It dominated the scene for quite a while. Looking at the Xyzzy award winners will confirm that. Even the first Best Game winner ("So Far" by Andrew Plotkin) was an Inform 6 game.
However over the past decade or so the scene diversified a lot more, particularly as choice-driven rather than parser-driven games became a lot more popular. Twine is a tool for making browser IF games that really broke through there, although in recent times Ink has gained popularity both for making IF and as a tool for making dialogue systems in other non-IF games. Most new parser-IF games I encounter though still tend to be Inform.
A small tangent: for readers interested in IF and curious about trying to write one, Inkjam is 3-day event where you make a short game with Ink, and this year's event is in early September. I participated last year and it's good fun and a good challenge!
I feel like this issue is brought on by the fact that it seems like copyright law hasn't been properly reexamined for the modern era. I support the notion that the ideal model is people pay the rightsholder in order to enjoy the media, and offering resale of pure-digital media instead leads to some weird "You make money based on the peak simultaneous number of owners, rather than the number of people who consumed it", which just doesn't make any sense to me as an economic model.
So yeah, you see labels calling the user purchase one thing, in order to make it line up with the intent of the consumer relationship, then on the royalties they call it a different thing to make it line-up with that intent. I don't feel the publishers are violating the intent of these relationships at all, it's just that copyright law being extremely out of date requires stupid language games. The correct solution is to re-examine copyright law to either establish that the intent is you pay the rightsholder to get access for you as a distinct individual, or that the intent is that you are purchasing resalable access, and be done with this nonsense.
I absolutely don't blame artists for trying though, the labels screw them so it seems fair that they should try to screw the labels.
There's definitely been improvements in level editors for existing engines. Hell, if you want to use those old editors, there are even plugins like HammUEr, to use Hammer with Unreal 4 and Qodot to use editors like TrenchBroom with Godot.
But I will say there was something particularly special about level editors coming with the greatest games of the day back in the 90s and early 2000s. Not only was this extremely accessible as it meant you could just learn level design, and not the rest of game development, but it meant you got to learn level design for your favourite games, something particularly enticing to young people. You also got a bit of a built-in audience.
The incredibly low skill floor is definitely a big part of the scene's success.
I know a number of veteran professional level designers are kinda worried about the field. So many of them got their starts playing around with the level creation stuff that frequently came with PC games. It got them interested and allowed them to walk before they could run. So they're a bit afraid a new generation might be starting from less than they had. There's some hope that maybe Minecraft and Roblox will be a younger generation's starting point.
Game development is the most accessible it's been in many decades, but level design specifically is perhaps less accessible than it was circa 2000.
Doom mapping is basically it's own weird artform that people have been refining for nearly 30 years. Unlike a modern AAA game it's typical for a single person to make a Doom map and so people can develop very unique and identifiable styles that come through in their maps. If you're an old Doom-fan and never got around the checking out the scene, Cacoward winners are a great place to find some highlights.
I don't think it gets quite mentioned explicitly in this video, and don't I know if it's what the parent comment was referring to, but I recently learned the interesting detail that this lock system was invented for the international NES, and was not a feature of the original japanese Famicom. And apparently Nintendo did have a bit of a problem with large numbers of bad unlicensed games in that market.
This apparently was a small motivator in the development in the japan-only Famicom Disk System, a floppy-disc-like drive addon, which did use a protection system that amusingly was based around trademark law. There was a number of other interesting elements about the Disk System, but I'll suppress my desire to vg history ramble :)
However over the past decade or so the scene diversified a lot more, particularly as choice-driven rather than parser-driven games became a lot more popular. Twine is a tool for making browser IF games that really broke through there, although in recent times Ink has gained popularity both for making IF and as a tool for making dialogue systems in other non-IF games. Most new parser-IF games I encounter though still tend to be Inform.
A small tangent: for readers interested in IF and curious about trying to write one, Inkjam is 3-day event where you make a short game with Ink, and this year's event is in early September. I participated last year and it's good fun and a good challenge!