Interactive Films – First contact

With the Band or Brand website done (I might write one last post about that one, I don’t know yet), we have a new project on our hands: make a static video, that can then be interacted with through a website’s interface. So, I’ll start by looking at some initial thoughts on this.

Dissecting the brief

What is this about? First off, let’s get to the technical requirements. We need to make a static video (so not animated real time by JavaScript), and fit it on a website. We’re going to have to make it enjoyable on a 1080p screen, but also compress it enough so that it doesn’t take insane amounts of data and time for a regular to access it. This part I can probably solve if I mess around with Adobe Media Encoder’s settings for long enough. As for what the video will contain, how it will look, or how I’ll film it… well, I have no idea yet. So I’ll just postpone that part until next time.

As for the second step: we have to design a UI for the interactions. This ranges from the logos and text appearing at the edges of the website, the background image if there is one, or the video’s container, all the way to the buttons or prompts for the user to interact with. Most of it will preferably be in vector format to save on both size and load times, which is crucial, especially if the video will require fast reactions for some reason. We also have to think about making all of this look and feel consistent (unless we have a very good reason to deviate from the rules we establish).

Finally, we have to build the website itself. This will consist of three steps: building the framework and inserting the video with HTML and CSS, writing the basic functionality in JQuery or JavaScript, and then finally inserting the JQurey code used to interact with the video. This should be the simplest part of the entire process, although as usual with coding, bugfixing is going to take some while to complete (to a reasonable degree). If I have the time I might even attempt some cross browser testing, just to make sure the site works on stuff like Firefox too.

But for now, let’s look at some of the examples I found:

Call Bulls#!t

This one is built around a simple concept: the user watches short clips from the debates that took place during the United States election season last year, and they have to decide whether or not the candidate talking is actually telling the truth, or just talking, well, bullshit.

The player has one button in the bottom left corner of the screen, using which they can express that they think the speaker is lying (at least to some extent). There is a limited amount of time to make the choice, and after each candidate’s clips are finished, the user is told what percentage of their guesses were correct, and also gives them the average score of everyone who played before. Depending on whether or not a choice is made during the allocated time, and if that guess was correct, the player gets feedback through a brief audio clip of the narrator saying something like “Yup, that’s bullshit.” or “No, that is not bullshit”, or some other relevant line, after which the player is shown a brief explanation of the particular topic the candidate was talking about.

It’s a simple, yet very effective work, which points out the flaws in politicians’ arguments, and promotes critical thinking. Also worthy of note is the fact that both parties are equally represented (3 republicans and 3 democrats), and no side is favored by the quiz itself.

Don’t Look Away

Regardless of what I think of the song itself, this is an interesting piece of media. The video consists of still images of black people who recently died due to police violence, and each photograph has their name, age, and cause of death overlaid in harsh, white text. The idea behind the site is to make users “face the facts”, and to achieve that, they constructed the site so it requires a webcam to use, which tracks the face of whoever is currently using the computer. If they look away, the video jarringly pauses, along with the music, and the only way to make it continue is to look back at the screen.

This is like that one episode of Black Mirror, except instead of pushing advertisements onto people, the face-tracking technique here is used to get a point across about current social issues. Having said that, this is a weird specimen. It is an interactive video, but the only interaction it requires is for the user to do… nothing. Which is the exact opposite of interaction.

Coldplay – Ink

I like the art style. Sadly everything else isn’t really satisfying in this video. How it works is that during the music video sometimes the main character, and thus the player, will have to choose between two (or in rare cases three) options, by clicking on one of the highlighted objects on screen. Depending on the choice one makes, the video plays different scenes, or the same ones but with some changes to the actions of the characters. So far so good. The problem is that since the video has to be synced to the music, it doesn’t progress immediately when the player clicks on their choice. Instead, they are locked into their choice, but still have to wait for the time to run out. This wouldn’t be so bad in itself if the site provided clear audio-visual feedback on the player’s choice, but that is sadly not the case. Multiple times during this short song, I was unsure whether or not the game even registered my clicks, since the only indication of it happening is the highlights fading away from every object, except the one selected. But this highlight is sometimes very faint, and in general isn’t visually clear enough.

