Original Innovations of AeroFTP

[EhudKirsh]

There are several aspects that make and will make AeroFTP unique. Some of these are ideas that:

• you @axpnet come up with over time by yourself.
• you come up with in collaboration with contributor feedback like the aftp alias and AeroVault V3 & V4 design.
• contributors come up with.
• are kind of obvious, but were never done (ideally) before. For example, coding Rsync for Windows and having a dual panel file manager on PC that bisyncs as many cloud drives as possible like RcloneView 🦉 and others can, with wrappers/overlays like Kopia has, with CLI and a mobile app, all coded in Rust for speed and safety. Those are actually all obvious to me, and with the exception of Rsync for Windows, which I only recently learned about in [ COMMUNITY UPDATES ] AeroFTP: Major Milestones, Checkpoint, and Thank You #271 (comment), I did envision such an ideal app and discussed it with AI before I found AeroFTP.
  
  I've known for almost a year now that there are various applications that are the most efficient when coded in languages that have a GC, whether thanks to JIT compilation hotspot optimisations like Java has, or because some applications naturally take occasional breaks, so "stopping the world" with Golang isn't too inefficient for them. But it seems that regarding continuous data transfers as in this case, there is no room for taking breaks where the coding lang can ideally "stop the world" and there are no hotspot optimisations, so a lang without a GC is ideal here. This is despite Rclone, Restic, Kopia and Syncthing all having been coded in Go, which seems to be mostly for simplicity, and that Rust hadn't been out for that long (2012) when some of these started (Syncthing in 2013, Rclone in 2014).
  
  By the way, I didn't discover AeroFTP via that AI chat. I looked for multi-cloud syncing apps in various places by searching for key terms as I'm used to doing as a researcher. Despite being a Windows user, I discovered AeroFTP on the Snap Store website of all places, because AeroFTP came up when I searched for "cloud drives", "webdav" and "s3".

My questions here are:

1) Is there a place where you already list every aspect of AeroFTP which makes it unique, better and original,
especially compared to the most similar apps?

I saw you comparing the encryption AeroVault has (and that you're working on) to Cryptomator and Rclone Crypt in several tables all over the place, but I'm talking about having a page that would list this as a single bullet point out of many, with a link to another page where you have a table that compares the various encryption standards.

2) What innovative aspects specifically do you believe you came up with originally from the very beginning of AeroFTP?
Ideas that you don't believe others came up with.

Of course, absence of evidence is not evidence of absence. You can't ever know for certain whether Others might've already thought of your original ideas before you. I had to explain in my PhD viva that if I research the first 100 publications that my search comes up with and all of them lack what I believe is my original idea which makes my research unique, the 101st paper could've been the one where the authors had already thought of and implemented my "original" idea before me. It's not certain whether someone is the first to think of an idea, just likely.

I think that the AeroTheme/AeroFamily is definitely the clearest example of an original, innovative aspect of AeroFTP that I certainly would never have come up with.

I had an original idea yesterday based on #266 of connecting the concepts of rewriting existing ideas in Rust, which is known as Oxidation, with the word Aero, because air, or at least oxygen, is needed for at least certain oxidations (losses of electrons) like Rusting, unless you came up with it first. I actually don't know. Did you choose Rust partly because you chose the word Aero, or did you choose the word Aero partly because you chose Rust, or were these choices unrelated?

Heck, now I'm even connecting this with losing the Electron framework in favour of React, and a Reaction is exactly what Oxidation and Rusting are.

[axpnet]

First of all, thank you for your comments about me and the app's features. To better understand AeroFTP, perhaps it's helpful to go back to its origins and tell me a little about myself. First of all, but perhaps less importantly, my studies and training are in economics, with a particular focus on mathematics and statistics, and scientific subjects in general.

However, during my studies and the advent of the first computers connected to the Internet with the first 16k dial-up modems, my interests shifted to a wide range of IT, from PC assembly to network configuration, programming with the first web-oriented languages, and using all kinds of applications. Like all developers (in the old days, I don't think this is true today), I had to choose the platform on which to develop. I couldn't jump from ASP to PHP, and my obvious choice was Linux, free, open, and above all, stable and high-performance. For those of us who come from that era, using software freely and with the utmost respect for our anonymity has always been a must. I don't understand why this attention, despite what is said, has been lost!

So, I started creating websites and web applications (using the technologies of the time) back in 1998. My programming studies led to the creation of a highly advanced, customizable CMS that I have been developing for over 20 years. It's now capable of handling any queue, from CRM to e-commerce, a booking system, a distance learning platform, internal management software for companies, and a file manager that has evolved into a cloud platform, finally (recently) becoming a complete WebDAV platform with a related Android app. It essentially transforms your FTPS server into a complete cloud platform that leverages your domain's web hosting. I never considered commercializing the CMS; it's my personal development tool, what differentiates my projects from others.

