Total ozone measurements using the Dobson spectrophotometer 56 (D56) have been performed on a regular basis in Oslo (59.9¿N, 10.7¿E), Norway, from 1978 to 1998. The data set for the entire period of observations has been critically examined. In this paper we present the results of the homogenization process and discuss the data quality of the record. Much effort has been made to optimize the zenith sky charts used for CC' zenith observations. The charts have been constructed using a semiempirical method, where a large number of quasi-simultaneous observations of direct Sun and zenith observations have been used in combination with a radiative transfer model. The same radiative transfer model has been used for constructing AD and CD zenith sky charts for various optical thicknesses of cloud cover. When the D56 ozone observations started up in 1978 the SO2 level in Oslo was significant. High concentrations of SO2 will generally influence the measurements of Dobson total ozone and might give fictitious ozone trends. Without the SO2 corrections the revised Oslo Dobson data shows a year-round ozone decrease of -5.22 ¿ 0.63% per decade from 1978 to 1998, while after SO2 corrections the decrease is reduced to -4.68 ¿ 0.63% per decade. The Dobson data from Oslo are compared to ozone data from Brewer spectrometer 42 and to Total Ozone Mapping Spectrometer (TOMS) daily overpass data from satellites Nimbus 7, Meteor 3, and Earth Probe. The mean and standard deviation of the monthly percentage difference between Dobson and TOMS amount to -0.91 ¿ 1.41%.