The “Magic of Tidying for Photos” App — Introduction

This post shares about the idea of applying Kon Marie organising method to your digital photos, and I developed an IOS app with AppleML and PhotoKits to solve the problem. I share my reflection on lean product development strategy and my journey of making the app end-to-end using a minimum, iterable tech stack and strategy.

It started with our own problem with managing photos on our phones (which, we found out that many other people also had). As parents with young kids, or more accurately, as humans who own smartphones that have cameras, we co-produce gigantic loads of photos. Phone storage is outgrown so easily (remember when16GBs was enough, then 32GBs, then 128GBs. No, they never really are).

How do people back up their phone photos? Our private Facebook or Instagram photo albums help a bit with curating the best shots we want to remember, but quite limited. There are definitely more photos that we want to keep, and we do not wish to upload many of them to these platforms.

Cloud photo storage solutions (like Google Photos) can be a back up solution, they keep photos safe. The problem with them is, these 100’s of GBs of photos will be stale, in other words, 99% photos have very little chance of us browsing through them again.

The problem of putting everything in the “garage”

Anyone that has a physical garage would know this “garage hoarding” problem — stuff that you stash in the garage that you hope will be useful some day. In reality, many of them will just sit there covered with dust forever.

Piling up stuff in the “garage” isn’t good for your soul. Photo by NeONBRAND on Unsplash

Come back to the photos.

We believe the fundamental solution is to keep our photos tidy and in manageable number. And so will be our memories.

In fact, a quick review of our recent photos easily shows the 90% — 10% line. 90% of photos are random snaps, spammy shots, the “9 other blurry copies”, cute-but-not-really-needed-to-keep ones. The 10% that worth keeping are the special occasions, the birthdays, the new picnic places, the unique moments. Plus occasional random shots here and there would be fine, too.

Indeed, if we could find the 10% we want to keep and clean up the rest, a quick calculation would show my phone storage would be full only after 7 years. And everything can stay on my device.

What is the problem we want to solve?

Each time the phone is full — we face that dreaded question — should we back up all these Gigabytes of photos? Often, we have to — because deleting everything is not an option. Selectively cleaning up thousands of photos is impossible.

Cleaning-up is best done on a regular basis. Some even say it is the only possible way. If you leave your kitchen untidy for a week, it is 10x as hard to tidy up.

We acknowledge that there will be people who simply ain’t interested in tidying or organising stuff . For those who are, there will be benefits far beyond reclaiming storage and keeping your photos in check. After all, people bought a best-selling book to learn just about tidying.

The Marie Kondo way

Marie Kondo, the author of “The life-changing magic of tidying up” (who admittedly was on trend an eternity ago) offered some parallel thoughts with her trademark KonMarie philosophy. To simplify your life, so Kondo suggested, you either put your possessions to use, or you should let them go.

Removing stuff, while counter-intuitive to human possessive tendency, helps you actually keep (stashing stuff in garage doesn’t count) more stuff by focusing only on those that matter. Consequently, it gives you the joy of living in the present.

The idea

As all know, once-off clean up is easy. Building a habit is really hard. Or else, the book would not be a best seller. Just look at kitchens or cupboards. This is where the idea of an app came in. How can it help?

• A reminder
• A “Swiss knife” photo clean up utility built in with KonMarie principle — the 90% -10% principle (explained above)
• An AI assistant that help suggest what to keep, based on intelligent image recognition (for instance, photos with faces, or non-blurry photos, on average, are more valuable) and understanding your preference
• A habit builder (via incentives and clear goal settings )
• A safe, friction-less way to clean up your photos with plenty of safety and reversibility measures in place.

The last point proved to be extremely important, as I learned after validating the product, which will be elaborated in the product development session.

This is how the end product looks like — I have published an app to Apple’s App Store. If you are keen to try the app itself, please visit https://three-a-day.app or get it from the Apple App Store.

Starting out

At my current workplace (ThoughtWorks Australia), I am blessed to be embedded in a place with excellent innovation and learning culture. (On a side note, with the current pandemic has been harsh for many people — I have to admit we are really fortunate to still have opportunities to work and learn during this difficult time).

People here love trying and building innovative things, especially in the data & AI/ML space. Our amazing colleagues have produced really cool stuff, from real-time AI-powered sock sorting using event streaming and transfer learning, to the “Strava for Wheelie” project using advanced analytics to guide improvement, to counting bees using deep learning on edge device.

While we all certainly had plenty of fun (you can be sure by the names), we love such projects also because of the learning opportunities. They afford us sharp opinions from a practical experience perspective.

In this series of blog posts, I am sharing my learning from this side project. I reflect specifically on applying product thinking, finding a suitable lean strategy that allowed me to finish the app end-to- end in tiny chunks (it took many months), AI/ML and image tech in IOS app development, and the minimum tech stack needed for get my app off the ground.

• Part 2 —Product thinking: on lean, setting up the project to fail fast, finding an iterable strategy, and finding suitable product goals
• Part 3 — Technical considerations and the minimum tech stack to finish the app end to end: tools, techniques, architecture, Swift, Apple ML and Photos Kit, App publishing and performance tuning
• Part 4— Final tips and thoughts— asking for help when you need, using external resources, and how to not let exciting projects consume your life