@Sam_Ursu I like it! random drunks are always badly informed! But they can come up with interesting points
Whatâs coming out of this discussion is that âwalls of textâ is placeholder to represent how text is presented. As pointed out by @Gwen itâs all about how the text is shown, when and relevancy. I also agree that typography plays a part too.
Something Iâve been thinking about is, what i call, the interaction cadence (perhaps thereâs already a term for this), but itâs basically the rate at which text (or output) and interactivity fit together.
I agree that large initial exposition and world building text puts people off. Personally, i always get a sinking feeling whenever i start a game that throws more than a page of text at me before i can do something (and pressing MORE doesnât count).
But the idea of interaction cadence applies to the whole game. I want to read a bit and do a bit, and repeat in small chunks of reading and doing, otherwise i get bored. My theory is thatâs what other people want too.
For me, this cadence, is an important part of game design and it more relevant for narrative IF than graphical games.
To address @MoyTW 's question, of what is the question. I think the question is turning out to be; Not whether less or more text is bad but;
What is the optimum interactivity level for interactive fiction?
Just to add, that it may well be, this âoptimumâ level differs with age group.
And audience. E.g., CoG audience has a specific set of expectations about cadence thatâs been worked out through years of feedback between authors and players.
Hear, hear! I generally gave up on the IF scene ⌠um⌠a long time ago, when I realised that the general view (at the time, anyway) was that players should be adjusting to conventions of IF parsing rather than the other way around. I recall trying a comp-winning game, which was positively player-hostile. IIRC, it opened with the player sitting in an audience in the middle of a play. So I typed WATCH. âWatch what?â was the response. WATCH PLAY. âYou canât do that!â AAARRGGGHHH!!! And it didnât improve after that stupendous opening. You are sitting by the wall, so EXAMINE WALL. âWhich wall do you mean?â I gave up soon after that.
I would say that as a bare minimum, parser games should offer the list of all verbs they understand as well as the list of objects (and pseudo-objects) that the player has already encountered. While it would not cure all problems, at least it would give a novice some chance to observe that EXAMINE is listed as a verb non-synonymous with SEARCH.
Interesting. I totally agree about listing verbs with a VERBS command. But listing all objects sounds like it would make the game harder. When youâre 15 rooms into a game and you type OBJECTS, that might be a huge list of things youâve encountered, many of which wonât be useful. And including pseudo-objects (like background stuff) sounds like it would make a headache for the player. You could list only useful objects, but that might be spoiler-y.
It really is a tightrope to walk to figure out how much help makes a game more fun, and how much help steals the playerâs agency. I am glad that we are wrestling with these questions, though.
I think there are occasionally cases where the fun is in the player discovering the action for themselves, and offering the verb from a list would spoil that moment of discovery. What should be avoided at all costs are the times when the player has had the flash of insight, knows what action is needed, but has to wrestle with the parser to actually get it done.
Not necessarily. By pseudo-objects I did not mean everything that appears in scenery, but whatever the player can sensibly manipulate, even if it is not an formally an object, as far as the gameâs engine is concerned. But, yes, I should have qualified objects and pseudo-objects as those that have been encountered and can be usefully manipulated.
Scenery scanning is another kettle of fish. I also experimented with that and it has to be implemented carefully to avoid causing confusion.
I speak from personal experience here. Have a quick look at https://mipmip.org/adv770/adv770.html, which implements vocabulary listing and scenery scanning (and automatic vocabulary abbreviations, and a single-typo vocabulary matching). It has a pre-game, but you need not bother with that â just type VOCAB (or HELP, which will mention VOCAB). It is a large game, but the vocabulary is quite manageable.
I am not âoffering the verb from a listâ. I provide a VOCABULARY command which includes (but is not restricted to) all verbs one needs for the ultimate success (any magic formulae exempted). Guessing what verbs a game might understand is one of the most frustrating pass-times I know.
In my game for Spring Thing, I have a VERBS command which tells you the verbs you need to win (a list of 9 excluding directional commands), and then it offers you the option to see other commands: like magic words, silly commands, help commands. I think splitting it up like this may make it friendlier to folks who arenât interested in XYZZYing. What do you think?
Elsewhere recently I mentioned Ascension of Limbs , which literally tells you every turn what verbs and nouns are available to you, with tightly constricted choices. Itâs an interesting approach. But I wonder if selecting choices from a menu would be a better interface once the choices are constrained that far.
