Scientific Method v. Rapid Prototyping v. Film Development
I’ve been having a debate with my friend Simon Hill about the difference between the Scientific Method, and the princples of Rapid Prototyping.
I have been arguing that there are close parallels between the two, to whit
The SCIENTIFIC METHOD, if one could distill it into a single line, goes like this:
Experiment –> measure –> alter –> re-experiment –> remeasure –> deduce
RAPID PROTOTYPING METHODOLOGY goes like this:
create —> test —> iterate. RAPIDLY.
To me the parallels seemed self evident. But here is Simon’s latest missive which lays the counter-argument rather persuasively:
“ACTUALLY THEY’RE NOT (SIMILAR)! Scientific method and rapid prototyping resemble each other superficially, but really, they couldn’t be more different. They are actually separated by a whole developmental stage. Let me count some of the ways
1) Scientific method is about proving hypotheses are true, consistent, explanatory—theory; rapid prototyping is about testing whether a machine, a device, a thing, WORKS and fixing it before you’ve gone too far down the road because that is expensive to change. A theory, on the other hand, can be changed with a stroke of the pen.
2) Scientific method is dedicated to producing THEORY, thought itself, that can then be applied; rapid prototyping is dedicated to producing THINGS. Those are the two most fundamentally different categories in the whole frickin’ universe: ideas and things.
3) Scientific method proceeds by reasoning and theorizing about the results of experiments; rapid prototyping most commonly fixes problems as they arise, and finds solutions through trial and error. There are no experiments in web development; there’s only trying it to see if it works. There’s no hypothesis being tested (something of the form, “If X, then Y”; and “If I press Start, then it will Work” is not what I mean!)
4) Rapid prototyping arose to cope with the complexity of certain kinds of “wicked” projects, it arose out of Cybernetics (if there is any foundational discipline that can claim parenthood), to make development go faster and increase its quality. Scientific method, on the other hand, makes everything go VERY SLOWLY.
I’m not saying there is anything wrong in rapid prototyping. It is absolutely a good idea. For engineering. For web development.”
So here’s my question, and why this debate is relevant to us at slingshot. Is rapid prototyping a useful methodology for feature film development? PIXAR, the company with the most successful track record in the history of feature development (seven features in ten years, all of which took $250 million or more at the global Box Office) certainly thinks it is: its development process is derived from the principles of software and game engineering. But how does that reconcile with Simon’s persuasive argument that prototyping is about things not ideas. Aren’t films more ideas than things? Or are they more things than ideas?
This might seem like a somewhat ontological debate, but it comes into real focus for us as we continue to refine and refocus our development process.
Development, incidentally, doesn’t stop when the cameras start rolling. Its a continuous process of honing the film at every stage - from the cuts we made as a result of audience testing on SUGARHOUSE (which I guess is an example of prototyping and testing) to the script re-writes which are taking place now on FRENCH FILM as a result of a read through we had down at our latest Film Foundry a couple of weeks ago (does a read through count as an early prototype, or is it a scientific experiment? Is a script an idea whereas a film is a thing?).
Have Simon and I got ourselves into a semantic twist?
February 18th, 2007 at 3:12 pm
Hi Arvind
I think that both a film and a script are vehicles for ideas; they’re a means of conveying emotions to people you’ve never met before through the universally understood language of story.
I’m not suggesting that every film has tout a message or leave you feeling as warm and fuzzy inside as say ‘Little Miss Sunshine’ (which I loved) but it’s this core of a film-the very reason you chose to write it, produce it etc -that has to be held onto.
So- whatever method that most allows this to happen has to be the best.
It’s also worth remembering that it’s been Pixar’s emphasis on story that’s put it ahead of every other company jumping on the CGI animation bandwagon (and co-incidentally, it seems that Michael Arnt, the writer of ‘Little Miss Sunshine’ is attached to write ‘Toy Story 3′).
All CGI animations now look stunning, but I’d be hard pushed to remember anything from ‘The Wild’ or ‘Chicken Little’ that has the depth and pathos of Buzz Lightyear discovering he’s a toy (and not even a flying one at that).
Personally, I’m all for the slightly-scary-but-exhilirating rapid development and execution of an idea than the long drawn out process over many years
(but then I probably would say that…!)
February 19th, 2007 at 4:17 am
Simon’s distinction between things and ideas seems to miss the point. If there is a difference between the scientific method and rapid prototyping, then it seems to me to lie in the nature of the goal, which governs the type of evaluation that can be applied. With the scientific method, there is (depending upon your philosophical persuasion, of course) some sort of external reality your theory must explain, and against which you can test your hypothesis. With rapid prototyping, there is no objective to point to; you may evaluate your results on some scale but in the end you must judge for yourself whether you have succeeded because you defined the goal in the first place. From this point of view, making a film seems much more like protoyping than science. Nevertheless, I’m not convinced that the superficial similarity you began with is unimportant: much of the virtue of either approach comes from the willingness to repeatedly try new things while refining a design, without becoming overly committed at any step along the way. It seems likely that this insight could apply very well to the film industry.
February 19th, 2007 at 4:34 pm
I think ideas vs things misses the point too, at least when you’re talking about film production: I think I had been spending too much time trying to get my vacuum cleaner to work and cursing the manufacturer.
Nick’s distinction between Explain Reality vs Achieve a Defined Goal is, I think, much the most relevant frame to understand and assess the differences between Scientific Method and Rapid Prototyping.
February 19th, 2007 at 8:16 pm
A further thought…
Film production: to the extent that it is creating ART, is probably a different beast again. You are neither trying create a testable theory that can
explain reality (although there is something about art that does
reach into the parts of the universe that science cannot), nor are
you just following a defined formula. Hmm, project planning for art
projects. Difficult one, if you ask me.
February 20th, 2007 at 7:51 pm
Is filmmaking always about making art? Ghost Rider just opened at no 1 in the US and it is by many accounts, erm, not to put too fine a point of it, a bag of cr*p. Although I still want to see it…
I think there’s are some very real analogies between ‘consumer product design’ and filmmaking, although perhaps not in every case, and perhaps less so outside of Hollywood… You are testing that something will sell, and then modifying so it sells. Not so its perfect, or conveys a perfect message, or so that it becomes a perfect work of art… Ergo iteration?