I have two main questions for this post, albeit a little broad and kind of vague.
What would you like to see more if in IF games? I.e different/unconventional storytelling methods, more intuitive features, better puzzles, etc.
What direction do you see IF going in now? I.e in some of the same broad examples given above.
I just thought this would be an interesting discussion for the community and I was curious about it.
EDIT - This started as an answer to your question, and along the way it became a rant. So I tagged it acconrdingly.
[rant]1. Personally, I would be very happy indeed if we went back to more linear narratives. Multiple endings, and even multiple branches through the story, are not things I can deal with very well - it gives me the illusion that I have a lot more freedom than I do, much like a bad homebrew parser. And if it doesn’t give me that freedom, it explicitly gives me key choices, so that I have to track each node individually. And at its worst, it pretends to hold my hand while expecting me to break off (I didn’t enjoy Nightfall). Finally, whenever something bad happens I’m left wondering if I could have - or should have - prevented it. The story goes on, but which branch am I suddenly playing? Am I locked out of the ending I was striving for, or was this scripted? (I find Pandora Directive to be infuriating in this aspect)
I don’t think I’m in the majority. But hey. You asked. I would like to see more linearity, so that an author could write a proper story rather than three potentially less good ones. I would also like to see all safe-combination-style and password puzzles banned forever into the darkest pit of hell. Not only it is not what I play IF for, unless explicitly clued it can send me reeling in wildly different directions - these puzzles assume that the player thinks the same way that the author does, and most players don’t. I end up overthinking everything. These puzzles are, for me, a special kind of torture, they really are.[/rant]
So, I am a newbie to the scene, and my tastes run counter to the norms I think.
But, in general, there is one very specific thing that draws me to IF as an artform: I’m interested in stories that can’t be told any other way, or at least that really gain something from the IF format.
I enjoy feeling like part of the story, even if I don’t actually direct what happens – interactivity makes me feel complicit, and that makes me more invested. And I like the way that you can use the format to affect pacing. Some IF really resembles excellent spoken storytelling more the written word, and that really appeals to me.
Broader genre representation. Fantasy, supernatural horror and sci-fi are well represented, and slice of life and surreal are not far behind, but there’s not enough mysteries or romances, and definitely not enough westerns. I don’t care for biblical fiction, but a little of that wouldn’t hurt either, and I will never, ever get tired of good historical fiction. Likewise, there aren’t enough “slice of life” settings that aren’t the suburban middle class variety. Other people have lives, too. Show me some of them. There are wonderful opportunities in IF to introduce players to alien lifestyles. Television is at last getting the hang of this, IF should too.
Pushing the boundaries of the medium, oh yes, there’s never enough of that. Self-reference gets old if overused, but peculiar interface tricks and truly subverting expectations makes me go “wow.” Likewise, if you’re going to have a game featuring magic, high tech or super powers, implement it really well and let me do lots of cool stuff with it.
Clever use of language, yes please. It’s a text-based medium, make some pretty wordstuff.
What direction is it going? Well, so far I see a boodle of choice-click games (which I wouldn’t consider truly interactive fiction, more like semi-static fiction, as the “interactive” factor is limited to clicking prewritten choices), written with greatly varying degrees of skill. I’m waiting with trepidation for the day when the fanfic community discovers Twine, frob help us all.
Gonna have to respectfully disagree with Peter Piers about linearity. I want more choices and more multiple endings. A game with only one ending has a lot less replay value. And yes, I persist in viewing IF as being, in a significant degree, about games. If there’s only one ending, then to me it’s not that different from reading a book, except a game requires more input (and in the case of a choice-based game, not very much more input, unless the author has been kind enough to implement puzzles). If I wanted a linear narrative, I’d read a book. Which I do, with great frequency and gusto, but my motives for playing IF are not the same as my motives for reading. There’s overlap – I like games that are good to read – but I also desire games that make me think.
Integrated CSS and graphics.
