Thematically it doesn’t sound like my jam personally, but I love the idea of MUDs and text in general. One of the few good things that LLMs have done from my point of view is to make people think about and help prove to a general audience how effective and powerful “chatrooms” using a text-based interface really are and can be. It’s foundational to the way we think. As humans we developed language for a reason, it’s a huge part of our survival and continuation as a species and how we have come to dominate this planet the way we do. It’s how we process information. It’s how we communicate information. We developed keyboards specifically to have a hyper-efficient and accurate way of entering those languages into a computer, and it actually makes sense that generative AI relies on that fact so heavily.

But at some point, some people decided that typing commands was too hard to learn and we wanted to make everything “easier” and more “visual” but that came at a subtle but significant cost that was never really accounted for in how efficiently we can individually convey our own thoughts. Anyone can write a message or description or a story in the same way that, yes, anyone can also draw a picture or act out a scenario, but the difference is in how broad the skill gaps become and how dramatically less efficient the information transfer becomes as you move away from written language into more complex ways to illustrate thoughts and ideas. A crude and amateurish story is not as far removed from a brilliant story as a crude and amateurish drawing is from a brilliant drawing. Fewer people are capable of being decently skilled artists than they are decently skilled writers. By the "making things more visual, more easy, more slick, we steadily put up heavier and heavier barriers to entry on the creation side at the same time it lowered the barrier to entry for the consumption side. It shifted the balance away from creators and towards non-participative consumption. Even when people do participate and engage, the most common form of participation we see on social media is people writing a comment. Again, it’s text. There’s a reason for that. It’s the lowest effort and most accessible means of communication in the vast majority of cases and situations. Even popular images and videos usually feature (or are exclusively!) filled with text.

Having your entire environment built out of text makes it accessible in a very fundamental way. It lets people interact with it on a more equal footing, everyone has the same tools and limitations, but the freedom is almost unlimited. It lets people participate in it more creatively, not just moving a pre-rendered graphic around or picking appearance choices out of a menu but actually inserting their own thoughts and feelings and ideas into a roleplay or a description with a level of freedom that no other kind of user-interface would actually allow.

All this to say, I appreciate how text-based environments like MUDs promote the democratization of creation, interaction and thought, and I think they or something like them will and should always have a place.