divulge even that much; they were required to sign a non-disclosure agreement. I know only this: I was on the short list.

Because so little is known about the ICANN selection process, it seems worth recounting what happened. Shortly before the nominating committee's late September selection meeting in Frankfurt, I got email from the chair, George Sadowsky, asking me to supply a phone number where I could be contacted on Friday, the first day of the meeting. If they wanted to speak to me they would call that number and schedule a phone call for Saturday. I should not draw negative conclusions if they did not contact me. But they did, working around a transatlantic flight, and the phone call was scheduled.

Now, I'm a writer, and kind of a literalist with language. I also have pretty much never applied for a job, and don't work in either the corporate world or academia. Therefore, when his email said they wanted to talk to me for "clarification" I assumed they meant they wanted to ask me questions about what I'd written in my statement of interest. So I reread it. I also spent an hour or two before the phone call reading news and other items on the ICANN site. One of these was the then newly released LSE report (PDF) on the Generic Names Supporting Organization.

None of that helped, because what Sadowsky, who conducted the 20-minute call with utter silence behind him, asked me were things like, "What, in your view, is ICANN's mission?" And "What are the three areas of ICANN you most want to be active in?" The first question made me think I was taking a test; the second seemed more like a job interview, or perhaps a theatrical casting call. You know, the kind where the director and his minions are all sitting, invisible, out in the theater where you can't see them because the stage lights are blinding you. When I asked who else was sitting around the phone they wouldn't say. (They did refer me to the Web page listing the committee's members, but I wasn't sure who might or might not have made the actual meeting.)

"What," the last question went, "would you want to say you had accomplished that only you could do" if I were chosen. I said I wanted to see ICANN become a more trusted and accountable organization.

There's no point pretending otherwise: I sounded completely lame and unprepared. Hardly surprising, because I was. I did manage to suggest that one reason I didn't know more about ICANN – enough, say, to know what they meant by "three areas" (countries? subcommittees? policies? "What are the choices?" I asked, and was referred again to the Web page) – was, as the LSE report agreed, the difficulty of navigating their Web site. It, like the European Union government sites, is perfectly understandable if you are already an expert on its content, but otherwise not so much.

In this sort of endeavor I have a second problem: journalists learn to deadline and forget everything the second they send the article in. In any given year I probably write 200 articles on dozens of topics. There isn't a single one of those topics where I don't have to reread what I've written to know what I said, even if I wrote it yesterday (I also reread my own work because I trust the research).

So, yes, I should have had the sense to be better prepared, but it was all, as I say, so unexpected. I serve on other boards. None requires anything like the effort the ICANN board does; and arguably none of them is as decisive in determining what the organizations do. In all cases I joined because I was asked; I never went through a selection process like this one.

As a reject, I can't really comment intelligently on how ICANN went about making its choices. I merely tell the story here in case my experience can help another applicant, somewhere down the line, be less confused than I was.

Thanks to all those who emailed or posted messages of support, and especially to my three referees.

Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, at her personal blog, or by email to netwars@skeptic.demon.co.uk (but please turn off HTML).

November 3, 2006

Where can we go?

One of the things about being an expatriate is this: whenever there’s a problem you always think that changing countries might be the solution. So one of the games we like to play is where-can-we-go. Global warming means the Gulfstream will stop or change direction and Britain will freeze? Where can we go? Britain is awash in CCTV cameras and wants to bring in a national identity card. Where can we go?

My usual answer is New Zealand, based on no knowledge whatsoever: it just seems so far from here that anything that’s a problem here surely can’t be one there, too. And I know so little about the country that it’s easy to fantasize being left alone to roam the hills among the sheep. Yes, I know: it can’t really be like that, and I’d hate it if it were.

But apparently when it comes to privacy the answers are Germany or Canada. In a pinch, Belgium, Austria, Greece, Hungary, or Argentina. At least, that’s the situation according to Privacy International’s new human rights survey New Zealand’s score is a good bit lower, although you could improve it by avoiding employment; workplace monitoring was the one area in which it scored really badly. Of course, the raw numbers never really tell you the quality of life in those countries; and most people don’t want to play where-can-we-go. Most people, being sensible, want the place they do live, and where they have their cultural and social ties, to be better.

The problem with talking about privacy is that it's so abstract. It’s arguable, for example, that the biggest worry in many people’s lives in the US is not privacy but how to pay for health care. A friend, for example, recently had occasion to go to the emergency room for a few tests; she estimated the bill at $2,000 and her share of it at $500.

In that sense, the Information Commissioner’s new survey, A Surveillance Society, launched alongside the annual data protection conference, is more alarming, in part because it investigates the consequences of constant surveillance and the impact it has on the realities of daily life. What can be done and is being done is to create a class system of great rigidity: surveillance, it says, brings social sorting to define target markets and risky populations. The airline that knows how much you travel decides accordingly how to treat you; the health service decides how to treat you based on its assessment of your worthiness for treatment. Welfare becomes an exercise in deterring fraud rather than assuring safety. It is, the IC’s survey says, risk management rather than the original promise of universal health care. That it’s not a conspiracy, as the IC survey repeats several times, makes it almost more alarming: there is no one specific enemy to fight.

This should not be surprising to anyone who read, some years back, the software engineer Ellen Ullman’s wonderful essay collection, Close to the Machine. Everything the IC is talking about was laid out there in detail, including the exact process by which it happens. Her story concerned a database created to help ensure that people with AIDS got all the help that was available to them. Slowly, the system morphed; in her words, it “infected” its users. The fuzzy, human logic by which one person might get an extra blanket was replaced by inexorable computer rules. Then it became hostile, trying to ensure that no one got more than they were entitled to – the precise stage Britain is at right now with respect to welfare.

In another case, the fact of a system’s existence led a boss to wonder whether he could monitor his sole employee to find out what she did all day. The employee had worked for him for decades and had picked up his children from school. This is, I suppose, where we are with National Identity cards. We *can* find out what everyone does all day, so why shouldn’t we?

Of course, Britain is famous for its class system and the anti-democratic nature of it. But if there was one thing you could say for the old ways, the class differences were clearly visible on the outside. Accent and habits of speech, as George Bernard Shaw observed more than a century ago in Pygmalion, determined how you were treated, and you knew what to expect. Social sorting via surveillance is more democratic in the sense that you don’t have to have a title or the right accent to be a big spender the airlines will treat like gold dust/ But the rules are hidden and insidious; rather than open and well-understood.

So: where can we go (PDF)? A lot of people like the sound of Ireland – they speak English, it’s close, and it’s pretty. But it ranks only a tier above the UK and will always been under pressure to adopt British standards because of the common travel area. Some people like Sweden, for its longstanding commitment to social welfare. It, too, ranks low on the privacy scale. It will have to be Canada. But it’s cold, I hear you cry. Nah. Global warming. Those igloos will be melting any day now. Off to Winnipeg (DOC)!

Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, at her , or by email to netwars@skeptic.demon.co.uk (but please turn off HTML).