Martin Rees

There are lots of ideas which extend the Copernican principle one step further. We went from the solar system to the galaxy to zillions of galaxies and now to realising even that isn't all there is.

Manufacturing doesn't just mean building cars and metal-bashing; it includes making pharmaceuticals and hi-tech electronics. A crucial part of the process is the research and development that allows better and greener products to come to market. Britain has traditionally had a strong science and engineering base.

I hope that by 2050 the entire solar system will have been explored and mapped by flotillas of tiny robotic craft.

In our interconnected world, novel technology could empower just one fanatic, or some weirdo with a mindset of those who now design computer viruses, to trigger some kind of disaster. Indeed, catastrophe could arise simply from technical misadventure - error rather than terror.

The images of Earth's delicate biosphere, contrasting with the sterile moonscape where the astronauts left their footsteps, have become iconic for environmentalists: these may indeed be the Apollo programme's most enduring legacy.

The extreme sophistication of modern technology - wonderful though its benefits are - is, ironically, an impediment to engaging young people with basics: with learning how things work.

Ironically, it is only when disaster strikes that the shuttle makes the headlines. Its routine flights attracted less media interest than unmanned probes to the planets or the images from the Hubble Telescope. The fate of Columbia (like that of Challenger in 1986) reminded us that space is still a hazardous environment.

The atmospheric CO2 concentration is rising - mainly due to the burning of fossil fuels. It's agreed that this build-up will, in itself, induce a long-term warming trend, superimposed on all the other complicated effects that make climate fluctuate.

Scientists surely have a special responsibility. It is their ideas that form the basis of new technology. They should not be indifferent to the fruits of their ideas. They should forgo experiments that are risky or unethical.

Some things, like the orbits of the planets, can be calculated far into the future. But that's atypical. In most contexts, there is a limit. Even the most fine-grained computation can only forecast British weather a few days ahead. There are limits to what can ever be learned about the future, however powerful computers become.

From a personal perspective, I am disappointed that we have yet to really achieve a full understanding of the origins of life on Earth. What was the spark that, billions of years ago, kickstarted the process of evolution that has brought us life as we know it today? I hope that we will get some answers to that in my lifetime.

And we should keep our minds open, or at least ajar, to concepts on the fringe of science fiction. Flaky American futurologists aren't always wrong. They remind us that a superintelligent machine is the last instrument that humans may ever design - the machine will itself take over in making further steps.