Ill give you one clue or more like a hunch...
If you have a tool that is trying to do everything it isn't going to be equally great at all those things and it will likely be confusing/hard to use.
The generic programming languages tend to be more popular but they do so in the same way TV programs or video games are trying to appeal to an audience as large as possible. (This is how we got all these absurd hacker movies for example)
In the future (lol) we will discover that single purpose languages are just better at the limited scope of things they do.
PHP is perhaps a bad example that I shouldn't even have mentioned here but such a language knows exactly what its goal is in life like PHP knows it is suppose to bake websites.
Maybe you've touched the awesome with your fun. Someone should try build a language entirely around the core mechanics of fun.
In games there is fun in the form of rewards for stuff that just takes a fucking long time to do, there is fun from rewards for stuff that requires skill, there is fun from rewards obtained though luck, there is fun from a storyline progressing, there is fun from unexpected things, fun from buying ingame shit, fun from selling ingame shit, fun from cooperation as well as growing to be able to do those things on your own.
But the real list is probably much longer.
The language should probably have a basic fun object that looks something like: { temporal: 0. skilz: 0, luck: 0, story: 0, randomEv: 0, pay2win: 0, progaming:0. coop: 0, solo: 0, [...etc...] }
Then you have to benchmark the fun people are taking out of a bit of code or graphics using real world data.
And then....
Then you would be able to focus your attention where your effort produces the largest amount of fun as well as see the areas where your game is teh suck. Answer the big questions like what parts are people playing and why? Where do they rage quit?
If they didn't give up on creating content for Diablo 2 I would probably still be playing it.
If they had a language where fun was the central mechanic they would have known that changing all the items and ruining all the heroes had a negative dev time to fun conversion ratio.
It wouldn't have to be limited to games at all. One could quantify the fun on HN using the same language. It would all of a sudden be obvious that the karma system and submission ranking lacks random rewards and events. A thing no one considered up to now but if we know it is fun and the system is lacking it it becomes worth considering it.
</fun>