Sometimes, simpler is better (in picture editing)

A Journalist’s Diary

A display of a DSLR camera and a paper notebook sitting on a camera bag.

Little did I know that the software tools central to my Cost-Benefit Analysis late last year would change only days after I had finished my evaluation.

However, the announcement that Photo Mechanic would soon transition to a subscription model coincided with the observation that my efforts to tame my photography archive had more rough edges than I wanted to see.

While I’m not against software subscriptions per se, I used the news to evaluate which programs I needed in my toolbox. Maybe a change or simplification could help me with those rough edges.

Today, I’ll share how I approach subscriptions, write about the problems I noticed in my workflow, and give you the actions I’ve taken to get my archive sorted.

Efficient speed machine v comfortable do-it-all cruiser

I thought I had it figured out. I looked after keywords and knew how I wanted to embed location data, people’s names, and other more basic metadata. I wrote about how I had kept my picture archive organized before.

I think it was around the time of writing that post that I started working through my archive to bring all images up to date with that approach.

I started in the present and worked backward. The goals were cohesion and using the reliability of well-structured metadata to build collections for easy access to specific images.

I decided to split the work between two pieces of software. I started editing in Photo Mechanic and used Lightroom Classic for toning and as a final resting place.

Photo Mechanic felt more efficient for the bulk of the editing. The software is highly regarded in documentary circles or anywhere where editing efficiency is valuable.

I felt comfortable with that set-up. I made decent progress with my archival updates.

But there were underlying issues I tended to ignore because I felt comfortable enough. More on those issues follows in a bit.

Do I need a subscription?

Years ago, I lost a writing tool to a subscription switch. I didn’t stop writing with Ulysses because it required a subscription in the future. I stopped because I disliked how the company introduced that subscription, which caused me to look at how much I needed that software.

I concluded it was not worth a regular payment because other tools felt just as comfortable and efficient.

I’m not going to add endless arguments for and against subscription software. To me, a subscription can be worthwhile and can make sense for both sides.

But given that there is an increasing stream of coffee a month we are asked to pay, I evaluate if that coffee has enough value for me.

When the “Photo Mechanic subscription future” was announced, it felt as bodged and unprepared as it felt in the case of Ulysses (but that’s a different discussion).

There’s been a change to the company’s plan, but I’m not writing to debate if going subscription was a good or bad decision.

I write these lines because the announcement made me confront the imperfections in my workflow. And while I moved away from the two software solution, that’s only the by-product.

While I invested lots of time and energy into maintaining keyword cohesion and location clarity, how one works with that metadata in Photo Mechanic still caused more confusion than I was willing to see.

The outlook of that additional coffee a month (or whatever Photo Mechanic might end up costing once the dust of the announcement has settled…) trained my thoughts on those imperfections and made me consider if my way was still the best. Put differently, I realized I needed to look closely to decide if that “coffee on the horizon” would be worth ordering.

Process Refinements

I hadn’t made much progress with my archive lately and was still stuck with 2022 (I believe). The reason, I believe, was those rough edges.

I still dreaded applying keywords and deciding how to tackle location information. Have I considered everything, or have I missed something? Do I need a keyword for this? How detailed do I want the location info to be, and is that exactly how I handled it in older pictures from the same location?

The more I progressed, the more I learned about the difference between the theory and practical benefits of metadata use.

Instead of clarity and ease of use, I struggled with knowing how consistent my keywording was and with an endless list of locations that grew continuously.

Using two pieces of software instead of one also adds complexity.

Easy in theory, confusing in reality: controlled keyword lists are still confusing and time-consuming if one has to pull up indvidual lists, which is how Photo Mechanic operates in the various places with access to keywords (1 and 2). Knowing what keywords are applied to an image and removing unwanted keywords was the most complex part of the process because it can only be done on one image at a time. Side-note: the only downside to using Lightroom’s keword lists is that the “parent” keywords that create order are visible when the RAW files are being viewed outisde of Lightroom. I find that a small price to pay for the ease of use. Lightroom can ingore them and exported images don’t include them (3).

Photo Mechanic auto completes text fields such as locations after the first three characters, but that’s not of much help if your list of locations gets confusingly long… (though, I wish Lightroom had a similiar feature).

I always thought Lightroom Classic was bad at those dreaded editing tasks.

My evaluation, however, surprised me.

I realized that using the keyword list feature gives me a simplified approach to keywords where I can apply and remove keywords with a mouse click and see at a glance if my keywording strategy gets out of hand.

