I had a lot of trouble with engineers, because their whole background is learning from a functional point of view, and then learning how to perform that function.

When you look back on a historical period of music, it seems so obvious to you what the characteristics of it are, but they're not obvious at the time. So, when I look back at my own work, I could easily write a very convincing sort of account of it that made it look like I had planned it all out from day one and that this led logically to that and then I did this and then that followed quite naturally from that. But that's not how it felt.

I got interested in the idea of music that could make itself, in a sense, in the mid 1960s really, when I first heard composers like Terry Riley, and when I first started playing with tape recorders.

Once you've grown to accept something and it becomes part of the system you've inherited, you don't even notice it any longer.

If you think of the way a composer or say a pop arranger works - he has an idea and he writes it down, so there's one transmission loss. Then he gives the score to a group of musicians who interpret that, so there's another transmission loss. So he's involved with three information losses. Whereas what I nearly always do is work directly to the sound if it doesn't sound right. So there's a continuous loop going on.

I often say to people that producing is the best paid form of cowardice. When you produce things you almost always get credit, if it's a good record, but you hardly ever get the blame if it's not! You don't really take responsibility for your work.

It infuriates me that stuff from the Internet routinely doesn't include all the credits. Because as soon as I listen to something, if I like it, I want to know, "Who's the bass player?" "Who did that?" "Who's the engineer on this?