We read the article by Fischer et al. [5] with great interest. It describes the case of a patient with an elbow fracture and luxation -treated with a fixator equipped with a limiter -who underwent a rehabilitation program. Recovery was monitored using an accelerometer-based application for smartphone measuring joint goniometry and motion frequency. Results showed a progressive increase in joint goniometry values but not in motion frequency. The goniometric measurement reliability was assessed correlating the results with those obtained by manual measurements made on photographs. In the week before removing the fixator, goniometric measurements were not reliable, even if "it is plausible to assume that the patient [had] reached the maximum available range allowed by the limiter." In the discussion, it is written also that "when the patient can exhaust the full motion range without effort, the movements can become so fast that no reliable angle measurement is possible."Accelerometer-based applications are one of two main kinds of software available for smartphones to measure joint angles [3]. The other kind of applications is photographic-based, which makes the measurement by positioning a virtual goniometer, visible on the smartphone screen, on a photograph obtained via the smartphone camera. The procedure is similar to that used by the authors to validate their goniometric data, i.e., manual measurements made on photographs. The main advantages of the photographic-based applications are that it is not necessary to print the images to get the measurements and that the data can be easily stored in an electronic database [4]. In 2011, our group published a paper on the reliability of a photographic-based application -DrGoniometer (CDM, Milan, Italy) -in the angle measurement of elbow [2].Both kinds of applications can be used to assess joint motion in dynamic condition, for example, during therapeutic exercises or sport activities. With photographic-based applications, in fact, it is possible to select the appropriate time-frame from a video sequence. Anyhow, to date, no goniometric application for smartphone has ever been validated in dynamic condition.The paper by Fischer et al. [5] does not explain in detail the assessment procedure or, in particular, how the accuracy of smartphone measurement was validated. The accelerometer-based measurements seem taken only in dynamic conditions, while the photographic-based measurements were obtained in static circumstances. We would appreciate if the authors could: i) confirm (explaining why) they assessed the elbow range of motion by the smartphone only in