As I’ve now established my selection criteria today I got to work whittling my longlist (about 80-ish images) down to a more manageable size. I started with a simple culling based on the easiest criterion: removing the images where there are two gloves. I then weeded out images with noticeable technical or compositional issues. I was left with 54 images to work with. This is where it gets physical :-)

I printed the 54 then sorted them into three piles: Yes, Maybe and No. I figure I need 21 or 22 images in total, presuming I’m going to produce a book of a similar size to the early dummy.

Looking at the piles in the order I picked them:

No

e3fd969c-5f69-4eb4-9106-3084ba4a51e3.jpeg

This was the easiest set to sort. Once I had them all laid out, I knew which images I didn’t consider strong enough. Too many of these are similar, too many are black gloves – they aren’t distinctive enough.

Yes

db8e6c02-a163-4da2-a8bd-08b10f8ee98b.jpeg

These are the ones I was drawn to based on the criteria I’d set myself: visually distinctive, a variety of colours and styles, good composition and figure-to-ground ratio, interesting shapes (gestures).

Some were in my very first draft set from February, staying in because they still worked for me based on my selection criteria. Some were included because they particularly link to one of the text fragments. Some were shot more recently and have the intention embedded, as opposed to some of the earlier shots taken when I didn’t know why I was taking them.

There are 18 in the Yes pile so far, so I’m three or four short. This is where the Maybe pile comes in…

Maybe

4ed11a5d-58f6-435e-90cf-a1812923dcd7.jpeg

Some of these I do like, but am unsure whether they fit in the intended sequence. Some are potentially too similar to existing Yes picks, while others are potentially too different (e.g. the second on the second row where the glove is very small in the frame). There’s one that really appeals to me because it reminds me of my dog’s fur; I’d quite like to work that in. Broadly speaking, the left hand columns are closer to being Yes and the right hand columns I’m slightly less sure about, but they aren’t definitely No yet.

Anyway, this Maybe set serves two purposes: I need to select three or four from this pile to make up my 21 or 22 in total; and it’s the fallback pile in case I decide to remove an existing Yes or two during the sequencing stage.

Next steps

  • Match the Yes images to text fragments
  • Identify the gaps
  • Select from the Maybe set to fill these gaps
  • Produce draft sequence
  • Mock up images and layout