Iâd leave the silly stuff for an amusing command available at the end of the game. One expects that one of the reasons a user might be consulting verbs is because theyâre confused, and itâs more helpful in that case to avoid extraneous info.
Thatâs very true and interesting. If Iâm playing IF, I want to interact.
But the thing is that kind of goes against
because VNs can be notoriously low-agency - âagencyâ in this case meaning how much the player can actually do⌠Iâve seen letâs-plays of visual novels where there is tons of clicking to advance but no actual choice until hours in.
I think in Doki Doki Literature Club the player enters their name and then doesnât do anything but advance until the end of the first day where they write a poem by clicking lists of words to tweak some relationship stats. Then they choose what order with whom they share poems on the second day, but thatâs not a choice so much since you have to lawnmower them all. Thereâs a subtle three-way branch thatâs based on the relationship meters set before, and then the player chooses whom to agree with in an argument, which branches a bit of dialogue but mainly tweaks relationships again. Thatâs maybe 2 or 3 actual choices in about an hour of play and two in-game days.
I suppose your point depends whether âinteraction cadenceâ includes âpress something to continueâ or true interaction. If some people are bored by a wall of text, others are bored clicking mindlessly forward without doing anything.
You initially posted âthe visual novel styleâ is what they prefer, but thatâs basically spoon-feeding them an enormous amount of non-interactive text one line at a time. Almost every parser game is more interactive than than if you consider agency vs âclick to continueâ.
That said, there can be a difference between reading VN dialog heavy text (where presumably some may assume more is actually âhappeningâ since a character is speaking) than reading expository world-building text.
Well, I started playing IF, afterwards got to adventures that switched input method to icons as Monkeys Island saga is.
I canât maintain my attention in a game when the adventure consists in some text walls wich through a simple choices lends to dead or even worse another text wall. My entire body protest painfully and try to enter hibernation mode closing eyes and sleeping. I cant barely maintain me awake. There is no coffee enough. This situation seems to me as if I have been cursed.
This, an no other, is the reason why I donât play choicescript or twine / ink games more often.
Yes to splitting things. In my case, VOCABULARY command gives you the option of seeing just the particular section, e.g.verbs. The game is large and lists 76 verbs and, depending on the state of play, up to 150 or so objects. Itâs the verbs list that is the most useful one, but the nouns list also gets consulted, as CGI build logs show. When first published (back in 2003) it seem to go down well with players. Some even took the trouble of saying how helpful it was.
Another bugbear for me is the lack of charity in interpreting what the player is trying to do. As far as I am concerned, the combination of UNLOCK, DOOR and KEY in any order at all, is a comprehensible command. Why insist on UNLOCK DOOR WITH KEY ? Yes, there may be specific cases where the order is important and/or the specific preposition need to be supplied correctly, but that is surely rare!
In philosophy of language, Davidson points out the crucial role of the âprinciple of charityâ when hearing othersâ utterances â we assume that they make sense and interpret accordingly. I do think parsers should do likewise.
Uh, well, Iâm definitely under 40 and I love IF.
I think itâs just a matter of what youâre used to and what you grow up enjoying. Clearly, IF has a lot features that visual novels donât, and puzzles that are impossible in visual novels. For most IF Iâve played I would struggle to see how they could be adapted to VN form (or even Twine form).
I grew up playing narrative IF games mostly from the 2000s-2010s like Galatea, Photopia, and so on, mostly from the Zork app. This was partially because they were free and I wanted to play games in a household that didnât allow it with devices that couldnât handle games very well in a era where most gaming was on handheld devices made for gaming or on PCs. IF was a natural leap from reading. So the barrier of entry was crossed when I was young and motivated, and the idiosyncrasies of the parser feel natural to me.
I donât think the problem is necessarily âwall of textâ because I think reading is still very popular, and thereâs even less breaking up text in VNs than in IF games. Instead of thinking about what command to input next, or deciding on what thing you want to examine and getting relevant information that way, you just click to proceed. I suppose there tends to be more of a dialogue and character focus than world modeling, but thatâs not the same thing.
Hm, this sort of reminds me of Grooverâs works. Eat Me has >eat, directional movement, and a few others as the only possible actions, and this is made known to the player at first opportunity. I noticed it had good reception outside of the usual IF circles, and by people whoâd never played IF before, though some of the puzzles were still considered difficult.
Midnight. Swordfight. also has a limited set of commands, all of which are listed to the player with >playscript. I think to be accessible requires some amount of built in tutorial.
