This file describes the changes to the software actually being implemented and not yet included in the public distribution. It also decsribes the known problems and bugs. (Updated 3 June 2011). 1) The problem of non gravitational perturbations is still unsolved in the stable version of the code. The merge of the non-gravitational aware version with the stable version is due, but we do not have a date yet. 2) The full error model with correlations, according to the formalism of Carpino et al. (2003), but with the new model and coefficients computed by Baer et al. 2011, is not yet implemented in 4.2, although most of the code is there, where it is not possible to reach it. 3) There are too many developments which are not yet included in the distribution, contrary to our promise that stable improvements would become available. The problem is, a lot of work would be needed for this, as well as solving some complications with different licenses, including "scienceware" type licenses. KNOWN PROBLEMS 1) Documentation: the project of a "Mathematical Specification" document has been abandoned, to be replaced by Milani and Gronchi, "Orbit Determination", Cambridge University Press 2010, 392 pages. 2) References. A serious work in providing references could to a large extent replace the ad hoc documentation. Just our own papers are dozens. The problem is how to link the subrotuines, and the algorithms used in them, to the references. Possible to use some BiBTeX machinery? 2) Cometary orbits. The module ever_pitkin is now quite stable, even if the "universal" variables are NOT universal (unless we have a computer with infinite mantissa). Still observing a Sun-grazing comet while very far from the Sun is not possible numerically, for our good luck it is also impossible astronomically.