Ferry started with me being forced to use macOS at my current workplace, which suddenly replaced a working environment I'd been using during professional career as a programmer: Windows for around 10 years, then Debian/KDE for the last 15 years.
Over that time, I'd grown very accustomed to Total Commander on Windows, and later Dolphin on Linux as a file manager, plus a collection of small but very useful tools I'd written for myself on Linux. Moving to macOS was quite an adjustment. I really struggled with Finder and the rather fragmented ecosystem of apps, especially when I couldn't find tools that covered the things I was used to having at my fingertips.
Having already written a few internal tools in Swift at work, I started to get familiar with the language while trying out several file managers for macOS without finding one that really suited all of my needs... So one thing led to another, and I eventually decided to build my own version of a better Finder (that's probably how most of the macOS file managers started their journey).
Obviously, the first versions were essentially a combination of a slightly improved dual-pane Finder spiced up with basic features from TC, but t'was more than good enough for me. Once I started sharing it with coworkers and friends they began suggesting features I hadn't even considered initially, many of which turned out to be genuinely useful.
I know there are quite a lot of file managers out there, so a "what makes Ferry different?" is a totally valid question.
Well, Ferry first and foremost tries to be as less opinionated and as much customizable as it's possible, so I have included A LOT of options for personalization. Eg. the layout, the UI, the menus, or the shortcuts are adjustable in many ways, you could even put together something which more or less resembles an old-school Norton Commander-style file manager with a modern toolset if that's your thing.
The app also includes a rather big collection of handy tools and tidbits which probably don't seem too interesting up until the point you run into a specific trivial task, and then you realize you have to search and install a tool for it.
Main features list: https://ferry.indus3.org/#ftrs
Another aspect that's important to me is how the app is made. Having been really involved in the demoscene, I wanted to handcraft as much code as possible and keep AI involvement to a minimum, while still using it to improve quality. Most of Ferry is still written manually, AI mainly helps with the boring parts: proofreading and translations, debugging, tests, temporary tools, code reviews and the site. LLMs have become so advanced that I'm honestly not sure whether this is a good or bad thing for software quality. My skills probably aren't quite on par with a modern LLM agent, but hopefully I can still build something useful and pragmatic by doing most of the work myself.
Licensing follows TC's approach: there's a reminder, but you can keep using the app for as long as you like (or until the reminder gets too annoying). Hopefully that's reasonable enough. But if I eventually become really happy with the codebase, I'll most likely release the source code too.
At this point, since there are plenty of features, I'm becoming much less interested in adding even more of them and rather focused on getting the existing ones as good as possible: size, stability, performance, memory consumption and above all, bugfixes and code quality. Of course I still listen to users, and if someone suggests a feature which is practical and it improves the UX, I try to include it into a future release.
Feel free to contact us, as suggestion and criticism is always welcome!