Would an XFCE Qt fork be a good idea?

ToZ · 2022-07-14 23:48:06

https://mail.xfce.org/pipermail/xfce4-d … 32939.html

Olivier is one of the core developers.

Skaperen · 2022-07-15 00:15:17

or make a case of the benefits of having and using a variant with Qt. maybe that can be an influence.

sfammonius · 2022-07-15 02:55:53

Thanks (to both replyers)!

EDIT: I only realised that Oliver had already replied to an earlier email after posting this reply. Knowing that even one of the XFCE developers is open to the idea is a huge relief. I'll try to finish this as fast as I can.

Last edited by sfammonius (2022-07-15 03:00:31)

KBar · 2022-07-15 04:46:50

sfammonius wrote:

Is there any way I can actually contact the XFCE maintainers to ask them weather they would be willing to merge the Qt versions?

You already did via the mailing list. I was going to quote Olivier Fourdan, the creator of Xfce, but ToZ (thank you) has already linked his reply.

If Qt is really easy to work with than GTK, then they might be willing to think about merging.

However, I don't think it's going to be enough. There's another part to this story and it's the users. An average user doesn't really care if it's Qt, GTK or even Windows, as long as "it just works". So if you can rewrite, say xfce4-panel, in Qt, and it performs as good as its GTK equivalent without feature loss, why wouldn't they like it? For users like myself, you probably need to go an extra mile and prove that Qt apps actually perform better (doubtful), can be themed more easily (doubtful), and come with a little bit more features (to incite interest and motivation). "GTK sux, Qt better" isn't good of an incentive, motivation or justification. You say GNOME pushes this, forces that, well, I don't really see any difference between their approach and yours, which involves coercion.

With that said, I wish you nothing but success. I hope you prove your doubters wrong, me included.

lastonestanding · 2022-07-15 11:05:18

There will be no merge (similar to LXDE and LXQt). You will have to work on your project with your own set of developers, especially since a different toolkit and programming language is being used.

Compositoron · 2022-07-25 12:50:31

Xfce with Qt! That would be amazing.

I am one of those who thinks Qt looks way better than Gtk.
Good luck big_smile

bernd_b · 2022-07-28 09:01:40

From a users point of view:

My killer app is krusader. As long as I use this, I depend on qt whatever desktop I am using.

What I am trying to say is:
For me, it is not a question which desktop I am running. I want a set of software and functionality to do the job. And whatever I did, I ended up with a mix of gtk and qt apps. So far, xfce seemed to me the most lightweigt and easiest way to configure and run this mix of gtk and qt.

But: Yes, having all functionality depending on one of the two families and set them up with tools depending on the same family sounds promising to me. If that's the meaning of porting xfce to qt, you have a fan!

But it is not because I could say I like gtk more or qt more. In my case, it is because of krusader. I wouldn't mind dropping mousepad, since there is kwrite, kate for example. But configuring my window manager with a gtk app would be back to the mix which is I think the normal state of current linux desktops.

Last edited by bernd_b (2022-07-28 09:02:29)

sfammonius · 2022-07-28 11:12:50

There will be no merge (similar to LXDE and LXQt)

Is there a chance that maybe the developers could make it into another repo like an xfce5? (also thanks to everyone else for the support!)

Danielsan · 2022-07-28 13:22:03

I would not ever use a XFCE-QT version, at that point I would move directly on WindowMaker...

But rather than moving toward QT I see more strategic moving to toward EFL (https://www.enlightenment.org/about-efl.md), probably also other DE projects, such as Budgie/Solus, may decide to jump-in...

It is clear that GTK/Gnome has developed an open-source hostile agenda.

So far XFCE is the ONLY sane desktop environment and must take distance from GTK/GNOME as soon as possible before the same devs decide to leave the project because the effort is becoming unworthy.

A proof of concept might generate an interest and help the project to take a new path toward a better future!

Last edited by Danielsan (2022-07-28 13:22:58)

sfammonius · 2022-07-28 14:35:25

Danielsan wrote:

I would not ever use a XFCE-QT version, at that point I would move directly on WindowMaker...
But rather than moving toward QT I see more strategic moving to toward EFL (https://www.enlightenment.org/about-efl.md), probably also other DE projects, such as Budgie/Solus, may decide to jump-in...

How could Qt be so bad that you'd prefer to use WindowMaker? A lot of people have been saying that they don't like Qt because it doesn't look good and isn't customizable, but that's definitely not your reason. I've also looked at EFL, but it seems to be more focused on shapes and text than GUI widgets, which isn't really something that XFCE would need.