I started with Windows and Linux, then continued with macOS, which was my web development platform for years, until I completely migrated to Linux as my primary OS for both development and personal use. I dual-boot all my desktop and mobile PCs with Linux as the primary operating system and Windows as a secondary operating system for any needs Linux can't cover. I continued studying other languages, not just web-based, and only six years ago did I discover Rust and start creating small apps for myself, just for training and curiosity, which has always fueled my journey of discovery. I've always developed on my own, learning by studying and testing. If there's one definition that definitely fits me, it's that of a Full Stack Developer, true and native.

So, as you can easily understand, remote servers, especially FTP, have been my daily bread for years. I've had dozens of projects on different servers to manage, countless profiles to save and manage, thousands of databases, backups of thousands of files, and transfers of entire projects from one server to another, not to mention synchronization and cloning.

I won't go into further detail here and get back to the main point: Rust and AeroFTP. A year ago, I decided to try my hand at a simple FTP client. I was really tired of the available apps, and I started a first version without aspiring to create something that could replace apps that had been around for years, even though, in my opinion, they were somewhat dated and had quite a few problems. Meanwhile, new AI tools have emerged, with great initial skepticism, justified by the complete incompetence of the first models for complex tasks. The project progressed and grew until I decided to focus on the GUI version, which had been my initial goal: a fast and reliable FTP client that could replace the old and dated client. In September-October 2025, something changed. AI models arrived that were increasingly efficient and precise in executing tasks, and I was immediately amazed by the potential and decided to really delve into the topic. I already had an app in hand, the ideal testing ground for experimenting with the potential of the new tools and pushing them to the max, as I have been doing in recent months. AeroFTP began to take shape. The name comes simply from the choice of a universally easy-to-pronounce and -understand name, representing speed and, in a broader sense, rapid transfer from one place to another. FTP because it is the protocol that was first integrated into the app, the initial purpose itself, and ultimately a name that encompasses a bit of everything. File Transfer Protocol is the name of the well-known protocol, but it also represents the app's primary purpose: transferring files. So Aero + FTP was the final choice, which in an early version of the release was installed on the system as aero-ftp, later becoming aeroftp. I wanted to honor the name of the historic FTP protocol while giving it a new look.

That said, the meaning you managed to extrapolate is fantastic!! (Aero + Rust + React: air reacting and generating rust, brilliant!) and adds a new interpretation that fits perfectly, it's certainly more poetic than mine :-) and I'm absolutely delighted to make it part of AeroFTP!

I've gone on too long and haven't answered the questions.

1. Is there a place where you've already listed every aspect of AeroFTP that makes it unique, better, and original,
   especially compared to the most similar apps?

First of all, I don't think AeroFTP is better than others, or rather, I don't want to be the one to say that. I must honestly say that I initially prepared a comparison table, which I keep among my many development documents, of all the file transfer clients, but I later decided not to publish it. For a young app like AeroFTP, publishing comparisons where it appears to be better than apps developed over years seemed inappropriate to me. Now the time is almost ripe to consider doing so.

There are many aspects that make AeroFTP a different and unique product, not necessarily better; it depends on your workflow and needs.

What benefits us is the fact that there aren't many potential competitors, perhaps because the industry is highly technical and no one has ever thought of investing money and energy into creating a new app. It wouldn't have been profitable, and that's why the industry has remained with decades-old apps that haven't been updated much over the years.

2. What innovative aspects specifically do you believe you came up with originally from the very beginning of AeroFTP?
   Ideas that you don't believe others came up with.

The goal of innovation has always been the guiding principle behind development, but before innovating, you must at least catch up or get closer to others before turning and taking the direction of innovation. You've listed some of the things that make AeroFTP already a product without many competitors, simply because others haven't thought to implement simple things, not because AeroFTP is who knows how far ahead of others; we're just talking about optional features.

The real innovation is certainly the fact that it's the first file transfer client designed for both humans and agents, and the highly advanced AI integration with AeroAgent already takes AeroFTP to a whole new level, incomparable precisely because we do something else. It's the first with MCP integrated with the app. And it couldn't be otherwise for an app created in 2025.

The level of security and encryption adopted is perhaps excessive, perhaps too complex, but there aren't many other examples in the file transfer industry with the same level of security.

There are many ideas in the pipeline, and I believe now is the time to focus on truly innovative solutions. The DAG-based centralized system is a crazy example that I still don't know where it will lead, and whether I'll have to go back on previous steps in the future. But the beauty of discovering new paths is precisely the risk that they might work or not, and if they do, then you're a winner.

