Producer Quincy Jones on producing Michael Jackson’s Thriller album:
Well, we had great rhythm sections, which we did first. We did what we called Polaroids. We must’ve looked at 600 or 700 songs. When you get a song you feel you like, you put it down with a rhythm section to get it on its feet, and then you hear Michael sing a couple of takes on it, maybe with a couple of background lines to see how it holds up, so you can see what it might be and you’re not just wasting your time. We called those Polaroids. Then, when something sticks, you develop it further, get into background lines and horns or synthesizers or whatever else you’re going to be using.
I like “Polaroids” as a way to quickly get across the idea of rapid prototyping. Go through a ton of stuff and give it all a quick shot. Then see what sticks and devote more resources to that. That way, failure is cheap. You’re actually expecting failure and embracing the idea that only a small percentage of your ideas are truly good enough to earn a big chunk of your attention.
The LAByrinth Theater Company, a collective in New York that specializes in new American plays, provides another example of the Polaroid approach. Every year, LAB has a two-week “summer intensive” workshop during which 35 to 40 plays are rehearsed and read. Company members then offer their critiques and the artistic directors then select the 10 or 15 plays they would like to see go to the next step. Again, failure is cheap.
Good inspiration. If LAB can put on 40 plays in two weeks, what can we get done in that amount of time? We always say we don’t have enough time, but maybe the problem is we’re just trying to do things too well.
Stacy
on 25 Feb 09I see your point but the analogy doesn’t fit. Design and coding takes far more time than sampling music. They are sampling “taste” among themselves, and we need to sample useability among paying customers. They know and understand their customers. Our customers may be totally different from project to project. I don’t think there is any shortcuts as seductive as music would suggest. We are more like making a movie than music.
Chad
on 25 Feb 09I absolutely agree with you! I use this approach in initial layout mocks.
David Andersen
on 25 Feb 09600 to 700 songs? 40 plays? Aghast, the tyranny of choice!
Ryan
on 25 Feb 09I think the analogy fits. The point is… experiment with several mini-projects before jumping into one with both feet.
Do pencil/paper mockups. Make a bunch of versions. Then move to Photoshop or OmniGraffle. Get a lot of eyeballs to look it over and give feedback. Then make a revision or two and get more feedback.
Only after that do you move onto the expensive, time-consuming coding process.
Jon Plummer
on 25 Feb 09The NeoFuturists (theater collective in Chicago) combine rapid prototyping with production, writing, rehearsing, and performing 30 two minute plays each week, 50 weeks of the year.
They have an interesting process which is best explained in their own words. A video of an interview with their Artistic Director Jay Torrance at last year’s UX Week conference is at http://vimeo.com/1551821 . The production process part of the talk starts at about 5:03.
Scott Semple
on 25 Feb 09”...trying to do things too well” [too soon].
Tom Kerwin
on 26 Feb 09“We are more like making a movie than music.”
Don’t film-makers do their own version of paper prototyping with storyboards?
As a recovering procrastinatoholic, I love the idea of just blitzing out loads ideas (which seems to be a new theme at 37sigs)
Phil
on 26 Feb 09The other thing we can learn from this example is to be open minded about the end result, and to see where creativity takes you. The didn’t set out to make the Thriller album. They set out to make the next Michael Jason record, and the creative process, and Polaroiding brought them eventually to Thriller. Important distinction. Great record, by the way.
Thibaut Sailly
on 27 Feb 09Yes, the analogy fits. The numbers, maybe not.
Music is art, design is not – you have to deal with the constraints. There is much more freedom in music so it’s “easier” to get to a 100 different songs in a given time. The design process leaves you less options, choices have to be balanced with both logic and emotions , it takes more time to come up with really different solutions.
I’m with Ryan on the use of sketches at first. I recently came back to use tracing paper as it’s easier to produce variations of a structure. You can use colors for elements families, you can draw big or small, precise or fast. The other advantage of paper is that you can lay down all your “polaroids” side by side and evaluate them. This is something you can’t do with a computer yet. Knowing how to draw really saves me lots of times, and on top of it, clients really enjoy looking at their designer sketching a website.
This discussion is closed.