XpdWiki
Set your name in
UserPreferences Referenced by
JSPWiki v2.0.52
|
Bold assertion: The single biggest problem faced by software development teams is the management of creative thought on an industrial scale. Interesting parallel: another industry that faces this problem is feature film making. Consider the creation of a contemporary feature film. There can be hundreds, even thousands of people involved, for periods ranging from months to years. There are specialists of many different kinds. There are hard technical constraints, there are time-tested conventions, there is innovation. Consider, in particular, the making of the three Lord of the Rings movies. This project is currently (2003) ongoing. It will eventually have consumed many thousands of person-hours of effort. How is the business of creation being organised? Well, if you get the "special extended DVD edition" of The Fellowship of the Ring the bonus disks reveal a lot. Here's a précis of some of it. Peter Jackson (director), and his collaborators got together to begin deriving a screenplay from Tolkien's novel. Before that was completed, artists had begun to draw "storyboards" to illustrate the screenplay. A storyboard is like a strip-cartoon, a sequence of drawing showing crucial shots, with the dialogue for the shot written below, and arrows and such to indicate movement. Jackson did a clever thing with these storyboards. He filmed them. By using a video camera to capture a few seconds of each storyboard image, and using zooming an panning onto the image, he was able to create a first, very low fidelity, cut of the entire film. He even hired a bunch of cheap actors (not bad actors, but not Sir Ian McKellern?, either) do record a dialogue track. The storyboards were reworked as and when required as the production continued. Next, small models were built (of about Blue Peter, "here's one I did earlier" standard), and action figure dolls used to film key scenes in slightly higher fidelity. Then, some of the sets were begun, and before they were finished Jackson and other crew members would set up a camera and play scenes themselves, to get a feel for how they would work in the set. About this time a key technical innovation was made. There are Men, Elves, Dwarves and Hobbits in the story. Men and Elves are around six feet tall, dwarves around four feet tall, and hobbits around three feet tall. As it happens, John Rhys-Davies who plays the Dwarf Gimli, is the tallest of the actors playing any role in the Fellowship. That makes him correctly proportioned in relation to the shorter, more slightly built actors playing the Hobbits. So, how to make the Dwarf and the Hobbits appear correctly proportioned to the Men and Elves? Sometimes through CGI, but often by using a technique as old as film-making: forced perspective. Forced perspective is a variation on deep focus, where the object to be seen as smaller than it is is placed far away from the camera. Traditionally, this could only be done with the camera locked off. But Jackson is a very dynamic director, he likes mies-en-scene shots with the camera moving around. So new motion-controlled rigs were developed, parts of the set would move relative to one another as the camera moved and so preserve the forced perspective. At other times, the hobbits were played by "scale doubles", actors of the correct stature, who sometimes wore full-face masks bearing the likeness of the actors playing the main part of the role, and sometimes wore blue screen masks so that the main actor's face could be composited in. Throughout production, complex effects and model shots were "pre-visualised", using simple computer animation with very limited rendering to indicate how they would look (and sometimes the cheap dialogue track, too). Working with the model-makers and set builders were two "conceptual artists", both of whom had worked on illustrated editions of the novel. They produced fully rendered drawings of set and prop design concepts. They worked on the models and sets themselves, they drew visualisations of completed sets based on the current state of the work. I think that we can see here clear evidence of iterative working practices. Shots and scenes were filmed, and reworked, and filmed again, and again. Each iteration produced a version closer to what the audience would see on the screen. That didn't stop with principle photography and editing. A lengthy post-production stage took place where the entire film was scanned into a digital tool that enabled Jackson to alter the lighting of scenes without re-shooting them, changing the colour balance, adding key lights to actors faces, and so forth. And it didn't stop with the release of the film to cinemas, either. The DVD version is longer, has a new score, and some scenes were re-shot. There were also evolutionary practices going on. If you've seen The Fellowship of the Ring, recall the scene prior to the Fellowship reaching the bridge, where the staircase collapses with the Fellowship on it. That's quite a long scene, and it isn't in the novel. It isn't even in the script! The screenplay says "the Fellowship run down the stairs to the bridge". However when one the conceptual artists was sketching some ideas for the set, he thought it would make the stairs more visually interesting to have a gap in the masonry, a couple of missing steps. From that one visual cue Jackson was inspired to add one of the most dynamic and exciting sequences of the film. Note that the screenplay was not altered to reflect this change. One person who's views are not recorded on the DVD, but from whom it would be very interesting to hear, is Barrie Osborne, the producer. If the director is the "architect" of a film, ensuring its consistency of vision and such like, the producer is the project manager, trying to get the thing done on time and to budget. And so far he's succeeded. So, There's a lot of talk about "software engineering" and "discipline envy". Many of these fall down immediately; the reality of what those other engineering disciplines really do (rather than what the aspirants assume they do) is compared with software development. One central difference with almost any other branch of engineering the huge cost of producing an executable at all, but the trivial cost of producing as many identical copies as are required thereafter. Another is the almost unlimited mutability and plasticity of the medium. Another is the curious mix of hard technique and technology with pure creative thought. But perhaps a discipline, one that shares those characteristics, we might do well to be a little envious of is film-making. --KeithBraithwaite Cynical assertion an essential problem of commercial software development is getting the management and customer to understand that it requires creative thought at all On second thoughts this has strong parallels to filmmaking as well... --TimBacon3host33date3February? 2, 20033ParallelsWithFilmMakin 33agent3Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.2.1) Gecko/200211303ParallelsWithFilmMaking
|