As the major k-film projects are being made something that is becoming clear is the, well I’ll call it fluidity, the fluidity of how to make a good k-film. What do I mean by this? Well, it is still not that clear as we are collectively inventing ways of working in a very new form, but it goes something like this.

A procedure is created that letscontent be created. Stuff be filmed. This procedure might be in response to a list. It might be based on a story. A poem. Some lyrics that you’ve created. It might be based on an arbitrary rule who’s function is to let you generate content easily, that uses a repeatable formula. For example filming each journey each day. Or to film something (an object, an observation, an event) at the same time or interval each day – 5pm every day, or every half hour for a day. Yes, they appear random, but they produce a body of stuff, an archive of material, raw stuff (think clay) that can be worked with. The point is that even where you are filming in response to quite specific guidelines (a script) this first step has been about being able to collect plenty of material, easily.

Then comes the second moment (and why this is called ‘making squared’). You collect and collate this material and now look at it. What patterns does it contain? There might be patterns here that are different, or complementary, to what you thought you were filming. For example many of the projects that I’ve seen as prototypes can easily have a simple keyword structure of night and day, or inside and outside, or still and moving, or looking out and looking in. This might, in fact, be the way to weave the work together – remember your audience will make assumptions (make hypothesises) about why things are related, whether they really are for you or not – and so by rethinking the patterns of your work you can provide a simpler way to build the project. Remember, these patterns are imminent to the material you have gathered. You are now concentrating on the material you have to hand, not the script, story, shot list, list.

Why do this? Because learning how to pay attention to the actual media you have, the actual material, is how you get to be good. You are poor editor if you think the storyboard has to be strictly followed when editing a film. Same applies if you think your essay plan has to be slavishly obeyed when writing. If the material (the shots, the writing and ideas) ask for other ways, then give them a go. So this is a step in learning how to listen.

Why do this? Because this is what your stuff, your media, is telling you it is. Here’s a scenario to help understand. Imagine you’re writing a song. You have lyrics and music. You’re sketching, riffing, and so on. A line forms. It is a bloody good line. You know that. You’re not sure what the music needs to be but you know that line has to be in the song. You don’t throw it away because you don’t have the music yet. You keep it because it is worth keeping and using. You then use it to help work out how the music can get there. Reverse it. You play some stuff and you know that that melody has to be in the song. You don’t know what lyrics will go there yet, but as music it is important for the song, as music, for that passage to be in there. It is a to and fro, a back and forth. A crafting.

So in your k-film you have this library of media, and now you can look at it with new eyes to see what it actually is. Ah, close ups and wide shots. Mainly things (objects), or just spaces (views). Perhaps moving camera, still camera. These are patterns, patterns that let you then link things together. Doing this might mean you can build the work with a small number of keywords. Why is this good? Because the patterns are stronger with small simple rules. Remember complexity does not come from lots and lots of keywords, but from a simple, significant rule repeated. This gives the work a strong dynamic structure that can be understood, and then your viewers are not wondering what the ‘structure’ is, they are now paying attention to playing, viewing and interpreting your work.

However, this also poses a problem, as you now have a simple dual structure (for example night and day), and you then have the problem of how to connect them. This can be solved by allowing a randomised link to appear when nothing matches a keyword search, but a more elegant solution is to use what I think of as dense nodes.