This is getting silly - most of my posts these days start out "A year ago, I said..."
Anyway, getting on for a year ago, I said:
The police have moved quite quickly - it emerged in September 2001 that some of the US hijackers were already under investigation, but nothing concrete had come up, so they were being left alone. As soon as they struck, the authorities were able to quickly track down their backgrounds. The same may well be the true here, and no blame would neccessarily attach to the police or security services - it could be very difficult to make the jump from vague suspicion to grounds for arrest.
On one hand, it seems that rather more information was available to the FBI in 2001 than we knew of.
On the other, my speculations about the July bombers have turned out to be accurate.
I find the news encouraging; it is somewhat reassuring to think that successful attackers slipped through our defenses rather than that we are wide open.
How do we prevent others slipping through? That's not obvious, but one thing we can say with confidence is that wider information-gathering is not what we need. Terrorist plots are not being carried out without the security services getting any sniff of them. What they need might be better management, it might be more resources, or it might be easier access to specific kinds of information on suspected individuals - what might be termed "deeper" intelligence-gathering. (Or, of course, some combination).
Even illiberal measures such as ID cards have to be considered in this context. If it were easier or cheaper to "drill down" from having suspicions of one individual to getting enough information to act, then that might have happened in July 2005.
However, it's hard to see what could have triggered action. What arouses suspicion seems to be contacts with known terrorist sympathisers. What would trigger action? Probably only getting hold of actual plans or materials for an attack - and those are hard to get. You can't follow every suspect around with armed police, (which is a good thing) - if the amount of preparation for the attack is kept to the minimum, it's very difficult to prevent it.
That's not a reason for despair. The sort of attacks that can be carried out with little preparation of the sort that risks exposure are the ones we've already seen - kitchen-made explosives or poisons, small arms, sabotage. These are not a threat to our society, and not a significant threat to us as individuals when compared to the non-terrorist risks we run every day.
The IRA was able to to mount much more serious operations - mortar attacks on Downing Street, truck-size bombs, sustained campaigns. I feel in less danger from terrorism now than at any time since I moved to London in 1989.
If counterterrorism is basically working well, what are the lessons? More of the same, I think. Gradual increases in resources for intelligence and policing, and for analysis of intelligence - nothing that will disrupt an organisational structure that is functioning adequately; nothing that will carry too much cost or cause too much disturbance. No overreaction to the fact that small-scale attacks will continue to slip through the net.