I agree with everything else you said though. I'll try having a usable panel finished as fast as I can for a proof-of concept.

EDIT: I take back what I said about ELM not having widgets, I just hadn't done enough research. It would be a great option if it supported themes.
EDIT2: ...It also supports themes. Why would it be better than Qt though?

Last edited by sfammonius (2022-07-28 15:13:54)

Danielsan · 2022-07-28 15:34:20

I don't like the looks and feel of the QT, never sympathized with them, never liked KDE etc... Therefore I would stop using XFCE if it would move on QT, and my other favorite WM/DE is WindowMaker. There weren't any technical reasons behind.

Why would be not a good choice being under the QT? Besides the usual QT licenses issues, besides the fact that if you stick witha stable version you have to pay for its use, the problem here is that the project would go over a similar situation (if not worst) as for the GTK: where there is another big DE which in someway is influencing the QT development, and if the this DE would start feeling annoying by another competitor in the same field, we are probably going to face the same situation were piece of code will be geared around the bigger project rather than being agnostic.

I do not have any technical knowledge to say which library is better but EFL seems a decent alternative to GTK and QT, and if the project must go away from GTK had better to find a safer place, QT doesn't appear to be such place.

Last edited by Danielsan (2022-07-28 15:35:50)

sfammonius · 2022-07-28 17:44:18

Danielsan wrote:

I don't like the looks and feel of the QT, never sympathized with them, never liked KDE etc... Therefore I would stop using XFCE if it would move on QT, and my other favorite WM/DE is WindowMaker. There weren't any technical reasons behind.
Why would be not a good choice being under the QT? Besides the usual QT licenses issues, besides the fact that if you stick witha stable version you have to pay for its use, the problem here is that the project would go over a similar situation (if not worst) as for the GTK: where there is another big DE which in someway is influencing the QT development, and if the this DE would start feeling annoying by another competitor in the same field, we are probably going to face the same situation were piece of code will be geared around the bigger project rather than being agnostic.

Qt has been free and GPL2-licenced since 2002, and it even signed a contract with KDE that basically said this will never change no matter what. It's very unlikely that Qt will go bad like GTK did, since it's main customers aren't desktop environments. Qt depends mainly on large corporations developing embeded devices, and those corporations could dump it in an instant if Qt even thinks of doing what GTK did. GTK on the other hand is *owned* by a desktop environment, so what happened isn't very shocking.

Danielsan wrote:

I do not have any technical knowledge to say which library is better but EFL seems a decent alternative to GTK and QT, and if the project must go away from GTK had better to find a safer place, QT doesn't appear to be such place.

KDE is almost the exact opposite of XFCE, so I can imagine why you would be against XFCE using Qt if KDE is your only experience with it. I'd recommend trying out LXQt on a virtual machine before making any more assumptions about what Qt typically looks like. A lot of people have been saying that they just prefer GTK's appearance over Qt, but if you would rather use EFL or WindowMaker over it then you definitely don't have the right image of what it looks like.

Last edited by sfammonius (2022-07-28 17:56:08)

ToZ · 2022-07-28 18:11:09

sfammonius wrote:

I'd recommend trying out LXQt on a virtual machine before making any more assumptions about what Qt typically looks like.

Here is where I struggle with your reasoning and approach. Why not just invest your time and energy in making LXQT better as opposed to trying to migrate over a whole DE? Xfce is currently based on GTK and will most probably continue in that manner. Of course, you can fork the code and do what you will with it, but then again, why not just focus on LXQT? If the concern is about GTK4, then I can see more value in having a discussion around forking and maintaining GTK3, maybe a collaborative effort between Mate, Cinnamon and Xfce developers, then about migrating to QT.

Danielsan · 2022-07-28 19:11:06

sfammonius wrote:

I'd recommend trying out LXQt on a virtual machine before making any more assumptions about what Qt typically looks like. A lot of people have been saying that they just prefer GTK's appearance over Qt, but if you would rather use EFL or WindowMaker over it then you definitely don't have the right image of what it looks like.

Of course that I took a look at LQtx, I took a look also when it was Razor-qt. I think has been a great idea the merging between the two projects also the adoption of the QT.

However If you like QT so much doesn't make any sense bring XFCE over QT, I don't see any additional value for the project, unless you are implicit admitting the superiority of the QT respect the GTK#, so you are the one that is having a bias actually.

