= Game doesn't die.
I hate to sound like an elitist because I'm happy to see new players learning the game and playing but the game doesn't need new players to survive and it doesn't need to become more user friendly to gain new players. There is literally nothing stopping anyone that wants to learn to play from learning to play and playing. Its free. There's tons of help and tutorials everywhere. And moreover with the current way things are (even in several year long droughts of new content) there is actually just no stoppage of fun in the game even for veteran players. This game will never die from lack of new players.
I've just seen so many games destroyed in the effort to gain a larger/newer playerbase that its become a surefire early warning system for me. The only good new player is the new player that comes to the game. Nothing good can come from new players that expect a game to come to them. Part of the reason why EVE Online is still trucking along (even though they are always trying to attract new players) is because CCP hasn't folded in and alienated their own userbase for attracting a wider audience. I doubt the UI improvements your suggesting are quite what I'm referring to in terms of building design towards attracting new players. However the current trend in large scale games and some indie games is that the game should try to attract every player it possibly can. I think they lose sight of just trying to make a good, focused game, sometimes people just don't want to play the game your making and that's totally fine.
Another genre that "suffers" from this is fighting games. However as much as people playing fighting games want new competition to join the community none of them are willing to sacrifice what makes competitive fighting games what they are. If you don't have the drive to legitimately practice, advance, and overcome the people that were destroying you before then you shouldn't be playing competitive fighting games, because that's literally the entire point of playing them.
Except that there's an easy way to bring new players into the fighting genre, and that's to make a fighting genre game that is easy to get into, with mechanics that can help the less-skilled player win sometimes... and it's called Smash Bros., and it's probably more popular than all other fighting games, combined. And when players have graduated from playing games that have random item drops to wanting to play a game without those random elements (Fox only, no items, Final Destination), they are ready to play the games that expect you to have the basics mastered coming in.
Taking the position that getting new players interested in a genre such that the genre doesn't asphyxiate from sheer player attrition is inherently going to ruin the game is a rather false dichotomy, and one that both hurts the genres you want to preserve and also justifies some really bad design decisions as things that just can't be helped when it's blatantly not the case. Besides, if those MMOs were willing to take desperate measures to get new players into their servers in the first place, doesn't that mean they were going to go under if they
didn't get new players, anyway? Doesn't EVE Online's lack of need to take desperate measures to get new players prove they
don't need to take those measures to keep new players coming in because their game already draws them in? This seems a causation/correlation confusion.
Anyway, there's a huge difference between doing something like destroying game balance with constant item power inflation to make all the items players were grinding for a few months ago totally useless and obsolete and simply redesigning the user interface to do the exact same thing in fewer button pushes.
I've been saying this for years in several of the other threads, but the biggest problem with Dwarf Fortress is its Information Invisibility problem. There was a bug a while back that had been in the game for years involving eyelashes that grew forever, resulting in dwarves with eyelashes several feet long because they would grow a millimeter every few game ticks (wasting FPS) forever... and nobody noticed (because the only indication of eyelash length is a details page text description that doesn't describe lengths past a certain point) until Quietust found unusual data when memory-hacking the game. Dwarf Fortress is a game where even the developer doesn't know what is going on under the hood, and only memory hackers can find out how things are actually working. You need things like DF Hack to see crucial data like whether your livestock is gay without excruciatingly time-consuming trial-and-error setups.
By far the most dire sin Dwarf Fortress frequently commits is that information (when you can actually find it) is often segregated from where decisions are made. Oftentimes, information you need to make informed decisions is only found in places like the details page, but that's often several button presses away from places like the labors menu, and you lose your place in the menus any time you try to go look for the information you need to make the decisions on those other menus.
It should be the credo of any interface designer that the information you need to make an informed choice should be on the same menu as the one where you make the choice (or at least, just a single pop-up away that you can go straight back to the decision menu without losing your place).
Even beyond that, there is the problem of there being so many different ways you need to navigate the UI and gameworld. I, for one, always make sure to rebind keys just to make things a little easier on myself. (5 and 0 for up and down make things vastly easier than < and > to go up and down while the numpad moves in all other directions... no wonder so many other players make ludicrously inefficient horizontal fortresses when it's such a pain to move up and down!) As already mentioned, controlling moving up and down in lists or extending lengths or areas of different constructions can be any of three different combinations of buttons. There are FIVE different buttons that have a "look" function, depending on if you are looking at creatures, loose items, workshops, areas, or furniture. Surely, just streamlining this so that there's a unified control scheme, and you don't need to memorize totally disparate controls for every different menu (because Toady now wants to use a different control scheme than he did 3 years ago when he designed the last menu, but isn't going to bother taking the 3 minutes it would take to rebind the keys in the old one or because he's afraid people would complain if he changed the old setup) wouldn't cause some sort of MMO-style new-player-hunting death spiral.