Rewind: Interactive Edition (from my book)

Brand new dystopian interactive fiction based on my book Rewind: The Empyreum War. I’d very much appreciate any feedback both here and on the Kickstarter preview. The Rewind Project now has a Facebook page at, where I’m posting regular updates. Please help support Rewind and help me create a new breed of interactive fiction for PC, Mac and iOS.

Thank you :slight_smile:

"Here at Rewind, we care about your future - and that’s why we have chosen to invest in your past. You wouldn’t keep all your tax records on one holoputer without making regular backups, would you? - and yet, every day, millions of people across the galaxy walk around with their heads full of precious memories and no way to back them up. Until now.

Why watch your loved ones go off to war, knowing they’ll never return, when one simple visit to Rewind could safeguard your future together? One hour is all it takes to back up your memories onto our systems. Everything that makes you an individual will be stored away safely, and the more you come back to visit us, the more up to date your backup will be - and If the worst should happen, Rewind will supply a cloned body to your own specifications into which we can restore your memories.

Think of it - you’ll never die. You’ll just come back as a younger model, different body, same mind. Call us today for your free information pack."

  • Rewind Advertisement

Official Project Website:

Follow us on Facebook:

Rewind: The Interactive Novel is brand new dystopian interactive fiction - playable either as a full length old school text adventure, or in the format of a choose your own adventure book depending on difficulty preference - for Mac, PC and iOS. It will be based on my script book Rewind: The Empyreum War, which was itself written to redress the balance between a multitude of epic fantasy series hitting our screens in recent years and the great SciFi multi-part classic not seen since the days of Star Wars (yes, I know it’s back!). Painting a bleak vision of a post-apocalyptic dystopia ruled over by The Company, Rewind: The Empyreum War (available to buy or sample through or Amazon) has been described as a mixture of Blade Runner and Scary Movie, referring to a dark humour interwoven with a plot full of shocking twists which will make you laugh, jump and cry in equal proportions right until the very end - and if it doesn’t, a couple of killbots from the department of corrections will be around to sort you out. Think “Hitch Hikers Guide to the Blade Runner” meets “Nineteen-Eighty-Fifth Element”. Or Something.

Rewind: The Interactive Novel will be written using a proprietary mix of blood, sweat and tears, and by sitting up all night over an extended period of time mashing a keyboard while cursing loudly at my programming software’s inability to magically know what I wanted it to do without telling me that everything is a Syntax Error. The game isn’t written using any sort of Interactive Fiction design software, by dropping a new vocabulary into an old parser, or by rejigging something I made earlier using sticky-backed plastic and Lego (there’s no s on the end, people) - this is an all new game, written from scratch in order to bring you just the right combination of retro and cutting edge and hopefully prevent latecomers to the world of interactive fiction from throwing both their hands and tablets in the air at the first sign of a tricky puzzle.

So what makes Rewind different from a traditional 80s text adventure, apart from the meticulously crafted graphics and 30 years of extra RAM and hard drive space just crying out to be filled with more story than a 16K ZX Spectrum could’ve ever imagined? Always assuming, of course, that said ZX Spectrum had the memory or hard drive space to run software capable of imagining such a thing, which it didn’t. Excuse me for a moment, I appear to be stuck in a loop.

Imagine a text adventure so rich, so immersive that it’ll be like reading a book and making a decision every couple of lines that changes the outcome of that book. Imagine a game where every command you type results in a paragraph that moves the story on rather than a simple impersonal “I don’t know what a sword is.” - a game with little repetition, big on story, and based on a full length Sci-Fi epic written by the same author. You’re imagining Rewind: The Interactive Novel.

Like such IF classics as The Hitch Hiker’s Guide to the Galaxy before it, Rewind will be an adaptation of an existing book. Unlike H2G2, however, the game will run on computers and mobile devices capable of holding the entire text of the book, which means that the story will be far truer to the original. It will actually feel like playing a book, and where old school text adventures will have you stuck in a room for hours trying to work out exactly what combination of hieroglyphics you need to type in order to pursuade a troll to give you a sword, Rewind will involve you in a constantly moving story which evolves and develops according to your decisions. In Rewind, there will be no “stuck in a room”, and the game’s responses will adapt to the location and number of moves made which means that there will be very little of the numbing repetativness of old school adventures where every command results in a stock response.

