Desktop integration and theme dependency

I can work with that. I will get this worked out. Cheers, Scarlett

1 Like

I think the adoption of these “solutions” should be undertaken by developers/software-house, they shouldn’t depend on subjective choices of build and distribution systems.

I understand that it is not easy to support so many things together.

What I think snap has no maturity yet. the linux environment has had an enviable packaging for years, tools like snap could replicate the same know-how. It is not so. Seems negligible, with snap the application icon is shown in a single size, XDG does not work. An image issue not to be taken lightly.

When you adopt snap and similar tools, you have doubts about what effectively you will see on the user’s desktop. Will it work or not? :exploding_head: Meanwhile things evolve: Qt6, Gtk4, many other new things, many less new things, for example: Tcl/Tk, Gtk, wxWidgets.

As a user, many popular packages come with old Ubuntu bases, bases that have had their life extended, for example VLC (18.04).

I understand the factory and communities make an effort, but as a snap user I’m not too convinced.

You can actually make a vanilla qt6 content snap for core22 by yourself. How ?? Well create a content snap and stage all qt6 packages from jammy in it and create a producer plug, then consume that snap from the snap you want to build, qt6 version 6.2 packages are available in jammy, you can also create a content snap with latest qt6 from lunar after adding lunar repos but thats unnecessary headache and complicated, so jammy one is the best option, that will also help others wanting qt6 with core22.

1 Like

I agree with you, Jammy is the best option, it’s the latest LTS, and it makes more sense to distribute following the latest LTS.

My humble opinion the factory, Canonical, is not doing enough. Documentation is outdated, Qt6 is not in the docs (it’s been 3 years since its first release), like other very popular toolkits, example: wxWidgets, Electron.

My impression 10 years ago when I tried snap, it’s not production ready software, it’s the same as today: Error, Error, this software doesn’t open, this one is no longer updated, where are my files?, etc. The experience is not comparable with traditional packaging, it feels like a big step backwards.

Imho It can be improved but not currently stable.

1 Like

Agreed, the main problem here is just like other projects Canonical is failing to integrate the community to take up several aspects of the snap world, take flatpak, the gnome,kde.ffmpeg,wine runtime etc. of flatpak arent driven by a single entity and are contributions of a large community whereas in snap world everything is in the hands of canonical leading to inefficiency, a single entity can never manage and control this, it needs to integrate the community etc. to have a equal representation in it, i always feel this as the main drawback of snaps, its not supported by community majorly, most of the work has to be done by canonical and other few individuals and corporate like the members involved here etc. Technically and theoretically snap has good potential but again lack of community contribution will just always be a hindrance. Overall Hope things will improve in near future, what canonical wants is control and centralization of control, nothing wrong in that as it brings quality control etc. but also has drawbacks.

1 Like

I too, I hope for future improvement. :crossed_fingers: Without going too OT, I often see a destructive behavior, the factory has important ideas like desktop convergence, but often destructive.

The factory should go back to being an industry reference imho. I say that as an user and developer, I don’t find documentation or support for a very popular toolkit in this environment, confusing things, so it becomes very hard to release software on snap.