From the category archives:


Cameras Are Spotlights

by Matt Blair on October 28, 2009

in Perception,Places and Contexts,Senses,Tools

People seem to be tilting their heads a little higher on the streets lately.

(No, not just because of the latest gushing story about Portland in the national press.)

Our trees — the moody ones that change their wardrobe with the seasons, not the stalwart evergreens — are baring themselves for winter, and Portlanders, often with cameras or camera phones in hand, are gathering evidence of autumn before it all falls away and leaves us with short days and drizzle.

This season brings all sorts of sensations: the first time in months when you feel cold even with two jackets on, the pumpkin lattes, the smell of roasting squash, the constant uncertainty over whether it is or isn’t actually raining, the seemingly endless variety of fresh apples, the piles of leaves that the kid in me wants to stomp through, and the intuition to look up a little more frequently than usual.

Life doesn’t stop, of course, and all the things that preoccupied us two weeks ago, and will preoccupy us two weeks from now, are still there, weighing on our minds enough to even our gaze, or turn it back down to the ground.

Looking down: Not such a bad view, either...

Looking down: Not such a bad view, either...

Whether absorbed in conversation, mentally re-prioritizing my reading list (again) or simply walking around mulling over nascent thoughts, whenever I see someone fussing with a camera, it acts as a silent, subtle alarm: something interesting must be happening here.

Hmm, a building — must be working for a real estate agent.

Or we see a toddler stumbling down the sidewalk towards the parent, who is documenting another step towards confidence.

Then there are those rare — and to me, beautiful — moments when a quick scan reveals no cause for photography at all. We can find no explanation for why someone has stopped to capture some part of this scene.  And we are left to wonder:  How often am I missing something among all that seems ordinary?

A camera is an attention-directing device as well as an image capture device. To point a camera is to convey to all those around us: I find this worth remembering.

When passing a woman carefully framing a shot causes us to pause, and wonder what she’s looking at, she has done us a great favor by making us more attentive to our surroundings.

Even just seeing a photo later, out of its original context, on Flickr or a postcard or an email, can have a similar effect. We think:

“I saw something like that last week, and I didn’t stop to notice the details.  Maybe I should.”

And with that in mind, I’m going for another walk, before all the leaves are on the ground.


In a recent tweet referring to Search Engine Optimization (SEO) tactics, I hinted at what I think is absent in digital systems:

My tweet on what digital systems lack...

One of my post-WordCamp Portland tweets

My meaning wasn’t entirely clear in the context of that tweet, so I decided to expand on it.

First, by “digital system” I mean any system built around the processing of numerical data. Examples include the internet, search engines, and the IRS.

A purely digital system is deterministic. The same input will produce the same output every single time. (I’m leaving out faulty parts or accidents for the moment.)

Whether a digital system does exactly what you think it will do or want it to do is another matter.  It’s only going to do what it is programmed to do.

While predictability is a desirable quality in an ATM machine or a heart defibrillator, it’s less useful when our goals are to be innovative, evocative and inspiring.

Preserving the quirky

How can we reduce the predictability?

Unreliability is one option: Poor quality parts can cause frequent and possibly interesting failures.

Intentional chaos is another method: If you build enough complexity into a system, or aggregate enough simple components, the system will start behaving in unpredictable ways. This is not an approach you want to take if you’re designing the braking system of a car, but it can be an effective way to generate a set of ideas you might not have discovered otherwise.

The most common — and in my view, the best — way to add ambiguity, uncertainty and maybe even serendipity back into digital systems is a thoughtful integration of people: allow human beings to be curious and playful and peculiar and idiosyncratic in their use of the system.

Quirky human beings breathe life into routine systems.

What worries me about some (but not all) of these guidelines around SEO, or any set of rules that we allow to burden our self-expression, is that they force us into certain predictable areas where our work becomes less interesting.

They encourage us to make decisions that dilute our ideas and diminish who we are, like watering down a well-aged whisky to meet some arbitrary local regulation thought up by the head of a temperance council.

Less than Human

Most music software packages have a feature called quantization. When enabled, the software alters a recorded performance according to certain settings: it can make all the notes equally loud, for example, and move them around in time so that each lands precisely on a beat.

Playing new ideas into a computer in a steady rhythm can be very awkward. Quantization has saved musicians countless hours of fiddling, editing, and reprogramming, especially given how crude the editing tools where when it was first introduced more than a decade ago.

But it’s also had negative effects.

The message of a system that will quantize you is that you can be sloppy. Don’t worry about drawing a straight line, or playing in time: the machine will fix it for you.  (Auto-tune, a more recent phenomena, applies the same logic to pitch correction.)

When “perfection” is a few mouse-clicks away, it can be come the default expectation, at least for a while, until everyone starts to realize that music “fixed” by machines tend to be very boring and repetitive.

So a few years after software companies introduced quantization, they released the antidote: another feature called “humanize”.

The computer goes through a performance that’s been previously quantized, or one that was played to a metronome or click track, or maybe even typed directly into a computer, and it adds random elements to the data: it plays each note a little softer or a little harder, or shifts it a few tens of milliseconds backwards or forwards in time to give it a sense of imperfection and “human-like” variation.

I love that it’s there, and I love the concept of it, but it’s always seemed like a peculiar thing to have to do. It was a recognition that computers tend to make our self-expression less than human. Feebly, we go to the Edit menu, and select “Humanize”, hoping that an artificial randomization routine can recover what we’ve lost.

The Norms That Lurk Within

Digital systems want to quantize us: they want to put us in boxes, attach us to tags and keywords and categories and clusters.

They ask us questions, and expect us to respond with a yes or a no, or by selecting from a short list of choices which don’t match our current situation. They apply algorithms to us, and expect us to conform to certain inputs and outputs.

Slowly, our instinct becomes one of self-surrender: we voluntarily algorithmize our own lives, if you will, so that we fit better inside their framework.

Of course, the real source of these algorithms and limitations are the designers of these systems which, in most cases, are still human. But we interact with the machine in front of us, not the person who told that machine how to behave. In this context, I’m personifying the systems, because they embody the designers’ decisions about the norms and constraints.

The Simultaneity of Square and Squishy

Machines and searchbots are a fact of life, and I’m not proposing that we all jam our shoes in their virtual gears.

The solution, it seems, is to explore the interplay between the deterministic and the chaotic, the predictable and the surprising, the explicit and the ambiguous.

There’s a quality present in many Caribbean pop songs that represents a kind of ideal to me: crisp drum machines form a structure as precise as the engineering of the chips inside of them, while above those relentless patterns, musicians add laid-back basslines, horns show up from time to time, and languid vocalists ease in and out of each entrance.

It’s lovely because it isn’t either/or: the musical interest comes from the tension between what’s on the grid, and what’s not on the grid, from the simultaneity of square and squishy.

I hear this same pattern elsewhere: In Joy Division, characterized by the contrast between the precision of Stephen Morris’ drumming and the mercurial vocals of Ian Curtis.

Or in Italian Baroque opera, as a soprano gracefully unfolds a melodic line over the tick-tock continuo of harpsichord and strings.

I see this quality, too: even Jackson Pollock used square canvases.

Systems and process provide order. It’s up to us to be a little quirky and chaotic within that, to keep it interesting.

Acknowledge the rules. Flirt with the guidelines. Follow some, avoid others. And remember: in the digital realm, conformity is built-in, and needs no allies.

The next time you feel overwhelmed by rules, how-to lists, keywords to include, tradition, convention or a statistical analysis of retweetability, please just stop.


And instead, choose to be the most interesting thing you can be: Human.