The game is designed to appeal to hardcore text aficionados and newcomers to the genre alike, allowing the player to play through the game either as a traditional text adventure or a more simplified choose your own adventure novel. When in text adventure mode, the player is in control, typing commands that control the story, while Choose Your Own Adventure mode will offer predefined options throughout - making the game significantly easier for newer players who do not wish to spend their evenings searching Ganymede for a key. The game will also include original location and artwork throughout (both background graphics and in-line scrolling illustrations similar to the sketches in choose your own adventure books), an original soundtrack, and built-in mini-games to break up the typing and reading - and where possible, you’ll be able to customise the interface, colours and fonts to your own taste rather than just take what you’re given.

Until now, the Rewind project has been a labour of love, squeezing every moment of spare time out of me in the hope that one day Mr Hollywood might notice my book and hand me a large wad of Arcturan dollars to direct a movie (yeah, right) - but now, I’d like to put my full time into the project and dedicate myself to producing an interactive novel to the same standard as the book, full of devious sci-fi puzzles and mind bending logic, but without a Babel Fish dispenser or colossal cave in sight. Except perhaps in parody. This will be a new breed of game - a game which will suck you in and involve you, and have you wanting to play it through again just to see what would’ve happened to the narrative if you’d asked the Squarglian bouncer politely to let you into Blue Johnnie’s rather than dropping that rock on him from the third floor balcony.

Good luck!

Thank you. If it’s a success, I’ll be creating a development system at some point for others wishing to create interactive fiction with the level of depth Rewind will afford. Of course, anyone using it will have to do a lot of work on the text :slight_smile:

You’ve obviously noticed my British sense of humour - a lot of Hitch Hikers influence in there.


That sounds interesting. Am I right in thinking that this magnum opus is like a Choice of Games story in tone but using a homebrew parser? (Yay, homebrew parser!) (Aha, I see the screenshot on the Kickstarter page.)

EDIT - I put the following in a rant tag, because maybe it’s a bit unfair. I don’t know how you’re approaching the parser, so maybe I shouldn’t butt in. Then again, I’ve been let down by too many homebrew parsers. The best I’ve seen, FWIW, are from The Adventurer’s Musem and Robin Jonhon’s games. Hopefully I’ll be able to add Rewind to the list, heh? :wink:

[rant]Very ambitious, and it looks very interesting. But I’d like to give you some advice regarding the homebrew parser there, because advice is free, well-meant, and often unwanted. But I’m gonna give it anyway.

I don’t quite understand what you’re doing with the parser - whether you’re going for the old-style model of having a fixed verb+noun format, or whether you’re going for more flexibility to allow for more apparent freedom. If you’re going for the latter, like those parsers that just do pattern-matching… well, let’s just say that I’ve played a fair number of games which tried that, and they were all pretty terrible. They were great when they worked; when they didn’t, they fell apart, because the pattern-matching approach is an unstable and fickle foundation.

If you’re doing verb+noun, I’d encourage you to check out “Cypher” for what NOT to do in a game. “Cypher” is also a homebrew parser, and one that ignores all sorts of conventions and shoehorns new ones in. It doesn’t allow for many shortcuts, and it actually has you type >CONTINUE when you are presented with some text and and to read on (a half-decent system would have you press any key instead).

Your work sounds like it’s very ambitious, which is good, but I’d feel a lot better about it if I knew that the freedom you’re going for comes from having a set of verbs, complete with lots of synonims and, in some cases, shortcuts (nothing fancy - n, s, w, e, ne, nw, sw, se, u, d, i, x, l - should be enough), implemented solidly enough that I can go nuts. I’d be a bit more wary if, instead, you’re encouraging the player to type crazy things. I mean, it’s amazing when you type a crazy thing and it works, but if you’re routinely encouraged to type crazy things, you’ll hit the limits of the simulation sooner rather than later and it’ll be difficult to steer you back into place.

There are some conventions to the parser, and they stuck as long as they have because they’re really effective. I would encourage you to use as many of them as you can.

And, of course, to gleefully ditch the ones you don’t need/want if you’re tsure of what you’re doing. It’s your prerogative.[/rant]

I’m not quite sure what just happened to my last post, but it vanished. Sigh.

