I've been using Lightroom since the 2006 public beta.

That's not a flex. That's just how long it's been a load-bearing infrastructure in my creative life. Twenty years of muscle memory. 100k+ photos. Keywords, collections, a catalog that remembers more about my photography than I do.

So when I say I'm leaving, understand: this isn't a hot take or a rage-quit. It's more like selling a house you've outgrown. You don't hate the house. You just notice, one day, that you've been renovating around its foundations for years — and the foundations aren't yours any longer.

The Catalog Problem

Here's the thing about Lightroom Classic: the catalog is the brain. Your ratings, your keywords, your collections, your develop history — it all lives inside a proprietary SQLite database that Adobe controls the schema of. Yes, you can enable "Automatically write changes into XMP." But even then, the catalog remains the source of truth. The XMP is a copy. A courtesy. And in this there's actually a grievance: Adobe embeds the sidecar info into non-proprietary raw files — a nightmare for backup schemes.

Stop paying the subscription, and you can still open the catalog in read-only mode. You can browse your photos. You just can't develop them anymore. Your creative decisions are there, visible, locked behind glass.

I lived with this for a long time. It was the cost of doing business.

Files as Ground Truth

The answer I kept coming back to was simple: the files should be the source of truth. Not the database.

XMP sidecars already exist. Adobe invented the format. Every rating, every keyword, every develop setting can live in a plain text XML file right next to the RAW. Portable. Readable. Yours.

The database should be a cache — a fast index built from the files on disk. Delete the database, rebuild from XMP. No data lost. No lock-in. The files are the archive.

This isn't a radical idea. It's how version control works. It's how the filesystem was always supposed to work. Lightroom just convinced us we needed a middleman.

Why Not Switch?

There are good alternatives. Capture One is excellent. Photo Mechanic is fast. Darktable is free.

But switching tools means adapting to someone else's architecture. Someone else's assumptions about how photos should be organized, processed, stored. After twenty years, I had opinions about all of those things.

So I built my own.

I wanted to answer the question: if you designed a photography workflow from scratch today — knowing what Apple Silicon can do, knowing what you know after tens and hundreds of thousands of photos — what would you build?

Two Apps, Not One

The first decision was to separate organization from processing. Lightroom merges them into one monolithic application. That made sense in 2006 when the alternative was Bridge + Camera Raw, and nobody wanted two apps.

But merging them creates coupling. The catalog has to understand develop settings. The develop module has to understand collections. Everything touches everything.

I went the other way:

Photo Archive Pro handles the library. Point it at folders on disk. It scans, indexes, and lets you rate, keyword, and organize — with smart collections, hierarchical keywords, and cascading filters. No import process. No catalog migration. Just your folders and XMP sidecars.

Photo Developer handles the RAW processing. Open a file, edit it, save the settings to an XMP sidecar. Eighteen custom Metal shaders. AI-powered masks via Apple's Vision framework. Film emulation presets. Everything runs on the GPU.

Both apps talk to each other through XMP. The same XMP that Lightroom reads. The same XMP that any other tool can read.

Some concrete things that feel different:

Metadata stays with the files. Rate a photo in Photo Archive Pro, open it in Photo Developer — the rating is there. Edit it in Photo Developer, switch back to Photo Archive Pro — the develop badge appears, the staleness indicator tells you if the JPEG export is outdated. All via XMP. No sync service. No cloud. No subscription verifying your license.

Lightroom shortcuts still work. 1–5 for ratings. P/U/X for pick/unflag/reject. R for crop. Cmd+E for external editor. This was non-negotiable. Twenty years of muscle memory doesn't retrain overnight.

What I Gave Up

Pretending otherwise would be dishonest.

Lightroom's virtual copies have no direct equivalent. In a catalog-centric world, virtual copies are elegant — zero-cost creative branches stored as metadata entries. In a file-centric world, you use snapshots (Photo Developer supports five per image) or explicit exports. It's different. Not worse, not better — different.

Lightroom's ecosystem is vast. Presets from every photographer on the internet. Plugin support. Deep integration with Photoshop. I don't have that. I have seventeen built-in film presets, a growing user preset system, and the freedom to build exactly what I need.

I'm okay with leaving that behind, because the things I gained matter more to me.

The Deadline

My Adobe subscription renews in April.

Between now and then, I'm doing what any responsible developer would do: dogfooding. Using my own tools on real work. Finding the edge cases. Fixing the deal-breakers. The apps are in alpha — honest-to-god alpha, not marketing alpha — and every day of actual use surfaces something new.

The tens of thousands of photos are already in Photo Archive Pro. The XMP sidecars are already being read and written. The Lightroom collections have already been migrated.

The question isn't whether the tools work. They do. The question is whether they work well enough to replace twenty years of muscle memory — and what has evolved into a more than capable tool from Adobe — without friction.

I'll let you know in April.

The Bigger Picture

This is about more than photography software. It's about the relationship between tools and the people who use them.

Lightroom was revolutionary. It shaped how an entire generation thinks about digital photography. Non-destructive editing, metadata discipline, the develop module — these concepts changed everything.

But revolutionary tools have a way of becoming infrastructure. And infrastructure has a way of becoming dependency. And dependency has a way of becoming lock-in.

The file-based approach isn't revolutionary. It's conservative, in the original sense: it conserves your ability to leave. Your photos, your metadata, your creative decisions — stored in open formats, on your own disk, readable by any tool that understands XMP.

Weniger, aber besser. Less, but better.

Photo Archive Pro and Photo Developer are part of the Photo Suite — four focused apps for the complete photography workflow on macOS. All are in active alpha development. No subscription.