One of the reasons I’ve been somewhat missing from the blog is that my photo processing workflow imploded — I came to realize it was broken beyond repair, and I didn’t know how to fix it.

That’s not a fun place to be.

The final straw was trying to integrate some more complicated processing techniques into the workflow, specifically handling multi-image processing for panoramas and HDR. The way I had everything set up in Lightroom just didn’t work for managing all of the pieces well, and everything I tried — well, all of the solutions were ugly and I realized they wouldn’t scale.

Ultimately I came to realize a decision I made when I first migrated to Lightroom was the failure point; I made a decision to use collections to store groups of photos instead of folders. Collections are a virtual grouping, folders are a physical grouping. I felt it made sense to import into a YYYY/MM/DDDD folder, and then use collections to pull related images together. Overall, that worked well (for a while).

Lightroom, however, has a — quirk — a design decision that is impacted by this, and that’s how sets can be used. Sets is another virtual collection that work within folders, but sets are incompatible with collections. that means when you pull everything together, you have to chose collections or sets (but not both). I chose collections. That works, until you need sets. Then all hell breaks loose. It really does make sense to use a set to pull all of those pieces together and tag them with the resulting image as the top image.

Unfortunately, you can’t do that if you use collections. sigh.

In researching options on how to do this (and more importantly, how to do this without tearing it down to ground zero and starting over), I finally decided the workflow I liked best was one outlined by Hal Schmitt at Digital Photo Experience as part of his Panorama screencast. But that meant — of course — starting from ground zero.

So I finally decided I needed to, and I’ve been spending my evenings recently taking everything in my Lightroom libraries and converting all of the collections to folders, one at a time. Of course, once you decide to open up the hood, you don’t just fix what’s broken, you start tinkering, and I did, restructuring my keywords, rethinking a few things in my metadata presets. Little things that flit in and tweak everything to some degree.

This, by the way, makes Time Machine crazy. That reminds me that I need to start planning to upgrade my disks to larger sizes soon. This means I have to think about my backup policies, and… and down the rabbit hole we go again. Fortunately I have a couple of months before I have to worry about the disks, and I’ve got everything back under control (well, mostly. I have a couple of thousand photos flagged with special keywords defining various “needs to be looked at and fixed” to-dos).

I’m happy with the structure of the files on disk and how the workflow gets me from import to flickr, and with the keywording and metadata (to a point; there’s more detail that I’m still thinking through and implementing, that’s the “to do” on a bunch of images…).

What I haven’t yet done is take it from “post to flickr” stage to the full portfolio, but that’s the part I’m starting to work on now. Most on that, hopefully soon…