The parser is exactly like any old school parser - verb noun - but with the intelligence of games such as The Pawn which harked on about being able to PUT THE SMALL BLUE HAT UNDER THE TABLE IN THE RED HAT ON THE TABLE. I’m not trying to reinvent the wheel. VERB NOUN alone is fairly restrictive and not very forward thinking though, so you’ll be encouraged to be a bit more wordy. Is that a word?

Well, the book is written as a script and is therefore in acts. The plan is that the game will play through act one in a sort of leading format so that the choices you make adapt the game later but the text moves on whatever you do - for example, if you stop to look at a market stall as you pass, you can buy things, and if not then you miss the chance. In this way, new players can get into the game without getting immediately frustrated by being stuck in a room. By the time you reach act 2, the game starts to become more location and puzzle oriented and movement becomes more freeform but still maintains the non-repetitive quality by having so many moves in which to solve an area. If, for example, you have 100 moves to solve a puzzle before poison in the air kills you, then there is no need to have repetitive responses to your commands. The game can quite easily have 100 programmed responses and therefore not repeat itself.

Obviously, it’s more complex than this in practice as you can’t have infinite programmed responses to every command, but it works remarkably well in the in-house concept demo, and the fact that each command typed by the player illicits a new paragraph of the story means that you very rarely notice when the game does get forced to repeat itself.

And yes, the parser is entirely homebrew. I used to write games back in the 80s and developed parsers that ran in real time - IE: it’s no good sitting back and thinking for too long because the game will continue without you running on a timer - “John enters, John leaves” etc. I fully intend to work some sort of real time aspect into the latter parts of Rewind.

That’s actually what I’m afraid off. Maher’s blog posts about The Pawn make it clear that, while the MS parser was great at handling sentences like that (because it was constructed to show those sentences off), it makes for a more frustrating experience if you’re actually trying to play the game.

Personally, I never saw the appeal of writing a sentence as long as that… A parser that showcases that leaves me thinking “If that’s the only way to get that done, something needs re-thinking and re-designing”. I’m looking forward to Rematch, but of course Rematch is a short game that does that as a gimmick.

Maybe if you can give me examples of why VERB NOUN (heck, VERB NOUN PREPOSITION NOUN, which is what we actually do have) is restricted and not forward thinking I can get around to agreeing with you, but as it is, I find it liberating instead of restrictive (with five senses we perceive the whole world; with a VERB NOUN structure we can do anything) and, against all odds, still as conducive to experimentation as it was 30 years ago.

EDIT - ANYWAY! I don’t really want to derail this, and your game does sound excellent. I’ve said my piece, and I’m curious to see the end result. :slight_smile:

You may want to consider making a playable no-graphics no-frills prototype at some point just to let some people play around with the parser and get that feedback. At some point.

+1 for a minimal, very simple demo!

Don’t get me wrong, I can’t remember ever playing a text adventure where anything other than VERB NOUN was ever required, although I may be wrong. It’s just nice when someone can be a smarty pants and type LOOK AT THE REALLY COOL GROWING SHIELD UNDER THE TABLE without the parser throwing a fit. Naturally, normal people will just type LOOK SHIELD and get exactly the same response :slight_smile:

Sure, there will obviously be the occasional instance when you need to determine which of two hats you’re referring to, but I see puzzles like that as the exception to the rule. You don’t want to alienate the player. In fact, Rewind aims to do exactly the opposite - to make the game easy to get to grips with for a modern generation of players for whom TAKE BOOK takes 8 too many letters to type.

First, welcome to the forum and good luck with your project.

Second, when I saw this thread was already up to nine posts I worried it would have turned into a shitstorm. I am glad this is not so. :slight_smile:

Third, your project is ambitious on several axes – any of which could fail and drag it all down. This is obviously scary. I do not want to be the wet blanket, so I’ll assume you’ve thought about all that and move on.

How much modern IF have you looked at? Many of these problems have been tackled, and there are known successful models. …only not for all of those problems in the same game!

So, for example, the Choice of… series does a great job of presenting character stories while letting the player move the story along (with a lot of possible variation) at each input. They’re menu-based choice games, rather than parser games. Then there are hardcore puzzle parser games, which are great, but they don’t try to be CYOA at the same time.

