Papers in medical journals are often difficult to understand and tedious to read. An editor's first loyalty should be to readers, by prioritising readability over merely producing a repository of data for the scientific community generally. The web now provides infinite repository space so there is even less excuse for journals to be unreadable. I give examples of how I attempted to improve one journal, despite external pressures and regardless of how it might affect the Impact Factor. As a postscript I outline increasing involvement in promoting honesty and integrity in publishing through the auspices of the Committee on Publication Ethics (COPE).