Have been determined to get this one done before "going public again," so it's finished with some sense of relief. After struggling on and off with it for years.
Shot April 2006 in Lexington, with the 10D and 24mm f2.8 lens, 1/1600th at f/4 and ISO 100. 24" print.
Obviously, the size really matters since the "action" is so small. Try to imagine two feet tall, and here's some print detail to help.
Some in-depth thoughts about reading time in composition and minor processing notes behind the fold...
First off, it always bears mentioning that elevation increases the likelihood of a dramatic shot. In urban settings, I've become a big fan of parking garages!
On the other hand, this one is very rare for me for a couple of reasons. Not prone to shooting random people (because I don't care to be shot myself), so you won't see much like this from me. But much more importantly, this shot was flipped. It obviously needed to be for the whole composition to tell its story.
Pardon me for another snob moment, but I am stunned at so much photography that's backward. Properly, the past should be to the left, and the future to the right. If forward-actions are aimed leftward in your image, it's already broken down on a fundamental sensical level. An image fighting against this will look at least substantially "not right." (Airplane wings in flight and cyclists in motion to the left are classic examples.)
(If you disagree with me, cool, let's talk about it. But if you've never thought about it, trust me - I've been scoping this out for years. And I don't know about the Asians, but even the Arabs "read" time in our direction.)
In this image for example, let's zoom in on this print detail below, but reverse it for the moment, back to the direction of the original shot.
So what's "wrong" with this? How about, "nearly everything?"
The major "arrowhead" in the composition is pointing "backward." The bottom dark wedge takes on much more substance, when the bright arrow should clearly be the center of attention. In this original orientation, the arrowhead becomes much less prominent altogether - the shadow over the top arm disrupts it more forcefully - and the sunbeam/sidewalk continuity seems to become a simpler path straight across the image.
The cyclist is pondering the threshold of the street, in his past?, while the path of his prior journey is "ahead" of him? Granted, our own minds "decode" this as best we can by the direction he's facing (just as we do with any cyclist heading the wrong way), but it just doesn't look right.
Now, granted, there are a couple of implications that are fine in this direction. In particular, the sun shining this way might be more natural than the other direction. (However, would you notice that it changes the implication to morning? And would that be as appropriate for facing a crossroads in his life? Are you starting to see the underlying gravity of all this?)
So now we're to the question of the story itself. That is, there's a woman walking in his direction, on not-quite-but-nearly a collision course, an opportunity to meet each other or pass like ships in the night. Obviously (from my title), I see two conjoined stories of interest in this image, their potential for meeting, and his own decision-making process.
As far as he goes, aiming to the right is a no-brainer. As far as they go, had to think a lot harder about it. Because each orientation tells a different story. The easiest way to think about it is that "left is premise and right is punchline."
And this holds up even when the viewer's eye is drawn to the right first - they still unconsciously perceive that they've "skipped to the end" and then backtrack to register the set-up. (I bet a critical image on this years ago and was amazed when everyone seemed to "read it right," even though the eye was virtually forced to go to the right first.)
So, in this case, her approaching from the right (in the original shot) wasn't without narrative benefit. He's sitting there at loose ends, and doesn't see that a potentially game-changing variable is walking in his direction. I.e., "the punchline." That would entirely make sense, and if not for the huge problem of him resting at the threshold of the past, I might well have stuck with it (for two reasons described further below).
But on the other hand, I didn't really see much loss to their story by flipping it. She can certainly serve as "the premise" herself, especially since our eye is probably drawn to him first anyway (thanks to the massive arrowhead itself, of course). Now he and his indecision become the punchline, which might actually be stronger - she's just trotting along and might change someone else's life, and lo and behold, there's someone's life that might be waiting for a change!
In the original orientation, not sure "her own life" seemed as important. She may have been more depersonified when she was only a punchline. I feel more subjective about this specific point, but she seems to have more personality to me when she also is now pointed toward the future.
So I think you see my logic now, but some other points bear mentioning.
One more about composition specifically (and in favor of flipping this image specifically): if you've got open air on one side, and any substantial "barrier" on the other, you almost certainly want the barrier on the right. There's no question that open air to the right allows the viewer's eye to track across and then just keep on going, to the next work of art or into the next room or out to the coffee shop or wherever. A barrier on the right bounces the eye back into the shot and encourages the eye to wander within, rather than away.
All this theory of composition can be tested - you're welcome to reach your own conclusions. But my experience is that it will hold up both within your own work and especially when evaluating the work of others. I extrapolated the theory from McCloud, and then set off on my massive commenting binge at dA in '03-04, and was quite amazed at the consistency of it. It seemed like every successful image I found was "playing by these rules," and every image that failed to live up to itself was thwarted along these lines. (Granted, many images don't test these rules in either direction. On the other hand, some fail in both directions. D'oh!)
But it must be said that flipping images (other than "abstracts" like clouds or gnarls of wildlife) is very risky business. Faces don't flip well at all (other than profiles, which sometimes get away with it). And generally, I think we're more prone to shooting it right to begin with, and flipping will look wrong on some intuitive level. That is, if we're "reading the world" when we shoot, then these same factors probably inform our eye. (But then again, if that's true, what's up with all the bicycles into the past?) All I can attest here is that I frequently saw an image at dA that seemed flipped, and usually turned out to be right. (I.e., the artist would respond, "How did you know?!?")
So didn't want to flip this image for two reasons, that above for one. But even worse, this image was populated with street signs! I had to use Photoshop's Vanishing Point tool to scrupulously flip all of them around, except one that ends up remaining as a subtle gag. Have been tinkering on this shot for years now - probably buckled down and did the reversals sometime last summer.
Finally, in the "every little bit counts" department, the street lamp in the corner was originally falling out of my frame. I look at problems like that and try my best to tolerate them (rather than hassle myself), but ultimately it just looked amateurish. So found another shot from a moment earlier that including the whole post, and cloned it into this one to pull it all "in bounds."
Ultimately, a really difficult image. An awful composition to manage on a cinema-ratio monitor, for one thing.
And don't think I've ever worked on anything so busy, with such conficts for attention, nor with the critical subjects so small. The tree isn't meant to be "the story" (though I like how it almost seems to be warning him of danger - go! GO!), but it is meant to be "the eye candy," so it was really tricky to establish its proper prominence.
Of course, almost all the shadows were much deeper, so pulling out of them naturally was a tricky balancing act. (I.e., too easy to pull too much.) But my personal development goal is to recreate the experience of the human eye, which can pull detail out of shadow whenever it likes. (And no, I don't ever use Shadow/Highlight - my strategy owes to the same underlying algorithms, but "by hand" and more versatile.)
So, had drafted this one twice before and never been happy with it, but think third time was the charm here. (Giving up on green in the tree was critical, as it turns out.) Not sure it's "great," but certainly very different and unusual in the context of my own work.
And just based on the title/premise/concept, it's always been really important for me to do, and get done. I feel like I'm at that threshold myself rather often.