The “modern standard” for parser IF (which goes back to 1990 or so) is best described as a three-term parser, rather than a two-word parser. There’s a verb grammar (which might be multiple words, like “PICK UP __” or “PUT ___ INTO ___”) and then object terms fit into the blanks (an object term might be “THE SHINY STEEL SWORD”.) It’s perfectly true that players tend to minimize (“PUT SWORD IN BASKET”) but we know that we can be more verbose, and might need to be for disambiguation.

Inform 7 handles all of that except for the relational part of the object term ("…UNDER THE TABLE"), which we all agree would be great but happens to be a weak point of that system.

The great Zarf :slight_smile: Bows down.

I think it would be wrong to look at Rewind as pure interactive fiction in the traditional sense - it’s almost a new genre in itself. You type commands in the IF way, but what you’re really doing is playing a book. Imagine reading a book and getting to interrupt the narrative every couple of sentences to make a suggestion - look at that man over there, talk to the beggar you just passed, etc. There are places where you are free to move around a series of locations and solve puzzles IF style, but mostly you’re moving the story on with your choices but with the freedom of a parser rather than picking from a list of limited options. The scoring system allows you to go back and play again to try to collect the points you missed. The option to switch to an option based narrative is mainly to address the vast majority of players out there who unfortunately have never had the pleasure of IF and yet have seen a recent resurgence of Fighting Fantasy style game books on iOS, who hopefully will be able to get as much pleasure out of the game as those of us who like to strain our typing fingers :slight_smile:

Oh, and the actual book on which the game is based is, of course, essentially a hint book…

Seems interesting. I’m really curious now. But I would again encourage you to, at some point, get people to play-test a prototype of your parser.

What I find especially interesting is how you’re trying to conflate the parser and the CYOA designs. Sometimes it looks like you’re talking about a parser game, and sometimes it looks like you’re talking about a parser game that really should have been a CYOA. There’s room for a lot of pitfalls, and I’m naturally wary… well, no, I’m naturally cynical, so I assume the worst… but you seem to have a strong direction in mind!

Parting thoughts[size=85] (EDIT - Is it really parting thoughts when you’ve added a bunch of stuff in after the fact?)[/size]: when you do give people a prototype of your parser in action… listen really closely to what they say. “Cipher” has soured me, they totally ignored every point that was brought up against their parser. You seem like a much more decent sort, so I’m sure you won’t do that, but if people point out ways in which the parser could be bettered… and if that bettering actually brings it closer to an existing model that you were trying to break out of… don’t fight too hard. :slight_smile:


Mind you, if the parser throws a fit and just says “What?”, that’s very unhelpful. But if the parser says “I only understood you up until…”, then you have a VERY useful error message! One that allows you to understand why your action didn’t work - you have to know whether you may have tried a verb the game didn’t know, or a noun, or even whether the object’s implemented. A strict VERB NOUN system usually wields the useful errors, the pattern matching system prints the infuriatingly unhelpful stuff. Usually.

Anyway, the best way to avoid fits is extensive beta-testing. :slight_smile: “Coloratura”, at points, was extraordinary in recognising non-standard input, because in some situations they were the commands that felt natural to type. I can’t see this being necessary for a whole game (if it is, you’ve failed abysmally at communication!), but when it IS necessary… oh hell yeah. I’m behind you 100% on that - it’s sweeeet to have the game recognise those commands.


Ever the optimist! :smiley:

What I’m getting from this is that you’re writing incredibly “cutscene”-heavy IF - long non-interactive sequences, followed by short opportunities to act. Is that correct?

No, not at all. As Rewind is a new genre of IF, it’s hard to show what I mean without a demo - which is in itself hard when the development system requires a large initial outlay of cash before it will compile anything distributable and funds are tight (hence the need to pitch on Kickstarter). I’ll get something done somehow…

In a CYOA, you tend to read a couple of pages of text and then choose to do something. Imagine Rewind as feeling more like a traditional IF game where you type a command and get a couple of lines, perhaps a paragraph in response and then you type another command. BUT - the story is constantly moving. For example (and this is not text from the game):

“I walk along the dark, gloomy road, looking left and right as a multitude of colourful characters pass by. In a shop doorway, a beggar reaches out pitifully toward me, his clothes nothing more than rags.”