The other issue is there is no way to know what these choices entail. In most interactive videos, or movie-like games, whenever the player has to decide between multiple options they have a rough idea of where each one will lead. Here, the stylized nature of the video, and the surreal events that happen in it, don’t succeed at making the user’s choice feel meaningful, since when two highlights come up, there’s no real ground for comparison between them, so the player just randomly clicks on one and watches what happens. Oh, and after replaying this a couple of times, the choices don’t seem to make too much of a difference overall, since the same choices come up every time. It’s like having several forks in a road, which don’t matter at all because both sides will join up well before the next one.

Five Minutes

ZOMBIES! On a more serious note, this one got really close to working for me! I liked the basic premise, the cinematography, the acting, the music, so pretty much everything that is related to the “video” part of “interactive video”. And it’s not even like the interactions are bad, far from it. But before I can delve into this, let’s step back a bit and explain what this is.

The story concerns a father and a daughter in the middle of a zombie outbreak. They are currently holed up in a small apartment, with zombies right outside the front door. The father is holding a gun to his own head; as we quickly find out, it’s because he may be turning into one of the undead too. In this universe, the first sign of the transformation is memory loss, which happens within five minutes of the infection. So if the father can remember after the countdown on his watch has finished, he is going to be alright. The player is told what happened leading up to this situation through flashbacks the main character has.

The main mechanic is using the mouse (or a touchscreen) to draw on the screen according to the little arrow prompts that appear. There is also a bit of point-and-click shooting, and a few moments where the mouse has to be kept inside a constantly moving target area. These work great in the flashback sections, since the quick curvy motions one makes when drawing the required lines fit the visceral motions the father does during the action scenes. The problem is that the way the player makes him “remember” is an overly abstracted version of the same thing: fill up a bar, by drawing arbitrary shapes, before the timer runs out. During these segments the focus isn’t on the story anymore, but the artificial barrier between the user and the video, which destroys the immersion, at least for me.

The biggest issue however, is that there is no branching whatsoever (other than the two second long clip that’s the equivalent of a “game over” screen). Failing at the “remember the past” sections forces the player to try again, and after three fails, it’s game over. The flashback parts cannot be changed, obviously, since they have already happened by the time we start the story. This undermines any immersion the gameplay facilitated during the action scenes, since failing at a specific action (like drawing a circle to turn the keys in a tense situation) doesn’t affect the outcome of that action specifically, but instead is handwaved away by the father failing to remember, and dumping the player back into the present.

And then there’s the ending, which pissed me off upon replaying it. Long story short: the daughter is holding his dad at gunpoint, asking him to remember what color her mom wore when they last saw her. The father quickly glances at a photograph, has a quick flashback, and then the player is offered three colors to choose from. In the flashback, and on the picture, the mother is seen wearing yellow. Choosing yellow makes the daughter leave her father alive, but she also says that she could’ve sworn she was wearing a red dress. Cue camera panning out into the hallway, where the zombified-mom is seen wearing said red dress.

“Okay”, thought I, “so this is the big twist. I’m gonna go replay the game just to see what happens if I pick red!”. And this is where my anger comes from: THERE’S NO OPTION FOR RED! You can only pick blue or purple, both of which result in the daughter shooting her dad (but for some arbitrary reason, eventhough she remembers her dress being red, answering yellow is the “correct” response now). This defeats the entire point of having interactivity in this piece! There is no consequence to any of the player actions, other than a “You suck, try again!” clip every time they mess up something, and despite the twist ending enticing them into a second playthrough, they are not rewarded for it, but are instead left frustrated.

Linkin Park – Lost In The Echo

Not available anymore, at least I couldn’t find the interactive version. Here’s the static one though:

Pointless gimmicks, ho! The music video for this song featured a bunch of photographs people were looking at, crying over, or carrying them around, etc. The interactive version of it let users link their Facebook accounts, which lead to those photographs being replaced by images from their own gallery. It was a nice idea in theory, except for people like me, who don’t have too many images up on Facebook, a lot of the photos in the video were switched to the exact same one, or a random Family Guy screenshot a friend of mine sent me two years ago.

