I've talked about the issue of control. If people were honest, that's one of the reasons why people are nervous about having a readers' editor, because of the loss of control.
On the question of need, I don't think actually -- until you have a readers' editor, you can't really tell whether you need one. I heard some editors saying, "We only publish X corrections a week", but it's only by having an independent system and encouraging people to tell you -- so I'm pretty sure that if the Guardian makes a mistake, that we're going to know about it because we are appealing for people to identify them. In terms of spotting the systemic weaknesses within the paper or of individuals, it's difficult, I think, for anybody to say they don't need that if they don't have it.
In terms of cost, I've also heard my colleagues saying there's usually somebody on the paper who deals with these things. It might be part of the managing editor's office. So it's not a given that having this person is going to be an additional headcount, because somebody -- usually editors don't have the time to deal with this personally. There's usually people within your organisation who is dealing with this. I would say I agree with what Chris Elliott said about why it is good commercially in terms of building trust and in reducing legal cost.
In terms of the much smaller papers -- I was trying to think if there was a rule of thumb. We have two readers' editors per a headcount of about 600. I would have thought that any paper with a staff of over 100, to have one person who is doing this would not be an excessive use of resource. In the case of regional chains, I would have thought a group like Trinity Mirror or Johnson, if they had one readers' editor who dealt with five or six or maybe even 15 titles, that would be a way of getting around the business of having one per title.