For both of these games itâs still relevant that theyâre parser games where you can enter unprompted commands, and to not be point-and-click, however.
An important piece of context here is that the 40+ crowd were the so-called âMTV Generation,â raised on TV and quick edits and instant gratification. In the 1980s, we too were supposedly unable to sit still for more than four minutes to read a book or a long news article.
I like to think there will always be readers who appreciate the unique power of the written word. Thatâs who I write for.
On the point of wall of text, esp. intro and world-setting, I think that the natural spot for word-setting is the feelie.
In the old days of long installation/generating game disk procedures from slow media, reading the manual AND whatever feelie when floppy and/or hard disk doing their work was a sort of ritual for me.
Ultima seriesâs documentation is perhaps an extreme case, but admittely very effective in introduced the players to Britannia (the game world of Ultima, for who donât know)
I donât hide that Iâm pondering how to expand (into feelie) and compress (the textual intro) of a WIP whose has admittely as intro a substantial wall of textâŚ
I remember those days! [visibly crackles into dust]
Some of that was before the technology provided a lot of memory and disk space, so offloading instructions into a paper manual and fun feelies such as an image of a map maximized resources for the game. Many games (including Infocom) also only displayed text in low-res capitals, so it was a much more comfortable reading experience for users to have a printed manual.
Nowadays, text and memory space isnât an issue, and using lots of packaging and paper for manuals might be considered wasteful.
Also user interfaces are a lot more friendly and games have tutorials (thanks to the extra space) so many people may not need a manual. But feelies were a majorly cool part of opening an Infocom box and other games that provided nice packaging.
Interesting. IF has changed again, and we havenât noticed yet. Once upon a time, the default output was typed onto a paper roll, which was why games defaulted to printing out the room description only the first time a player entered it. This saved paper, plus it was often quicker to scrabble back up the paper roll and re-read the description than to wait for the type head to s-l-o-w-l-y r-e t-y-p-e e-v-e-r-y-t-h-i-n-g.
Then the default output became a CRT screen, but for years IF stuck with this conventional default, despite there being no paper to save. But in both cases, the normal input device was the keyboard. Players had already achieved some familiarity with it, even if it was only at the hunt ânâ peck level of typing.
Now the default input has changed from being a QWERTY keyboard to touchable links on a web page (probably displayed on a phone.) New IFers regard keyboard with much the same curious disdain I feel for fountain pens. Sometimes beautiful, but slow and impractical.
The IF-systems are already evolving, so that they now bundle the game file along with the 'terp to produce a web page, but it looks like we need to assume that touch sensitive surfaces will be the normal way that players will interact with the game, and incorporate object links in the text into an interface which facilitates that* as the default.
*A scrollable verb-strip down the left hand side of the screen? Touch the link âchestâ to examine the chest, (examining something being guaranteed to be harmless) but scroll to âopenâ, touch it to confirm, and then touch âchestâ to assemble the command in the text box at the bottom of the screen where the prompt is.
Iâm sure that this has already been done, what Iâm suggesting is that this should become the default, to lower the technical bar to producing games friendly to those outside our magic circle of IFness, in the same way that the Inform 7 interface aimed to make writing IF accessible to a larger number of non-coders.
(I can still recall the alarm I felt when I discovered Inform (I6) required me to put something Iâd created into a device, and then use the output of that to make a game. It seemed to me about as sensible as controlling fast moving machinery by jamming a stick into it. I decided to defer this dangerous and possibly lunatic experiment until I had my own PC to mess up.)
The thing about âwall of textâ is that a wall of text is bad writing. In the most egregious case the writer isnât even offering the reader the support of paragraphing to help structure the information being presented.
This is not a problem confined to IF, as a quick glance at any fan-fiction site will show, but I think the problem is slightly different in IF, because of the imbalance between coding and writing skills.
Coding skills are gated by the medium. If you donât learn the syntax, your game wonât compile. If you donât learn to structure your puzzles into if/then/else cases, then you canât express the problem youâre presenting the player in code. So IF enforces a basic minimum of coding skills.
But itâs entirely possible to release a meticulously coded game with text like
Town Square
Newtown is high on a hill, overlooking the countryside. A blacksmiths is to the east and a store is to the north. The town gate (closed) is to the south.
The above satisfies one requirement that ordinary prose stories donât have: IF descriptions have to tell the player how to navigate that space, but that chunk of writing above achieves nothing else.
Writing is hard. Writing good is more harder.
