I think a lack of communication is the main problem. Lack of internal testing is the other. And they are related, when we find bugs they should make tests that detect those bugs and run them. I am a software developer and tester and I have been on more than dozen professional software teams as a contractor, and the teams that communicated with stakeholders and put real effort into testing got the best results and got them faster than teams that presume they know what they are or don't test constantly. From many players perspectives CIG went off hidden away in some coding cloister and came back with this large mostly done thing. That new big thing doesn't do what what many players. many customers, want. This big change makes their testing hard because such large changes break all the tests. They should iterate on smaller changes and understand what in the tests will break and dedicated testers and test automators and SDETs (Software Development Engineers in Test) should help maintain that. Superficially, this sounds like a lot of work, and it is; but is less work than the alternatives. Alternatives such as constantly fixing the same bugs many times, accidentally causing new bugs that tests might detect, and filtering Issues from the issue council because 1 of X causes of a bug was fixed. Because they didn't have basic test scripts that walk a simulated PC into the hangar and retrieve a cutlass black and fly somewhere the cutlass black didn't fly for some people in 2q3.23.0, this is amateur hour levels of mistakes. This is too common, something similar happened to the san tok yai or khartu al (however that is spelled) a few patches ago too. Every ship should have such a script, and it should exercise the most commonly bugged features of that ship. This is one place where they did listen to feedback, this was fixed immediately in the .1 patch. They had no feedback on ship UI from colorblind people (CIG has hundreds of employees, by numbers some are colorblind, where was their feedback). They ignored the limited feedback they got on MM. They didn't give more options to try MM and the old flight model side by side. How big was the test for the new starmap? And did it work for those people? They make these big changes ignoring feedback then are surprised when there is resistance. Why do some missions simply not work? Why are we all still a little concerned about dying in elevators? They have had success with small iterations why not more of that? Get feedback on parts instead of just forcing huge changes. Everything and be put out in parts. This would also let them check and see if they broke something else. Again, I propose automated tests scripts. For every mission they should have a script that spins up an image of the game servers and client (docker and other containers/VMs are simple and miraculous) have a bot go through pre-recorded motions and preset the random elements so it behaves identically each run except for the intentional changes the team made. Then if the mission doesn't complete either the test script broke and needs an update of they found a bug. A cool thing about this is that this doesn't need to be done in real time, most of these things can be done without graphics and can fast as fast CPUs allow! Run these script for every change and do a battery of hundreds for change and tell the developer what they broke on every change in minutes. Why are we still falling through planets and why are ships still disappearing or exploding in hangars in a game where these are foundational behaviors everything else depends on? Because they aren't listening to players upset about this and clearly haven't built tests that walks a simulated player around places where these were reported. We are falling through the ground because of bad communications and bad testing.