At least it lead to some unintentional comedy when the actors got all emotional about Peter Griffin being Peter Griffin.

Set Wars and IKEA: Where Good Days Start

Sadly the actual website for the IKEA one isn’t available anymore, but from watching the trailer for it, I feel like I know what it was mostly like. Both it, and Set Wars utilize interactions as a means to let the user expand on the video’s content. In Set Wars this means opening up a slideshow, or old photographs from the Star Wars sets, or other things which pause the video and let the user explore a topic they may be interested in. In the IKEA video the user can pause the video and look at the products that appear around the rooms, tag them for later, or just learn more about them. There are also some gimmicky “gameplay” elements, like a very short sidescrolling platform game, but that’s not the focus here.

Honda: The Other Side

The original is, once again, not available, however we can still discuss what it was about. The short film told two different stories about the same person: one at night, and one during the day. The user could switch between the two at any moment by holding down a button. To further drive home the point of the two similar, yet polar opposite stories, every single shot was made so it was almost a replica of the its pair, but with major thematic differences. For example a car driving by is seen in both versions at the same time, from the same angle, but during the day it’s some friendly neighbor driving, and during the night it’s a police officer looking at the main character with suspicion. This is a great core concept well executed!


From these examples alone, we can clearly see that some videos focused on a set narrative, while others instead based the entire thing on their mechanics. Five Minutes is a great example of the former, where the interactive sections feel like afterthoughts for the most part. The Honda commercial on the other hand, is an example of the latter: they thought about an engaging core mechanic, and based the entire video around that one thing. From this admittedly small-sized sample it seems like the narrative focus is the rarity. Most of the time the interactions are just added as some sort of gimmick, or extra feature in the hopes of capturing the attention of the viewers. These are ultimately shallow, and inconsequential to the story of the piece.

The true “choose your own adventure” style interactive movies are not usually found on the internet; they are instead video games on Steam or consoles. Until Dawn,  Heavy Rain, or Telltale’s The Walking Dead all fit the brief more than any of the examples above. So let’s involve those in the discussion too from here on out!

Different ways of interaction

When we want “interaction” in a video, that could mean a bunch of different things. I feel that it is important to clarify this early on, so we can build cohesive mechanics for our film/game.

A lot of the examples listed above featured fairly barebones, somewhat gimmicky interactions. The Set Wars documentary, or the IKEA commercial, or the Don’t Look Away video from Usher all have  interactions that do not affect anything, other than the user’s own experience.

A bit more complex is the addition of failure-states. In this case the user has to perform a task in a limited amount of time, or just do it correctly. If they fail, then the film is over, and they have to start again. Otherwise, the video still plays linearly, and the only impact the user interactions have is whether or not they can keep going forward. This is exactly what Five Minutes did, except it seemed like it was actually attempting the next category, which is:

Roleplaying, or assuming control over a character, or multiple ones, in the story. This kind of interaction ranges from moving the player avatar around, all the way to sprawling dialogue trees. This is a perfect fit for narrative games with meaningful interaction, since the player is directly engaged in the story, by virtue of being one of the actors. Until Dawn, Mass Effect, The Walking Dead, and many more games use this formula, and make sure that each interaction has some narrative consequence to it.

Then there is a weird category, where the player can affect the story, but acts more as an outside observer, or like the director of the movie, instead of one of the actors. Interactive movies, or “FMV games” released in the 90’s with the advent of the CD attempted to do this. While I don’t know many examples off the top of my head, I know that the infamously bad Plumbers Don’t Wear Ties used this format. No, the quality of the game itself had nothing to do with the player’s viewpoint.

Overall, I think that although the first two categories are a lot easier to film and implement, the latter two fit the brief a lot more, and I feel like I should strive to achieve one of those.

Next time on Interactive Films: Scene transitions! And other stuff!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.