Parser/Link Hybrid
No typing. Mobile devices are the future.
More IF playable on mobile would be great.
As for:
Technically this is true of parser games too. The only differences are that you can support a lot more choices because you don’t need to list them all and you can have hidden choices. But the concept that you have true agency is just smoke and mirrors, sooner or later you try something the parser doesn’t understand and the illusion is broken.
Oh, I totally second that. I got so engrossed on the actual games I forgot this. More interpreters for more platforms. I would so love to see that.
While we’re at it, I hope the paradigm “online games only” never really takes off. I really, really do.
FWIW, this is all a discussion that was had relatively recently. Someone said, I think it was HanonO, that the biggest strength of the parser is that “it can reject the player’s input”, and I think that’s so damn elegant HenonO should be considered a poet laureate for that alone.
My point is that CYOA vs Parser is possibly a topic for another thread (said the thread-derailer extraordinaire, oh well).
I quite literally made a thread for that exact discussion, as seen here: Choice-based IF vs Parser IF
No typing is an interesting point - the following just came up in another thread, but I’ve copied over to here as it’s more relevant to this thread:
Voice command I wouldn’t use…I couldn’t do it with anyone watching.
I want a sort of SCUMM button array of verbs which you can tap and build your command sentence with words in the source text.
It’s sort of been prototyped with Texture and Interactive Story https://www.interactive-story.com/is/
Voice command could be good, especially if the recognizer were integrated with the parser (so its guesses were informed by the grammar lines the game could understand).
No way. No waaaaaaay.
I’ve written about this before. Typin on an actual keyboard is key to the parser, and key to the communication between the player and the game. Because there are so many possibilities and combinations, and because it’s the same device that we use to communicate with others - e-mails, chats, word processing - it’s a tool rather than a barrier, and a meaningful tool at that. There’s nothing meaningful about moving the mouse around and clicking certain parts of the screen. With a keyboard you’re building sentences, you are communicating your intent. The keyboard effectively disappears underneath your hands as you focus on what you’re trying to communicate to the game.
Unless this new generation never did any typing and types with one finger in each hand or something. In which case I suggest you make wonderful CYOA games. I’m not being sarcastic, either. I like wonderful games, irregardless of medium. If you don’t like the keyboard, you actually have an alternative.
I mean, you couldn’t possibly have a cleaner interaction. With the mouse, a click can mean anything, which is convenient but also reductive. Or you have a fixed set of verbs, you scour the screen for hotspots, then you consider those hotspots in terms of the verbs you have right in front of you. The strength of parser IF is precisely to do away with those commodities (or, to be more precise, it predates those commodities) and asks you to actually communicate with the game in a more natural fashion. Notice something about the description? EXAMINE it. SEARCH it. SET IT TO another setting. FLY away. You’re talking in English, and you’re communicating in English. The only thing less obstacle-y would be…
…well, granted, the only thing less obstacle-y, failing the game actually reading your mind, would be voice commands. I do think they’re a better idea than the mouse.
But… it’s a nice enough gimmick but it shouldn’t replace the parser entirely. Speaking in public places? Pretty much out. Speaking in noisy places? Having to shout at the game for it to understand me? Out. And just generally speaking? I’m a singer, and it’s my bane that my spoken voice is actually detrimental to my vocal health (I have yet to find the correct placement for my spoken voice, as opposed to my singing voice), and in my day job I work 5 hours at a call centre. I don’t want to speak to play a game - quite the contrary, I want to enjoy my vocal rest.
Plus, what about playing foreign games? If your accent isn’t up to scratch?
EDIT - Passionate much? Me? Heck yeah, I am. 
I tried enabling Siri as a dictation device for iFrotz and played a little bit. Have you ever actually tried that yourself? It’s way more cumbersome than typing. It also managed to mistake my saying “pillows” for “killers”. I can’t imagine entering command after command after command this way.
