I would also note that, whatever the API, parser games still leave something to be desired in the year 2022 compared to what is now possible with other gaming frameworks. I'm thinking of indie games like Papers Please and Beholder. But also Twine, et al.
The parser interface accomplished beautiful things in its heyday, but now feels to me to be needlessly constrained and difficult. I think there is a future for text-based IF. I think there are entire genres of games that are only (theoretically) possible with that interface. But the usual <verb> <object> + inventory puzzles formula is just too narrow, too time-consuming to play anymore. And I say this from a place of love.
In my parser game _Aotearoa_ (https://ifdb.org/viewgame?id=lrbpmlpsknsgvgem) I worked really hard to provide player affordances such as an in-game interactive tutorial, exhaustive noun and verb synonyms, plus keyword highlighting so players literally _couldn't_ miss a primary interactable item.
Even with all of this, people still got stuck, and properly pacing a story that uses an unconstrained parser world model is an exercise in frustration. But I disagree that the games are too narrow. If anything, the frustration comes in because they are too broad. No other game genre has the sheer hubris to present the player a story, tell them they are the main character, and then challenge them to literally do anything they can think of to overcome the puzzles they are presented with. "I got this," they say, presenting that blinking cursor on a blank line. "Anything you can imagine, I can handle."
Of course every parser title fails spectacularly at this. But the best ones fail in impressive and fascinating ways. S. John Ross's _Treasures of a Slaver's Kingdom_ explicitly limits the number of verbs down to six, but succeeds in creating a game environment that is enjoyable on both the direct and meta levels. My own _Aotearoa_ does a lot to remove the normal barriers to entry to interactive fiction, at the cost of some immersion should you choose to use the cues and tutorials. Andrew Plotkin's _Hadean Lands_ and _Spider and Web_ both have high-stakes, complex stories with detailed world models. This is usually the recipe for frustration and failure in parser IF, but both titles implement wonderful unique, _diegetic_ mechanisms to motivate failure and experimentation.
I would hate to see parser IF go away. I think its innovations have been more valuable and influential than many people realize, and I think there's still a lot more that blank line and that blinking cursor have to offer.
I would also note that, whatever the API, parser games still leave something to be desired in the year 2022 compared to what is now possible with other gaming frameworks. I'm thinking of indie games like Papers Please and Beholder. But also Twine, et al.
The parser interface accomplished beautiful things in its heyday, but now feels to me to be needlessly constrained and difficult. I think there is a future for text-based IF. I think there are entire genres of games that are only (theoretically) possible with that interface. But the usual <verb> <object> + inventory puzzles formula is just too narrow, too time-consuming to play anymore. And I say this from a place of love.