We'll have the opportunity to delve deeper into this discussion, and I'd like to extract some feedback from it to publish on the web and GitHub channels.

[EhudKirsh]

Thank you for the reply. I don't mind if your answers are a little long, as I find the extra context interesting.

Your comment, by the way, partially answered a future question I was going to ask, which is the rationale for the FTP in AeroFTP. This does actually beg the question: will FTP ever be deprecated? What if FTP ends up being always less efficient than other data transfer protocols? Namely, S3 and APIs. Because these can handle lots of small files better, and FTP can break firewall rules. Will it still make sense to have the name of a deprecated or at least inefficient protocol in the name of the app? I actually was going to suggest that maybe AeroCloud or AeroSync might be better names for the AeroFTP app, and that AeroCloud might be renamed as AeroBox, because it acts like Box and Dropbox.

Regarding your direct answers:

1) Comparison Page

I was also thinking that it might've possibly been better to wait with publishing a comparison page and that it might soon be the right time for this. As the saying goes: “Be humble or be humbled”. I do have some ideas for bullet points for it, though. The approach of starting such a page should be to look at it as an aim and as a theory, and not to advertise it yet. AeroFTP will need a longer track record of being free of bugs first as a basis to consider comparing it to other apps. But I think this is important to have eventually.

2) Innovative aspects of AeroFTP

I appreciate that you come in good faith to answer this main question, though I must say that I disagree with your answer here. I make a distinction between features that are creative/innovative/original and those that are just unique. The former are always also the latter, but not always the other way around. For a feature to be unique, it must not have been implemented (well) in existing apps. For an idea to be creative/innovative/original, it must not be obvious, and others must not have thought about it before.

I agree that the AeroFamily fits this definition, but not AeroAgent, the integration of AI into a multi-cloud syncing app. When people started realising that AI became actually pretty good in the last ~3 years, companies like Microsoft tried to push AI in various places. Of course, AeroAgent is unique and is privacy-respecting, unlike probably other AI. But my point here is that it’s obvious to imagine AI integration into various apps, especially file management. I did actually think of mentioning AeroAgent in my question, but decided not to as I see it as the unique-only category.

I looked it up, and it seems that even AeroAgent’s ability to both use its own dedicated backend AI commands, as well as aeroftp-cli as a fallback, is an idea that’s already similarly implemented by GitHub CLI Copilot and Docker AI.

The point of my question and now also reply is not to demean AeroAgent, but to narrow down further aspects like the AeroFamily. Ideas that even if they seem obvious in hindsight, they clearly weren’t obvious. Sort of: “Why didn’t I think of that?”

Here are some examples of AeroFTP that I don’t think I would’ve thought of on my own, nor did I see any other app doing or notice anyone suggesting:

• 🚀 The rocket icon, which reminds me of the Kopia arrow that’s pointing up to represent upload backups. It also represents speed.
  In this context, the word Aero reminds me of Kopia’s cloud symbol in its logo.
• Pre-configuring MEGAcmd, WebDAV and S3 in the Add Service page.
• The badges in the My Servers and Add Service pages, including LOCAL for MEGAcmd and Filen.
• AeroCloud provides a “Box” (or Dropbox) experience in the same app as having AeroSync. You convinced me that this is a good idea to let the user choose their preference in AeroSync, AeroCloud and Crypt #145.
• Always having personal Client ID and Secret for OAuth, to avoid hitting the queries per second limit that all Rclone users face together when they use Rclone's default Client ID, as is described here: https://rclone.org/drive/#making-your-own-client-id. There's no ambiguity when connecting via OAuth in AeroFTP, that it's always by using personal Client ID and Secret. Sometimes in Rclone I'm not sure if my Client ID and Secret are working.

Storing the credentials in a more protected way than plaintext rclone.conf (or keeping its decryption key in plaintext, as if there’s much of a difference really) is also unique but obvious.

Another example for an aspect that’s unique, but obvious to imagine, is estimating the storage of cloud drives like S3Drive,
which RcloneView, RClone Manager and RcloneUI don’t seem to be able to measure.

And of course, those are just some of your ideas as of now, never mind your ideas in the future, and the ideas of contributors as well.
So there should be a categorised list somewhere, and I’d love to help with writing it.

Please don’t be offended if I disagree with you on things from time to time, and certainly if I ask questions in good faith.
If all think alike, it means no one thinks. I appreciate your appreciation for FTP, but I still think long-term.
Also, I bring a different understanding of the difference between features that are creative/innovative/original and those that are merely unique, because I believe the former often stand out more for some people.

[EhudKirsh]

I'm not sure if this post will become that complete list of unique aspects of AeroFTP. I didn't intend for it to become that, but it might just be that, at least partially and for the better.