[en] Complete text of publication follows. The ⁶⁶Ga radioisotope is important e.g. in the high energy efficiency calibration of γ-detectors. Therefore, the precise knowledge of its half-life is crucial. In 2004 a critical review was published about the half-lives of radionuclides considered to be important for detector efficiency calibrations. It was found that the precision of the ⁶⁶Ga half-life is by far not enough for the requirements posed by the International Atomic Energy Agency. Since 2004 two new high precision half-life measurement of ⁶⁶Ga became available whose results disagree by about six standard deviations. This strong deviation indicates that the knowledge of the ⁶⁶Ga half-life is still very far from the required precision, therefore, new experiments are clearly needed. In the present work the half-life of ⁶⁶Ga has been measured based on counting the γ-radiation following the β⁺ decay. Special emphasis was put to the experimental implementation of the measurements to reduce the systematic uncertainties and to increase the reliability of the measured half-life value. Six sources were produced at the cyclotron of Atomki by the ⁶⁶Zn(p,n)⁶⁶Ga and ⁶³Cu(α,n)⁶⁶Ga reactions. Evaporated Zn targets and thick Cu disks were used for these two reactions, respectively. The γ-radiation following the β⁺ decay of ⁶⁶Ga was measured with three shielded HPGe detectors. A sufficiently long waiting time was inserted between the source preparation and the beginning of the counting in order to reduce the initial dead time of the counting setups below 2 %. The reliability of the dead time values provided by the data acquisition system was checked by measuring the decay of one source in parallel with two different acquisition systems. In order to check the longterm stability of the counting systems, longlived reference sources were measured together with the ⁶⁶Ga sources. The reference isotopes were ⁵⁶Co, ⁶⁵Zn and ¹³⁷Cs. The ⁶⁶Ga half-life was determined based on the analysis of the seven strongest γ-transitions. The decay was followed for up to 87 hours (about 9 half-lives) and the spectra were recorded in every 30 minutes. The half-life was determined from the parameters of the exponential curve fitted to the peak area vs. time function. The final value was calculated as the weighted average of 37 individual half-life values (six sources with six or seven γ-transitions). The obtained half-life value is t_1/2 =(9.312±0.032) h. The quoted uncertainty include the statistical uncertainty as well as systematic uncertainties from the stability of the counting systems, dead time determination and peak integration. Further details of the experiments and the data analysis can be found in [5]. The obtained half-life value supports the validity of one of the recent measurements while it is in contradiction with the other one.