Worse, I think that descriptive prose writing and coding use different and incompatible parts of the brain, so you more or less have to make separate coding and prose writing passes over each part of the game to produce something acceptable. Since in order to experience the prose in context, the coding must already be in working order, I think the reason for some of the poor, unengaging writing is simply author exhaustion.
But only some of it. Iâm amazed by the way people, who would never dream of inflicting their lousy hangman-grade drawings on others, are of the opinion that if you can write a shopping list, you know all you need to know to write a novel.
Plus, the penalties are different. If an author releases a mis-coded game, then people are puzzled by why they canât drop or give away the wine bottle (needed for a later puzzle) but if they walk through the kitchen again, it mysteriously pops back into the wine rack where they found it. But a game with poor prose is simply not very engaging, and players looking for a story will get bored and quit. Only the puzzle-heads will keep playing, because theyâre not here for the props and scenery; theyâre here for the cunning brain teasers.
I think that VNs get a lot of slack because even with no choices to make, there are interesting pictures to look at - whether those are cool space stations & robots, or big-eyed busty babes, the player remains engaged and entertained until the choice point.
Again, this is not unique to IF. In stories itâs a mistake known as âfront loading,â and itâs something to which Science Fiction in particular is prone. The story opens with Brad Broadjaw captive, strapped to the Infernal Plot Machine in the laboratory of Professor Pluggenheim who bores on for five pages about his new lightspeed propulsion system.
Book goes back on shelf.
Re: the EXAMINE vs SEARCH thing: Iâve always understood that examining something was looking at it closely but without touching it, whereas searching something implied physical interaction. So itâs about being fair to the player where traps are concerned. Examining the chest wouldnât cause the player to encounter the poisoned needles (and is probably the way to get clues about how to deal with that puzzle) but rushing in and searching it would find the hidden compartment, but penalize the player by having them going through half the game with one hand swollen up like a cantaloupe (so they couldnât wear the Gauntlets of Itch Scratching until theyâd found the antidote.)
With modern IF games moving beyond the stereotypical âadventureâ trope, I agree that behind the scenes it makes sense for search to map to examine almost all of the time. I suspect that 90% of the time search gets used, itâs because a player has tried to look through a window, âlook throughâ being one of searchâs synonyms.
Every genre has its conventions, e.g. D&D players expect that when they first explore a new area, theyâd better be loaded up for defense. Itâs only once theyâve gotten out alive and returned to the Radished Orc to quaff mead and tell tall tales that they arm themselves with the right gear to counter the specific threat theyâve discovered. The process of reconnoitering is considered part of game discipline.
But I take @mlaâs point that this can go too far â it should be us striving to accommodate the playerâs expectations, rather than expecting them to adapt to our habits, otherwise we end up only talking to ourselves. Iâd hate for IF to recreate the tragedy of Superhero Comics, where the only comics that sold well enough to be commercially viable were Superhero comics, so the only comics that were published catered to Superhero fans, so comic stores sold superhero comics and nothing else⌠and comics basically ate themselves, instead of sequential art becoming a major narrative form as it is in Japan.
The cases are not exactly parallel, of course. I think in IFâs case, our Achilles heel might well be nostalgia. I appreciate that for a sizeable part of our community part of the fun of their hobby is not only writing a game, but writing a game that Neil Armstrong could have played aboard the Eagle when he got bored with the view.
You need really good coding chops to do that, and god knows Iâm aware that people like me, who donât know an argument from a punchup, wouldnât be here if it werenât for the goodwill of such coders whoâve made, and freely share, all these tools, but most people who play games want stories, rather than puzzles.
I worry about IF, this odd text child constantly punching above its weight. It is to modern gaming what kitchen tabletops are to animation: small scale enough, and low budget enough that enthusiastic amateurs can learn the ropes, and actually produce work, without having to master another three disciplines first.
The entertainment industry has huge problems, surrounded by âthe Moneyâ who are only interested in making sure of a good return; by sponsors anxious about âunsavoryâ content, and by censors and pressure groups insisting that some things must never be shown, and other things be always a consequence.
IFâs strength is that itâs small enough that it can sidestep all these issues; it can say things that art should be able to say: itâs important.
So I think that itâs important that we continue to struggle to make IF accessible to as many people as possible. If that means abandoning conventions that were once necessitated by hardware, we should do so, no matter how fond our memories of calling âYeah Mom!â down the stairs and typing softly into the night by the screenâs green glow.
