It's been ten months since I started photographing for my Convolution project, and while I've mentioned it several times here, I've never used its name in a blog post title. Funny. It's almost as if I was waiting for it to be finished or something.
It's not finished.
I'm also surprised by how comparitively little I've written about something that's become a huge project for me. I thought I was trying not to be too single-note with it, but somehow that sense of inhibition didn't stop me from turning this into a blogsite about pinhole cameras. Clearly I have some issues about this work that I'm still trying to work through; the fact that it has taken me two weeks to put together this simple post is also a bit of a clue.
I know what some of those issues are, but don't want to write about them. So instead here are links to my previous posts about the project:
Somehow even registering its own domain – convolution dot ca – for the project wasn't worth mentioning on the blog, despite it being the first time I ever did that. The domain name might eventually direct to a free-standing site, but currently drops people right into the photo gallery without any sort of warning or cushion. There's a Frequently Asked Questions page, too, but it's intentionally hard to find unless you've seen at least a few of the photos first.
At the end of May, ten months into the project, I did a major revamp of all of the existing photos. While the way I record the images has remained remarkably unchanged since the end of July 2020, how I digitize the film and edit the results has improved. So I have revisited all of the photos, removing a half-dozen from the set completely, going back to the original scans to find candidates that I'd overlooked previously, and making the tonal edits a bit more coherent from earliest to newest.
(It's weirdly important to me that people know that these photos are all taken on film, with each 35mm roll of Ilford Ortho shot over and over again, predominantly in toy panoramic cameras. These are not after-the-fact digital composites, with juxtapositions retroactively clevered into existence. I realize that my process of curated accidents might not matter to anyone else; I know plenty of photographers who enjoy slapping that big shiny "Make Art" button in their editing preset program of choice and think their results are perfectly valid examples of creativity. Whatever floats your boat, I suppose, but this isn't that.)
I also fixed some holes in the photo numbering sequence that makes up the titles of my re-merged scans, which was awkward but necessary. The wholesale renaming of the photos breaks a lot of mental links, and means that if I've ever mentioned a photo by its sequence number there's no way to go back and figure out which one I meant. Sorry. But better to do that now, as I prepare for the one-year anniversary of the project, than to just leave it as it was.
And for the sake of comprehensiveness, I'll also note that there are now an even one hundred photos in the Convolution gallery. Yay! That hundredth photo is #319, which is the lead image for this post. (You can see the enthusiasm.) The first image was #100, so that means my 'keeper rate' after the preliminary edits isn't too far off from the 5:2 aspect ratio that I use for these photos. Fitting.
I'll also point out that, with one exception, I don't manipulate the sequence numbers.
Each roll of film comes out of the developer as a continuously-exposed strip without interruption, which I then need to scan in overlapping 3:2 chunks. When I see an interesting section in the scans I'll use a panorama merge tool to reassemble that part of the negative, and those are saved as new files with the Convolution-Sequence file name. The number is determined by wherever the "sort by date" function in Capture One puts it – using some occult parsing of the creation dates of the original scans, somehow, not when I exported the merged file from Affinity Photo. So each roll of film will create seventy or eighty overlapping frames, from which I'll find five or ten interesting candidates to merge, and then maybe post about half of them. So the gaps in the numbers, or their proximity, suggests where they were on that particular roll of film, but says nothing about how far apart they were from each other. Sequential numbers might be widely separated, while ones with a gap between them might actually have overlapped on the negative.
Again, I don't know why it matters to me that people know this, but for some reason it does.
And that one time I played around with the sequence numbers? There's no photo titled "Convolution 152". Skipping a digit means that it's Convolution 153, seen above, that incorporates the building at 53 Fraser Avenue.
That's important to me – but, for once, I won't go on and on about why.