probonopd
commented
3 years ago Hello @BennyKJohnson
Thanks for your thoughts.
Building a distro is a huge undertaking.
Indeed. Hence we don't want to "build a distro". helloSystem is still FreeBSD, but preconfigured to do out-of-the-box what it is supposed to do. The base system and almost all of the packages come 1:1 from FreeBSD and FreeBSD Ports.
What we are building, though, is a desktop environment. We are putting it together from existing components rather than strating from scratch, but then go on to simplify things. In choosing components, we are going for the ones with the least amount of dependencies outside of Qt.
All these components are lacking in a lot of features compared to the main alternatives such as Latte Dock and Dolphin file manager.
Sure, but that is the point. Our Dock, for example, has no configuration. Our Filer is getting the features we really care about (icons for volumes on the desktop, Command-I for Get Info, etc.) and maybe some day we will find a way to add [Miller Columns].
In its current state, Hello is not suitable for day to day usage.
As someone who is using it every day exclusively since half a year I have to disagree.
I would start with a fully functional KDE environment on top of BSD
Let's distinguish between the KDE project and the KDE Plasma desktop. helloSystem is happy to use components from the KDE project, but instead of the KDE Plasma desktop we are using our custom helloSystem desktop environment.
While we do use some KDE components and applications, we think that the architecture of the KDE Plasma desktop is overly complex with layers upon layers of infrastructure, and way too many configuration options. Our objective is to make computing simple again, and KDE Plasma seems to value flexibility and configurability over being lean and being simple.
Can you point out concrete examples that make helloSystem "unusable as a daily driver" whereas KDE Plasma seems viable to you?
It could come with the global menu widget and latte dock out of the box.
Those do not work exactly in the way we would want them to work. As for the Dock, it is possible that it will go away entirely in favor of an Applications menu.
It would ship with the most common apps like Firefox, Thunderbird etc.
Those are not native Qt. Our objective is to remove Gtk along with other unwelcome technologies.
adding your special sauce by standing on the shoulders of others
Fully agree. Hence we are starting from existing pieces. But in line with our mission, we are usially choosing the simplest ones we can find, not the complicated ones. I call it "software minimalism".
BennyKJohnson
kettle-7
Hi,
First of all thanks for your Linux usability articles and your contribution to the Linux community with the App Image system. As a current MacOS user and someone who is dipping their toes into the Linux/BSD OS world, the Hello System project is quite interesting to me.
I agree with the majority of your ideas around user experience and would love an open OS that embodies the design principles you have discussed. As a software engineer however I do have concerns about your approach. Building a distro is a huge undertaking. What is concerning me is you're building it using half baked components. I've noticed forks of a file manager, a dock and a menu bar. All these components are lacking in a lot of features compared to the main alternatives such as Latte Dock and Dolphin file manager. This is a problem because instead of focusing on what makes Hello unique, you are spending a lot of time maintaining all these components. I think you are a bit too ambitious with what you're trying to achieve and I fear your going to burn out before this project even gets off the ground.
In its current state, Hello is not suitable for day to day usage. Someone can admire the project from a virtual machine, but then they will go back to their main OS. I believe this is a huge problem from a motivation standpoint. It will be difficult to attract users and contributors as there are currently too many compromises.
My suggestion would be to focus on adding your special sauce by standing on the shoulders of others. Why spend time adding basic features and maintaining when you can use other peoples work that do a better a job? Choose areas where you can add the most value by filling in the usability gaps. If this was my project, I would start with a fully functional KDE environment on top of BSD. It could come with the global menu widget and latte dock out of the box. It would ship with the most common apps like Firefox, Thunderbird etc. This would be unique enough combination to be interesting. I would then focus on areas where I can add the most value. Maybe starting by adding deep integration of AppImages into the OS. I would continue with this approach slowly but surely chipping away at the usability gaps. With this approach, the OS would be usable as a daily driver from the start and the user experience would improve over time. I get that KDE is not perfect, but in my opinion you are better off having an Distro that's not perfect but functional today than having a Distro thats barely performs what is needed in a modern OS and has a high chance of being abandon because the scope of the project is too large.
I wish you all the best with your project.