Plus, here’s what happened when I tried to say “Infocom, A Mind Forever Voyaging” in increasingly clearer (I thought) intonations:
“Infocom in mind for ever for your drink a mind for ever for your drink in mind for ever for you to ring”
BTW, that part-time job in a call centre I alluded to? I’m calling the UK, and never have trouble making myself understood, plus the few people who know I’m calling from Portugal ask me whether I’m English because, although they can’t quite place the accent, I sound English.
Finally, Siri requires a WiFi connection. That’s something else I don’t want for the future of IF - anything that requires an online connection.
iFrotz already does that, and in a mobile device I have to say it makes all the difference. Autocomplete the verb, double-tap the word in the source text.
It’s still more communicative than clicking hyperlinked words. 
There’s a qualitative difference between typing and clicking.
A handful of commands are most often used. Those would be a practical target for speech recognition. Speech recognition works well for common words which are often used. But each game has new nouns - accuracy is poor for novel words.
Instead of hoping to recognise random nouns, after recognising a verb, the game could display a menu to select the object.
In terms of voice recognition support, Quest is exemplary. Rather than being a black box, Quest game files use an open metadata format so that a third party speech recogniser could parse the verbs automatically.
as for html/css/javascript interpreters - there are already some, but I am thinking more about some next-gen multi-interpreter OS application based on web browser core like Blink, that has its own if-oriented interface, can play both local and online game. It should also support themes - both user defined, and those coming with the game(so authors can tailor visual look of their games)
This way you can even have different layouts and interface elements that are switched automatically based on device you are using. It’s html/css/js after all.
It can have split screen functionality, where in left panel you play a game and in right panel you have tabs with mapping application, walkthrough, ifdb automatically loading entry for the game you are playing and notepad for taking notes, or some web page, like Google Drive.
This is pretty alien to my personal experience. I’ve replayed many games that only have one winning ending. I’ve probably replayed Riven and Final Fantasy VIII more than any others. Six, seven times each? I’m not doing this to find new endings. I’m doing it to replay what I already played. And books, of course that goes without saying. You can reread a book over and over.
I think of interactive fiction the same way. If a game is just a puzzle to be solved, sure, once you’ve solved it, there’s no need to solve it again. This is probably why I like text games that are more than puzzles. Actually I don’t even need puzzles. howling dogs and Lime Ergot both have (to my knowledge) two endings each, but I’ve replayed them both many more times than that. It’s because I like the writing, the story, the experience. These games are essentially linear, but they absolutely could not be told as traditional static fiction in a book. They’re built around interactivity.
So, that’s what I enjoy about the medium. That’s what I want. Games that are as good as books, that need to be games in order to deliver their narratives a certain way.
An interactive work with one and only one ending, raises the question of what is the meaning of choices which always lead ultimately to the same ending?
If the only ending is successful then no choice nor accident can cause failure - a Panglossian Disney world. If the only ending is failure, the work may be disparaged as depressing or defeatist - in real life there’s always hope, but not in the work.
It’s valid for players to find meaning in play from outside the game - the journey/experience can be more important to the player than the ending.
A game can raise that question, but it doesn’t have to. That’s like asking what is the meaning of sitting down to watch a movie if it will always end the same way? or what is the meaning of turning a page if a book will always end the same way?
Interactivity =/= agency.
It doesn’t need to be used to explore agency. It shouldn’t be anchored to discussions about agency.
As GlassRat said, one simple but important thing interactivity can do is control pace. A game that’s written with an attention to this detail can be 100% linear and you still won’t be able to transfer the text to paper, because the text will be structured differently to capitalize on the interaction.
But pace is just one element. Both howling dogs and Lime Ergot do a lot more than control the pace. They actually deliver information to the reader in ways that a book couldn’t reproduce, and this mode of delivery is what shapes their stories.
I find your comment illuminating. Perhaps I need another term like agent fiction to describe what I want from an “interactive” work.