Maintainer(s) wanted

I’ve recently had trouble finding enough time for developing manjaro-architect. I’ve now been almost half a year without a proper hardware for testing, so it is difficult to get things done. Furthermore, I’m now focusing more on other projects like gnome edition and calamares (hybrid netinstall and better btrfs support).

Manjaro-architect codebase is quite messy, and if I had time, I would probably rewrite it from scratch with narrower and more maintainable scope. I don’t want to abandon a working installer completely, since I know some people like it quite a bit. So, If someone is interested in developing manjaro-architect further, this is a good opportunity to get started.

3 Likes

Another project I know nothing about to tinker with?
Sure … why not … I’ll probably take a look on one of the long lonely nights.
Like tonight … or tomorrow night … or … :wink:

2 Likes

Not knowing how to do any of that, sure :smiley:

I definitely want to try testing, even if I don’t know coding that well (just started python), I would look at this as learning experience… being unemployed and all

1 Like

Invitation sent :smiley:

I’m not disappearing if you have any questions or need any help.

1 Like

Same as @olividir …I can only test :slight_smile:

Im happy to test anything and give feedback but I suck at git and coding. I once accidentally synced my home folder with an android kernel I was compiling, when I deleted the kernel I lost my home directory :rofl::rofl:

3 Likes

Testing is also super useful

1 Like

Is this still looking for testing or anything?

I don’t really know how to code, never really sparked interest in me – it seems like coders are needed - I don’t know how much I can help with that, but I’m fine testing out shiz on the Architect releases.

Just let me know what I can help with(testing or whatevs) and I’ll see if I can do it.

1 Like

Thanks for having provided this useful tool for a while.

I think there are two basic problems with Architect. On one hand it does too many things. It will be better having those capabilities broken into separated tools, divided into the smallest functions, and then compose them into a simple overlay installer.

Secondly it’s hard to test changes on the code. It would be useful having certain way to simulate the installation on a regular system, so you can easily test any small modification without much previous building or configuration.

I think the way to go is to start coding any of those small tools, or preferably the one that eases the last problem. Then just make Architect call them.

I’ve been planning to do this for a few years now, but haven’t had time to do it yet. I would start with simple noninteractive installer that reads a yaml configuration file. And even that could be broken into several submodules. Then, make a TUI that generates the configuration file and calls the actual installer.

If you have time on your hands and want to take a look at it, you are welcome to do so and I’ll help wherever I can.

1 Like

Yeah, I will do it.

I will start with something simple, like the locales configurator, to get an idea of how the usability should look like.

I will come back later. Ciao!

3 Likes

I have decided that I’m not interested in engaging with Manjaro on coding.

That what I code I will post on places where I can have a reasonable degree of independence to do so.

Places where these contributions are generally well received, and people is encouraged by design to be reasonably flexible and welcoming. Specially important who has powers to moderate and silence others.

I see that this is usually the case when moderation request are reviewed in the last instance by appointed professionals. By people who had a real selection process, and that are held accountable and honored by getting a pay.

i hope Architect will not get abandoned and will be kept maintained.
i think it’s a very important addition to the other installers that many people appreciate.

nowadays when I’m testing Distros in VMs.i always look if a minimal installer exist,
and my next Manjaro install will be through Architect.

I’m sure a lot of people loath what seems as bloat for more advanced user.

unfortunately I’m unable to help since it’s out of my scope.

2 Likes

It’s not exactly abandoned, I just don’t develop new features for the time being. At some point I hope to replace it with something simpler, but it’s going to be a while before I have such time. Currently I’m focused on improving calamares and gnome edition, but hopefully next fall I have more time for other projects.

I still try to fix manjaro-architect issues as they arise.

3 Likes

if i may suggest,
it’d be nice that the installer give the option to create a swap partition when choosing automatic disk partitioning.

Not a bad idea. Until it is implemented, I suggest using graphical partitioning tools like gnome-disks or gparted for the partitioning.

1 Like

I feel obligated to answer here and volunteer. Not only that I think the architect is the best way to install Manjaro, I just had a look into its sources today for what I need to do to add proper ZFS support as well as functions to automate the install via CLI or config files to use it with Ansible. @Chrysostomus I joined the Manjaro Gitlab. Anything besides what I want to add and change that needs to get looked at currently?

10 Likes

Awesome! I’ll add you to the maintainers of the repo. Do you have telegram handle? And your gitlab username?

Send you a private message.

1 Like

Is this related to why the Architect Edition was removed from the Downloads page? I was trying to find out why Architect was removed, and this seemed to be the most relevant result outside of some false positives.

What does this mean if I were to download the ISO for Manjaro Architect 20.0.3, install it to a new computer, and then fully update via Pamac? Theoretically, it should work like normal and bring me up to the equivalent system as if I had installed from the 21.0 ISO?