QT have a double licenses you'll never know what is going to happen, GTK are moved behind the scene by RH/IBM, EFL aren't that better since are (or were) sponsored by Samsung. For me, I speak for myself, changing libraries (but not QT) would mean saying to GTK and Gnome: «we do not depend anymore by your stuff»; also GTK are so geared around Linux/Systemd that is complicated keep any project, based on, fully agnostic.

Probably the solution for minor DE based on GTK would be to recover everything that is going to be throw outside the window inside some sort of band-aid library. The problem here is the GTK folks, to make this harder to accomplish, have made the GTK a moving target, just to play square and fair with the others...

Last edited by Danielsan (2022-07-28 19:23:08)

sfammonius · 2022-07-28 20:33:39

The main reason I'm trying to get XFCE to switch is because I hate to see Unix as a whole slowly get worse thanks to GTK and GNOME. I think if XFCE switches, than it will encourage other apps and DE's to do it as well. It will do this by adding some variety to the examples of what Qt can do (Imagine if all desktop environments except GNOME and LXDE used Qt, everyone would think that GTK was slow/ugly. This is the same thing that's happening to Qt right now). I considered asking the same thing on the MATE forums, but I figured MATE users/devs would be less likely to like the idea since they mainly try to follow the designs of GNOME 2.

I do actually prefer LXQt to XFCE in a lot of ways, but I've used XFCE in the past and I only stopped using it after starting to dislike GTK. Originally, my only issue with GTK was that it took too much screen space, but the issues have gotten much bigger now.

TLDR: I'm forking XFCE because I think it would be better than LXQt if it used Qt, and because I think it will give Qt a small boost over GTK in the Linux/Unix world (which I think is a good thing).

Danielsan wrote:

QT have a double licenses you'll never know what is going to happen, GTK are moved behind the scene by RH/IBM, EFL aren't that better since are (or were) sponsored by Samsung.

Qt's licence model means that a project can use Qt for free if it is open source, or it can use Qt for proprietary software if it pays. As for the other stuff, I don't see how any of those things are directly bad. I just don't like GTK because it's forcing it's design, but how are corporations themselves the problem?

ToZ · 2022-07-29 11:28:59

sfammonius wrote:

The main reason I'm trying to get XFCE to switch is because I hate to see Unix as a whole slowly get worse thanks to GTK and GNOME.

(Imagine if all desktop environments except GNOME and LXDE used Qt, everyone would think that GTK was slow/ugly.

These are very subjective statements. Not everyone agrees with them.

TLDR: I'm forking XFCE because I think it would be better than LXQt if it used Qt, and because I think it will give Qt a small boost over GTK in the Linux/Unix world (which I think is a good thing).

And this is the crux of your motivation. Not to improve Xfce, but to make Qt "win".

I'm closing this thread. You will be better suited posting this to some sort personal blog somewhere.

Xfce Forum

Sub domains

#26 2022-07-14 23:48:06

Re: Would an XFCE Qt fork be a good idea?

#27 2022-07-15 00:15:17

Re: Would an XFCE Qt fork be a good idea?

#28 2022-07-15 02:55:53

Re: Would an XFCE Qt fork be a good idea?

#29 2022-07-15 04:46:50

Re: Would an XFCE Qt fork be a good idea?

#30 2022-07-15 11:05:18

Re: Would an XFCE Qt fork be a good idea?

#31 2022-07-25 12:50:31

Re: Would an XFCE Qt fork be a good idea?

#32 2022-07-28 09:01:40

Re: Would an XFCE Qt fork be a good idea?

#33 2022-07-28 11:12:50

Re: Would an XFCE Qt fork be a good idea?

#34 2022-07-28 13:22:03

Re: Would an XFCE Qt fork be a good idea?

#35 2022-07-28 14:35:25

Re: Would an XFCE Qt fork be a good idea?

#36 2022-07-28 15:34:20

Re: Would an XFCE Qt fork be a good idea?

#37 2022-07-28 17:44:18

Re: Would an XFCE Qt fork be a good idea?

#38 2022-07-28 18:11:09

Re: Would an XFCE Qt fork be a good idea?

#39 2022-07-28 19:11:06

Re: Would an XFCE Qt fork be a good idea?

#40 2022-07-28 20:33:39

Re: Would an XFCE Qt fork be a good idea?

#41 2022-07-29 11:28:59

Re: Would an XFCE Qt fork be a good idea?

Board footer