The debate, at the moment, swings between two poles: the law-is-the-law side argue Twitter ought be treated like any other type of publishing, thereby subject to the same libel, defamation, and other legislation that covers publishing content. The-Internet-is-outside-the-law side argues Twitter is an unmitigated free and open space, where national legislation has little businesses. It is a battle of two world views, and my sense is the law-is-the-law side is gaining the upper hand.
Of course, I'm simplifying positions to render my own appear more reasonable, but something is definitely missing: that technology sometimes changes the nature of jurisprudence: and forcing offline law to cover online spaces is disrupting legal concepts and enforceability. The law cannot act entirely at odds with social norms, and those norms are being reshaped by technology.
Take the concept of 'public', which is at the heart of much of the recent bother. Simply put: certain things are illegal in public but not in private. You can be racist in your home in front of your family, even to a small group of likeminded people in a pub. But you cannot grab hold of a public address system and start spewing race hate. Similarly, you might accuse someone of being a paedophile to a friend, but broadcasting that on television is quite a different matter.
So what is Twitter? Recent judgments suggest it is public. Journalists, such as David Aaranovitch in his 10 golden rules of Twitter, appear to be taking a similar tack and so does the government. But this is not strictly true. First, many Twitter users have their tweets as private: only viewable by their followers. Automated data mining software might be able to find them, but they are by any reasonable measure, private.
However, one of your followers might re-tweet your original missive: and suddenly your private chat becomes a public statement in a way never intended. I do not think someone else repeating what you say makes it a public statement at all - but the interface of Twitter, the way it appears on the screen - doesn't lend itself to that interpretation.
Then there is the platform itself: its speed, limited characters, gossipy nature: all lend an air of insignificant to content. People often report things they disagree with but find of interest, without endorsing it: sarcasm is a popular currency. In fact, it can even be helpful: during the 2011 summer riots, looters using Twitter to incite riots/post pictures of their loot were re-tweeted by citizen watchdogs in the hope the police might notice, and to preserve the incriminating content.
It gets even more complicated, because a number of cases are being brought against algorithm writers for the auto-complete: a former first lady of Germany is suing Google because if you type her name the word 'prostitute' comes up as a prompt. As Google argued, they merely catalogue other people's content - so take it up with them. Lord McAlpine was trending on Twitter, and many people were simply reporting that fact (thereby making it more likely to trend, incidentally).
Of course, much of Twitter is public, and is intended to be. The root of the problem is that we suddenly have 10million Twitter users subject to quite complicated publishing law that used to only apply to a small number of trained journalists with legal support. Clearly everyone at school now needs a crash course in publishing law - as they are likely to one day be subject to it. 'Twitter made me do it' is of course not a defence. But 'Twitter made it a lot easier for me to do it, it was taken out of context, and frankly, I didn't really know exactly what I was doing' ought certainly be a mitigating factor.