There are indeed countless examples on TMO of the cycle you describe, but it also is due to the fact that we are a small community to start with, and a community left with no device upgrade in years. This inevitably means developers are hard to keep on the long term, and even those that were very dedicated to Maemo/Meego/Sailfish may eventually stop maintaining their applications because the userbase is nearing zero, or because they need to use another OS or device for their personal use and can't be on every front at the same time.

To me it's a tad harsh and unforgiving to accuse the Neo900 to bear even the slightest responsibility for this vicious circle. The project is long overdue, and as time passes, its added value decreases (there will still be benefits in using a Neo900 though), communication has been a chronic issue, funding for final success is extremely uncertain, and the performance boost we wanted won't be ground-breaking now that we've been used to more performance. That I agree, but if we had to name what are the prime factors responsible for the erosion of the TMO community, I would certainly think about our beloved Stephen Elop, as well as the many companies that disregard anything that is not Android and therefore decide for us what one can or cannot do if not using the mainstream OS that generates the most profit per developer work hour*. The economic reasons are obvious and understandable, but the statement stands that it is not a step in the right direction for diversity beside Google and Apple, and for the long term health of the TMO community overall. The never-ending Neo900 project may contribute to losing faith among TMO members, but it's hard to support with evidence. Actually, the Neo900 is one of reasons why I keep checking TMO over the years (there are others, I'm still a Sailfish user and Maemo lover) because I will always be excited about any progress on this front even if it turns out I can't pay the final price; I will still think making this device happen will be an improvement for user privacy and a message showing that some people demand it.

Your willingness to invest money and kickstart a TMO-compliant project has been severely impacted, that is a very interesting point indeed. Many users have also invested in the Jolla Tablet, and/or its Youyota rebirth, and it's true that many have said they won't get tricked a third time. This is something I overlooked and you are right that the long Neo900 project is not helping, however I'm not totally certain that an announcement of a final and definitive failure would make those who invested feel better about future projects. At least at the moment there is still some hope, and the money was not stolen, which is kinda comforting.

* : I can't even pay parking in my 70k-inhabitant town without an Android device, and this town is on an island where free parkings do not exist. At all. How stupid is that. Then there are bank applications, some of them are mandatory to actually confirm transactions because the websites are by design meant just for monitoring while the phone is the key to interact. In the end, your OS can dictate if you can drive and park a car in your town, if you can improvise a bus trip instead (paper tickets are only available in selling points here, not in buses, only the Android application allows buying e-tickets at any time), what bank you can use, and we all have other examples of our own of applications that we want or even need, but can't use. These are some of the external pressures on the TMO community, together with the obsolescence of our devices (Neo900 tries/tried to do something here) and lack of official maintenance of Maemo/Meego and base applications (there is hope with Leste, Mer, PostmarketOS), everyday decreasing the number of users, developers and maintained applications. I doubt internal pressures like the Neo900 delays are causing significant population drops in comparison.