Moving in formation in Charente, France

On August 27, 2017 at around 10:30 PM, two witnesses in the middle of nature observe the starry sky. They are intrigued by the jerky movements of luminous phenomena in a restricted part of the sky. Two testimonies are collected. The observation shows very little strangeness: - When you stare intensely at a star (W1 says "I started looking at a star" W2 says "W1 is fixated on a star"), it is classic to have the impression that it moves, this is the autokinetic phenomenon. Moreover, the method of detecting movement ("I was taking reference points in the sky, the clouds, or with my finger without moving it") is unreliable, especially in the presence of changing wind and therefore irregular cloud movement (the Angoulême weather station records a 90° change between 10:00 PM and 11:00 PM (see on Infoclimat). - That a star is accompanied by two other less bright stars in a triangle position is nothing strange provided that the observation conditions are good and allow distinguishing faint stars as well. The autokinetic effect is not necessarily the same for each of the stars because it can depend on their intensity. - The observation lasts 40 minutes and the UAP moves from NW high elevation to N0 elevation 15° lower, which corresponds to the movement of the stars (and rules out a satellite hypothesis). The fact that W2 confirms W1's perceptions regarding the movement in no way objectifies this movement. The autokinetic effect gives a perception of movement but whose direction and intensity are hardly quantifiable and easily influenced by the perception of others. As it stands, to counter the autokinetic hypothesis, a photo recording, or two independent, precise, and timed measurements showing a concordance would be needed. And even in that case, the "clouds" origin of these movements would remain the most likely hypothesis given the detection method used. One may wonder whether the prior knowledge of a seemingly much stranger testimony in the same place and also involving a triangle has not influenced W1's perception of strangeness. Finally, the testimony is of average consistency insofar as what constitutes the strangeness (the jerky movement, the triangle formation) is imprecise in geometry (no drawing or values of movement and angular separations) and sequencing (we only know that it occurs within a 40-minute range). Consequently, GEIPAN classifies the case as A: observation of stars.