Platform for Internet Content Selection

I wrote this tretise on PICS about six months ago. Back then most of us were busy protesting against the CDA and deluding ourselves that Australian governments would learn from the minefield of mistakes US legislators made and refuse to repeat such a debacle here in Australia (that's most of us: See here for a notable exception).

During the fight against the CDA, civil liberties groups argued that it was not constitutional for the US Government to regulate Internet speech because the law was not the least restrictive means of achieving the Government's compelling interest of protecting children from pornography. Plaintiffs argued that the existance of blocking software meant that individuals would be free to control their own exposure to content.

With that argument in mind, it made sense to discuss and promote PICS, the Platform for Internet Content Selection. So I did -- I published a description of the system to inform South Australian ISPs of its properties.

Therein lies a danger, however: In making the argument that the Government could have protected children by encouraging the use of filtering software and technologies like PICS, plaintiffs created a fallback position for the US Government in the event that the CDA was eventually defeated. Put simply, the Government could mandate that all Internet content should carry a PICS label containing a content-based rating to ensure that blocking software would be effective. In essence, they did propose a simpler method of doing this in their defense to the ACLU's legal challenge: The -L18 tag.

The -L18 tag was a content labelling method that appeared to have been thought up on the spur of the moment by the Government. They argued that a sign of intent to "protect" children from exposure to offensive material would be to "voluntarily" tag that material with "-L18", meaning that it was only suitable for readers over the age of 18. That intent could be converted into a defence against prosecution in the event that a legal complaint was ever raised against the presence of the material.

Quite rightly, the three-judge bench of the Philadelphia Federal District Court dismissed that solution: They concluded that it was unreasonable to force speakers to rely on a defence that wasn't feasible yet because the technology had not been deployed: there were no browsers that used the -L18 tag to filter content, no web publishers making use of the tag, and no procedures in place to determine adequately whether any particular item was unsuitable for minors.

But what if the technology was deployed? What if browsers adopted support for such a scheme? What if publishers started to rate their sites with something like -L18? What if there were procedures in place to determine the rating any given Internet resource should carry as a label?

That's what's happening now. The PICS system allows ratings to be attached to Internet content in the form of labels. There are also two major censorship systems which utilize those ratings to determine whether someone is permitted to read a document on the web. The ability to utilize those censorship systems is currently built in to Microsoft Internet Explorer, and will be built into Netscape in its next major release.

The points that made the three-judge panel conclude that -L18 was ludicrous (subsequently affirmed by the full bench of the Supreme Court) are no longer the case. We can attach ratings to web pages. We have Microsoft RSACi and SafeSurf censorship systems to choose from. There is a substantial (but probably not yet significant) number of sites using those ratings systems. And each system has formalized procedures for determining which rating a web page should attract.

This is scary: the infrastructure for censorship on a massive scale is now in place. And if the US Government chooses to mandate its use, it can use precisely the same arguments Plaintiffs used in their briefs against the CDA to justify it!

It gets more insidious, though: It's more than a little bit likely that the US Surpreme Court would still rule that such a law was unconstitutional. But that's ok, because the Government doesn't actually have to enact the law: it can merely threaten it.

Last month, President Bill Clinton and Vice-President Al Gore called a summit at the white house to discuss the "problem" of Internet content regulation. It invited prominent Internet software companies, search engine companies, and some of the same civil liberties groups who had supported content labelling in their CDA briefs. It refused to invite the ACLU, but they kicked up a fuss in the press and eventually received a late invitation.

At that meeting, the President threatened attendees with "Son-of-CDA" style legislation. He also implied that a good way to stave off those laws would be to show good faith by voluntarily (!!) censoring the Internet.

As a political move, it's devilishly clever: Firstly, while Government censorship can be appealed on first amendment grounds, censorship imposed by a private company (or coalition of private companies) can't be, even if they've implemented that censorship under threat of retribution from the Government. Also, since the companies themselves haven't been threatened with censorship (only their users, right?) they have nothing to lose by going ahead with it. And finally, nobody wanted to litigate another CDA case: If by doing this they can avoid more stupid laws, they can buy themselves an easy way out.

Almost immediately after the summit, Lycos threw down the gauntlet to other search engine companies: It challenged them to expunge sites which don't carry PICS labels from their databases. Also, Netscape promised that its next Navigator release would observe PICS labels.

What the hell is going on here?! These are the guys who WON the CDA case -- It should be the Government who's doing the compromising! But no, the companies at issue here are too gutless, and the civil liberties groups seem to be (by and large) falling over each other to "do a deal" that will keep another CDA away. It won't work, though: Once the Government has won this step, it'll move onto the next... and the next... and the next, until we all learn.

The Australian Government is insisting on the same thing: It wants all Australian ISPs to sign up to codes of conduct to avoid prosecution under obscenity laws, and those codes of conduct must contain clauses to say that the ISPs will participate in the development of a uniform system of labelling and rating for content in this country. This will be law by some time next year, and there isn't a lot we can do about it because our elected officials don't actually listen to their constituents.

So where does that leave us? Up shit creek. We have an Internet which is "infected" with a system that makes it censorship-ready. We have governments around the world who are intent on using that property for their own purposes. We're close to losing the war.

I was wrong to word my statement on PICS in a way which supported its use. Although I was very careful to word it in a way that made it clear that it'd only work in the absence of Government interference, that was a naive view. And although PICS was developed with the claim that it could be used to locate content of interest just as easily as it could be used to censor, there is a dearth of applications which support that use: the only use PICS is being applied to is the restriction and suppression of material that makes governments uncomfortable, rather than the enrichment of material a user might find valuable. In retrospect, my closing statement on the matter seems banal and trite. I have erred.

Fundamentally, PICS is censorship technology: that's the beginning and the end of it. My intent in writing this is to warn allcomers of that fact. I don't want to convince anyone that PICS is a GoodThing;. As a major part of a censorship infrastructure, it's responsible for furthering the scope of the greatest and most persistent evil mankind has ever faced: Censorship.