Monday, June 22, 2015

The Future of Imaging II

This is something of a followon to this post from a couple weeks ago.

Two themes I revisit here a lot are: shooting portfolios, and the gigantic river of pictures being uploaded.

In a way they seem to be related, and that's probably not an accident. Both are about the relative unimportance of the single picture in this modern age, and that is a direct result of the glut of pictures. I can't value a singleton photograph in my own body of work, in part because it's just another instance of something of which we have billions. No matter how artfully I have made the picture, no matter how perfectly it conveys my thoughts and feelings, it is nonetheless one of The Billions. This is, ultimately, why I shoot collections, portfolios, books. This is why I no longer even press the button when I see a single picture without knowing what context I can fit it in to.

So, it's all related.

My wife is in the throes of making a picture book of the first 5 years of my daughters life. I have a 5 year old and a 2 year old, and several thousand quite decent snapshots of them.

The process of making the book is proving maddening. Even companies for whom The Collection is their bread and butter (blurb, myportfolio, shutterfly, etc), the companies who Make Books as their thing, even these companies are locked into the single photograph model. You make a book by placing photographs, one by one, on pages. Picking out the photos is your problem, one by one. Then you place them, one by one, in photo-containing boxes. One-by-one you adjust the box properties and placement.

In this era of The Cloud and 4 gigahertz 8 core 64 gigs of RAM retina displayed monster computers, that creating a book of 300ish photos out of 3000ish takes dozens of hours of manual labor is beyond ridiculous. This should be an hour or two.

My wife should be able to point some software at some archives and say "Get me all the pictures that have these two kids.." "Ok, now throw away the fuzzy ones and the badly exposed ones.." "Ok, cool, now flick them past me in chronological order and let me rate them, 1 or 2 seconds each.." "Great, now pick out the top 500 and lay out a book, find the birthday parties, and give me one chapter per birthday party, and one per year.." "Ok, too many pages, try 200 photos.. Nope, 300.. " "OK, flip, flip, flip, these need to go.. that one goes earlier.. this one isn't that birthday party... Perfect. Print it."

The ratings and metadata should obviously be preserved from session to session, so the next project is easier, but starting from scratch I can't get my arms around more the 2 hours of work here. Except that the software thinks that the way you make this book is you select a photo from an archive, and place it on a page, 300 times.

But it's not really about this particular workflow.

It's about letting go of the single picture. A photograph isn't a solitary 2 dimensional image any more. First, it's a digital object, it lives in a digital context. A picasa album, a flickr stream, a facebook timeline. It has a title, comments, likes, EXIF data, a shared-on date. It's before this other picture, and after that one. It contains recognizable people and objects, Second, it's not alone. It exists as a minute speck in a vast sea of pictures, with relationships to other pictures based on all the things noted in item 1, at least.

We need tools to manage these specks, in their multitudes. We need to be able to sort them, sift them, slosh them around and manage them. Pare down the sea of infinitesimal specks down to one collection now, for one purpose, and tomorrow, for another one. We need to be able to unify the several places we keep these things into a single giant heap, and then to reach into that heap with both hands, and in a moment pull out fistfuls of the right ones, for right now. This is a solvable problem. It's not even very hard.

Incidently, google Photos does a lot of this now. Given that google is pathologically incapable of keeping their eye on the ball it is a near certainty that this will not actually go anywhere, but they have demonstrated that the basic sorting/sifting problem is solvable, indeed, solved.

1 comment:

  1. Photobook software workflow? Why are you, a software monkey, surprised at all? It is what you get when you let clueless children build software. [I carefully left out the word "design" as most of what we see these days looks as if it was made up as they went along.

    Lets hope they don't move on to making bridges and skyscrapers.