I've long held to the opinion that what makes a great image maker is his or her ability to control what the viewer interprets from his or her images. As image makers we need to strive to "control the mind", sort to speak. I'm still working on that skill myself. When we point our cameras, we must be able to see the story in the frame and capture a composition that works for the story. When we fabricate an image in 3D software, we need to be able to see the story in our mind and build the composition, which involves all sorts of distracting technical details. Whether by camera or software, if we don't see the story before the image, we get what's called "happy accidents". It's good to have lots of happy accidents, but we also want to be able to control what the image means to viewers. When we can control the viewer's interpretation of our images we are better able to deliver images that rise above the untold millions of images that get produced in these digital daze. But we need to know the what meaning we're after if we're to control the meaning to others.
I don't hold to needing to have a story before we take the photo or create the image, but I believe that the author must know the story by the time a photo or image gets viewed beyond the author. As you described, the image evolved as you pursued various compositional experiments, such as blending the water line into the clouds. So, compositionally speaking your image evolved. But you still had a story in mind, whether you knew it or not. Something wanted to be expressed that took you down what seems to be a meandering design path. The idea then is to be able to articulate what that story was or became. By being able to express the story (your interpretation) of the image, you are able to keep the technicals in their place – good technicals do not make a meaningful image, rather help a image's meaning not be obscured by clumsy, distracting or ugly elements. When improperly used, HDRI is a good example of where the technical obscures the image's meaning, where HDRI becomes a process rather than a technique which all too often results in images that are not believable or downright ugly. Our 3D software is also a candidate for allowing the technical to come before the meaning. If, whether doing HDRI or 3D, we can keep the image's meaning foremost in mind we'll produce better images that will survive the test of viewer interpretation. But, even if the image development evolved, if the author can state its *intended* meaning, the image will be more likely to be meaningful to another in a way that is consistent with the point of the image.
QuoteYes, I know my 'final image' doesn't really fit the title of "The Forever Tree' as much as the first image did, at is it more crowded in, almost threatening to be choked out of existence by the dark clouds around it. Perhaps, 'The Lonely Tree' or something similar might be a better name for the darker image.
A couple points if I may. First, try to not design an image to fit a title. Let the title come last, if at all. Titles have the unfortunate consequence of suggesting an image's meaning. Leave an image untitled for maximum freedom of interpretation by viewers. When we sell our images, we want the buyer to believe what they want to believe, which a title defeats unless carefully chosen. Yes, we want to lead the buyer to believe that which is consistent with our intention for the image, but we want to do that with composition, not titling. Second, there is nothing wrong with choking an element...if that's what you intend to do. But be sure always to give strong visual clues as to what the primary element is (the tree in this case). For instance, bring the tree closer to the foreground to give it dominance in the scene. Make the tree bigger. More color to the tree. *Center* the tree (which breaks conventional compositional rules, but does clue the viewer that that's the primary element.) A busy image is ok as long as the viewer can identify the primary element in which case the viewer is "grounded". When grounded, the viewer can take it from there, else the viewer is left visually floating around the image confused. We have about 5 seconds to win the mind of the viewer. Use that precious time well.
-Pat
