I'm sorry that you feel this way. We're more than happy to discuss things with you - on a personal note, this is my first day in the office for 10 days so I have a lot to catch up on.
I'm also sorry if it's not entirely clear what's going on - we do try to make things clear when we post and when we interact with the community, and much of the time when specifying a solution we do try and propose the spec. for further feedback. I'm not aware of any discussions being deliberately deleted and when I have tried to start discussions on the forums, I've had little response to my questions which has been very frustrating on my part.
I am aware that Martin Dougiamas has changed the way in which we present the roadmap - this is something out of my control, but perhaps Martin can comment in the forums on these changes to the roadmap.
With the current forum work that I've been doing, we've been working through MDL-39707 trying to specify the different components. In doing so, some parts have been prioritised (i.e. by Martin, or as a result of discussion within HQ, or groups such as the Accessibility group). Sometimes as a part of the investigation into an issue, it becomes necessary to try different approaches to fully consider the ramifications of that solution, and in doing so we actually solve the problem. That was certainly the case with per-discussion subscription. Because of the nature and variety of the issues in MDL-39707, and the fact that some of the issues are poorly defined or incomplete, it's quite a daunting task and difficult to get right.
In the case of this specific issue, it has been prioritised as an alternative way of interacting with forums for users with certain visual disabilities. We also intend to improve the UI for those users, so please don't think that we've forgotten about that.
It was felt that e-mail would provide a good alternative though and many people use e-mail all the time and are familiar with it. Personally, I agree - I don't think that e-mail should be used to replace the forums for a majority of users, but it can make a good complementary solution for many users.
The changes to the User Interface are much harder to get right as there are so many different views on how it could and should work. It is a much more subjective area which is fraught with challenges and difficulties. I feel that this is already demonstrated by the fact that we already have two solutions in Advanced Forums, and ForumNG. There has already been anecdotal suggestion that these don't quite meet the mark still even if they are an improvement.
Conversely, reply-by-mail is something which is far less subjective in terms of design and will bring a usable change to the same subset of users. I realise that this is a strange way of looking at this change, but if we can provide additional and complementary options, we can be more inclusive.
Although it sometimes looks like we aren't actually dealing with the challenges of the forum, there is a lot of ongoing background work.
One of the challenges we've faced is that the forum code has not been updated much for some time and many parts of it are very complex and steeped in long history. Part of the process has been that of deprecation of old code (I think I've deprecated over 700 lines of PHP code which was previously unused but not fully deprecated). We've also moved a lot of code to use the new class loading architecture to both improve performance, and to make the existing code much more manageable. With these changes it's gradually becoming easier to work with the forum code, and we're also getting a great number of people people more familiar with the code and it's various quirks. This knowledge sharing has been invaluable, both in term of idea generation, and in terms of preserving my sanity - With this it becomes easier to implement in the other features that have been requested and proposed.
With regards the other components from ForumNG and Advanced Forums, we're keen to bring as many of those over as possible but some of them conflict with one another, or are incomplete in their descriptions. There is also some concern that some are not pedagogically sound (e.g. sticky posting) - these are features of standard forums, but not necessarily appropriate within an educational setting.
As an example, both of those solutions provide a feature for anonymity but neither solves the problem well in my opinion. They're good solutions which are implemented entirely differently and serve entirely different purposes. They are both targetted at Higher Education, and are fairly mutually exclusive and do not solve things for other user groups. That's something I would like to solve more widely within Moodle. Rather than implementing the same system in five modules which all require some form of anonymity, it would be better to solve them in one place and make it reusable.
I do agree that archival is a priority, and it's one I'd like to see happen. I also agree that an inline editor will be a great addition and I don't believe it's a great deal of work but it does require some work in the background first if we're to achieve it in the best way. Again, I feel that we should be able to insert a true editor rather than show a textarea - This requires us to have an editor manager so that we can insert an editor whenever we require (I've begun work on the Atto version of this in MDL-45787 but I need to change my approach). This will also require us to work out the best route for inserting files, and again I'm not convinced that the approach by Advanced Forums is ideal here - we can do better but it's new ground for Moodle and it requires preparation and thought.
I'm not sure what the issue with image handling is - perhaps you can detail further?