No label defined (Q4798587)

The paper presents the basic results in the course of the history of the subject, which spans more than 140 years from the pioneering papers of Riemann and Cantor to the latest ones. We recall the notions indicated in the title. A set \(E\) is said to be a \(U\)-set if every trigonometric series converging to \(0\) outside \(E\) vanishes identically. A set \(E\) which is not a \(U\)-set is said to be an \(M\)-set. It is well known that every countable set is a \(U\)-set, while every set of positive measure is an \(M\)-set. The subtitles are the following: 1. The prehistory: Riemann and Cantor, 2. The new start after 1900, 3. Zygmund and Bari, 4. Zygmund and Marcinkiewicz, 5. Zygmund and Salem, 6. Variations around \(U\)-sets, 7. Another variation: Zygmund's \(U(\varepsilon)\)-sets, 8. Thin sets and functional analysis, 9. The construction of thin sets, 10. Construction of thin sets -- continued, 11. The new role of set theory, 12. Several variables, References contain 56 items. This enjoyable paper is warmly recommended to everyone interested in Fourier Analysis, Functional Analysis, Measure Theory and related fields. For the entire collection see [Zbl 1007.00037].