While all the other refinements are independent of software, I don’t think I could find the same ease about applying keywords in Photo Mechanic.

It felt more efficient than ever before. That’s a conclusion I wasn’t expecting.

Lightroom also offers several ways to manage keywords. The keyword list section is the most logical to me. I find it much more efficient and assuring to go through that section and click on keywords I need to apply. It’s also an easy place to get an overview of one’s keywording strategy and if that needs adjustments because it shows the number each keyword has been used, and you can jump to a filtered view for each keyword. Lightroom’s “persons shown” meta data field is not very accessible (and not included in the software’s search and filter capabilities), so I add the data via keywords. Not ideal, but still good enough and easy enough to work with.

I knew there wasn’t much of a difference in renaming pictures. However, I was curious to compare and contrast the remaining tasks.

Location data was the next area I looked into. To be fair, Photo Mechanic makes that easier. And the simplification I implemented has nothing to do with the software.

But in the grand scheme, I can live with how I need to input location data in Lightroom.

The only way to store a location database in Lightroom would be to create metadata presets, which would result in an awkwardly long dropdown menu. I simplified my approach (which I could have done with Photo Mechanic just as easily) and for ease of use and consistency keep a spreadsheet, which I can use to copy & paste the data quickly. For my needs, that’s sufficient. Usually, I modify the one preset I have and apply the location data in bulk. If I spot an error, I use the metadata panel to correct it.

Lightroom’s caption field needs work. For some reason, I have to activate the spell-checker every time. Worst of all, when I start typing and hit a letter that’s also a shortcut somewhere else in Lightroom Classic, it tends to kick me out of the caption and sends me to some part of the software associated with that shortcut. That’s annoying.

However, I like to see the final state of the image when writing captions. So, I almost always wrote captions in Lightroom anyway.

Lightroom is too slow (especially with importing and getting the images ready for inspection). That was the consensus in the past. And yes, it can be cumbersome at times (especially with older hardware).

But color me surprised when I found no difference in the time that passed from starting the import process until I could begin culling the images.

One feature I miss but have learned to work around is Photo Mechanic’s variables. Pictured here is the “base caption” I applied to the entire contact sheet (after replacing “base caption” with a quick sentence that was “good enough” for a start. The words in brackets are variables. Photo Mechanic automatically pulls the corresponding metadata from the files, meaning you don’t need to write out the date and location.

To cut this short, I don’t need all the advanced tools in Photo Mechanic (like code replacements or built-in FTP upload, for example), prefer applying keywords in Lightroom, and can live with how I have to handle location data, people’s names, captions, and culling.

Lightroom Classic is good enough for my needs. Sometimes, simpler is better.

Bonus benefit: the archive is making progress

Since I came to that conclusion, I ramped up my archival overhaul procedures again.

I opted for a clean slate and created a new catalog. I’m taking it one folder at a time, taking out metadata I don’t want any longer before adding that folder to the new catalog. I then apply the simplified metadata from within the new catalog.

That process guarantees a keyword list that grows organically, even though it’s a bit more work.

I’ve now got 2023 fully updated and under control. Everything new goes straight into the new catalog. And I’ll work my way backward until the old catalog is empty.

It all feels more intuitive than ever before. I’m happy, even though I’m slightly sad to let go of Photo Mechanic, which I decided to do because of efficiency and what I need then because of the upcoming subscription.

As a side-note: I moved from color tags as main sorting tool to star ratings. That’s again independent of the software, but I enjoy the difference. I now use color tags for special references, say purple for composites (focus stacks or panoramas). I use flags only temporary to remind myself of pictures that I might want to include in projects that aren’t finished yet.

As a final note on using AI in picture editing, I don’t think AI tools are the solution for easier keyword applications or culling and sorting.

Yes, it would remove dreaded tasks. But would that be worthwhile?

If I can’t trust that keywords are applied consistently and for the things I want/ need to be tagged, what use would I get out of them? Some tasks just cannot be automated. And maybe that’s okay.

As for culling, given the huge impact that image selection has on the ethics of stories, it shouldn’t even be a question in journalistic circles.

But going beyond that, there’s monumental value in “manual” editing, in looking at one’s contact sheet, in seeing how one worked through any given moment. It would be devolution if we gave that up, no matter how tempting it may be.

But that’s just my two cents as someone skeptical about seeing AI as some kind of magic wand humanity can simply throw at everything.

That’s it for this week. Stop by next week for fresh reading recommendations and an update on my creative writing endeavors.