Like you said, I'm pretty much stuck on Boom compatibility and that most likely won't change. But there are what I think valid reasons why I chose Boom over, for example, ZDoom UDMF. I'd like to explain them so you can better understand my position :)
One reason, albeit a small reason, was to force myself to stay within the Boom-set static map limits. I really don't want to make something like this or this for 32 straight maps. And I'm pretty sure no one wants to a play a megawad full of 2-hour long maps. But like I said, that's a pretty small reason. I could [or should] be able to make small maps no matter how large the static limits are.
Another reason why I went with Boom was because I needed no more or no less features than it offered. My goals really didn't call for lots of additions or changes; there's still the core of vanilla Doom resources there which the project would rely on. Boom's dehacked format allowed me just enough wiggle room to add and change what I want without the need to escalating to a format like DECORATE. The map actions in Boom provided me with plenty of room to get creative; there's the generalized action format, voodoo dolls, MBF sky transfers...and that's all I really needed. Even with the scope of the project steadily expanding, Boom still offered all I needed.
Probably the main reason why I went with Boom, though, is because it is considered a standard format. Because of that, many source ports are either derived from Boom or are compatible with Boom features. So it is very portable, meaning it can be played in all sorts of ports - (G)ZDoom, PrBoom+, Eternity...hell, even the original Boom engine if you wanted (ok technically you'd want to play it in MBF, but it should still run in Boom). If, for example, I restricted the map set to GZDoom, it would only run in GZDoom and only people who play wads in GZDoom could play it. I wanted a wide playing audience, and Boom certainly provides the opportunity for that.
There are downsides to working in more advanced formats as well, at least for me personally. I could start on a relatively humble project using an advanced format. As I work, I keep seeing those nifty features dangle over my head, and eventually I decide to try and make use of them. That continues until all the bells and whistles I've made use of creates something way more complex and huge than I intended. And the more complex it gets, the harder it is to maintain quality throughout. This happened with a previous project of mine and it ended in disaster. That also sort of happened to me with Vela Pax, where my port of choice got support for a node format which allowed me a huge amount of breathing room for my maps. And that's how those maps became so gargantuan (I set the bar so high so now I gotta keep it up for 3 more maps UGH).
I can understand why you might want to see this being made for GZDoom and UDMF. Ultimately those more advanced formats don't automatically create a better end-product. If anything it can make it much harder for the final release to be of high quality. Boom offered everything I needed and nothing more, so that's why I went with it ;)
I totally agree on what you have say. The format boom was always an excellent choice, in particular for the competitions of demos. There was a time period where I mapped in format Zdoom, I abandoned this format, because a lot of feature, even if they are interesting, are not largely useful and are more suited for the aestheticism of the map. I also find all as it is necessary in boom format, which I need, because the important for me, it will always be the apsect of the gameplay that details superfluous which make certainly more beautiful a map, but useless rest, the atmosphere makes everything.
I hope that you are well to move forward on this new project, I remember that you have to mention the Supplice's project, one or two years ago maybe, accompanied with some screens, the result is totally different now with compared with my memories, the result are very surprising :), as most of people, I like the new pallet, and the design of the shotgun. Such a project (a megawad of 32 maps) needs to be supported well by other mappers, I am delighted to see that Jimmy contributes to the project. :)