"The beggar bows down, his voice trembling at a level of kindness rarely seen on the streets of New Amsterdam. He takes the shiny coin I offer him, staring almost unbelievingly at it as he turns it over in his hand. Then, as quickly as he came, he is gone, disappearing among the crowd.

I reach the street corner, where a collection of shopping carts block my path."

[Command Me]

The story moves on, but you get to interact with it at every step. Sometimes, what you do will profoundly affect the story or even make the game different down the line (giving the beggar 20 credits might cause him to help you later), at others your choices will be essential to continued play. In Rewind, for example, if you’ve missed the opportunity to solve a puzzle and gain the code to your door, then getting into your unit without being killed by a security droid later becomes harder - not impossible, because Rewind rarely leaves you without some sort of option for continued play, but harder nonetheless - and you get awarded less points for missing the correct solution, so you’ll want to replay the game later for a higher end score.

Think of the street in the example above as a “location” - just one that the game wanders around by itself while you interact with it. In other situations, a location is much more traditional - an example of this would be Kantrell and Marie’s accommodation unit, where you are situated for up to 30 moves. During those 30 moves, you are witness to a conversation between Kantrell (you) and Marie, but during that time you can explore as you like, examining things, taking things, talking to Marie about things, as in traditional IF. After 30 moves, when your conversation is over and the game has told you everything narrative wise it needs to tell you, the game moves on. If you solve the puzzle inherent to that location before the 30 moves, you’ll move on in your own time. You also play as several different characters throughout Rewind, so sometimes it is neccesary to leave items laying around for someone else to pick up later.

There are also plenty of places in Rewind where you are free to explore as in traditional IF, moving around locations and solving puzzles to progress. In these cases, the non repetitive narrative is maintained by giving the player an artificial time limit (say, 100 moves before something terminal happens), so that the game can have 100 different bits of story to tell you between moves.

Wow, this sounds exciting, and also very ambitious. I think this kind of time pressure immensely reduces the player’s/reader’s tolerance towards imperfections of the parser. If I imagine that I have only a brief time frame during which I can do something, and I don’t succeed just because I did not phrase it right, the story moves on, and I arrive at the point where it would help me to solve a later puzzle… aaaAAAAaaargh. :smiley: I would also love to see a demo, though.

Don’t worry, the time sensitive sections are few and far between. Since I’m retelling an existing story from my book, I’m very keen that the player experiences all the thrills and surprises of the story and feels as though they’re part of it without getting stuck too often in one place for an extended period of time. Remember that the constantly moving narrative is offset by the fact that the game is long - covering the entire first script book which itself covers two two-hour episodes. That’s a big game with a lot of story. You’re not going to find yourself being rushed through and going: “Oh, is that the end?”

The points system is also designed to add replay value, because you’re going to want to go back and work out what you missed and how you should’ve solved the problem that you hacked your way through :slight_smile:

In terms of “not phrasing it right”, the parser is actually fairly forgiving. When just the right command is required, the game will actually look for the right words somewhere in your sentence rather than expecting them to be in exactly the right place grammatically. So you can play with the language a bit in these situations.

Makes me shiver, that…

I find your description of your design extremely interesting, though, and maybe - just maybe - that design will avoid all the traditional pitfalls of a matching-pattern parser. After all, you’ll mostly have self-contained situations, where only a few actions will be meaningful. That may just make it work much more smoothly than I feared, significantly reducing the chances of the parser completely misunderstanding the player’s input.

Best of luck! It’s looking more interesting all the time!

Well, I’m glad the project is being generally well received - I wouldn’t expect it to be all things to all people, but I would be a fool to aim it squarely at either old school text adventure aficionados for whom graphics are the devils work (!) or hardened newcomers for whom text is something you find in those boring papery things made from trees. There’s clearly got to be a middle ground so that Rewind appeals to everyone (well, lots of people anyway!)

Just to keep you all in the loop, I will be working on a screen video of the gameplay through the first couple of locations tomorrow, so that you can get a preview. The issue with a demo still remains lack of funds to purchase a non-development only version of the compiler that will output a functioning application, but in the meantime I can at least give you a gameplay video. Watch this space :slight_smile: