Why XFCE gona die if no action taken

paweuj · 2013-05-29 08:59:56

Probably the topic of this post is annoying to developers who spend private time for their little, sweet baby - xfce.

BUT

As I am still satisfied and concerned in the same time user I would like to act before sad things come true.

So why in my opinion xfce gonna die?

- No clear direction of development, no consistent HOLISTIC VISION
(if exists why don't share with us? ) development is driven by users requests, making it fragmented and reductionistic.
- Hermetic team, no new members inflow
- Focus on efficient, not resources consuming framework, but mainly for old devices
- No self-motivating, team building stuff
- No broader cooperation

My advices, I hope some other users also give advices or argue with me or just say I am a stupid jerk.

- Porting everything to QT
- Design also for modern, small devices with different user interfaces
- Write more about your development, share your ideas, attract more developers to be involved
- Create a roadmap with your VISION and brainstorming feature!
- Try to involve summer-code google students and other projects
- Use crowd funding

Give me more!

the_file · 2013-05-29 17:19:15

paweuj wrote:

Probably the topic of this post is annoying to developers who spend private time for their little, sweet baby - xfce.
- No clear direction of development, no consistent HOLISTIC VISION
(if exists why don't share with us? ) development is driven by users requests, making it fragmented and reductionistic.
- Hermetic team, no new members inflow
- Focus on efficient, not resources consuming framework, but mainly for old devices
- No self-motivating, team building stuff
- No broader cooperation

It doesn't need any of the things you mentioned. Otherwise its going to turn into a piece of shit that gnome 3 is. With all those "developers" listening to the "users" and making it "better" bs when infact the developers listen to a bunch of kids who have nothing better to do than make useless mockups of how they think the user interface should look like, and disregard the majority of the users. Xfce should stay the way it is. If you don't like it go fork it and work on your own changes.

- Porting everything to QT

The team is already committed to GTK why would they suddenly "port everything" to QT.

Oww and this:

- Design also for modern, small devices with different user interfaces

Hell no, go fork it and do it yourself. This is a huge waste of effort as there are already great desktop implementations for such devices, and yet there are is very few hardware that can actually run them without having to unlock the device.

Last edited by the_file (2013-08-08 04:41:49)

MountainDewManiac · 2013-05-30 02:11:30

paweuj wrote:

- Design also for modern, small devices with different user interfaces

Like Microsoft's Windows 8 and Ubuntu's mess?

No... Please, no.

PaperNick · 2013-05-30 05:15:03

As the guys above said, I think xfce is doing great and the only problem that seems to exist is that there is no inflow of new users. Xfce just needs to be advertised more...

BearOso · 2013-06-01 03:31:17

MountainDewManiac wrote:

paweuj wrote:
- Design also for modern, small devices with different user interfaces
Like Microsoft's Windows 8 and Ubuntu's mess?
No... Please, no.

Or like GNOME 3, too.

If desktop environments die, then XFCE, and others like KDE, etc, should die with them. Taking the name and applying it to something entirely different like GNOME did is completely idiotic. There is no name recognition with these projects and no market share. If you're going to make something new that bears no resemblance to the original, then call it by a new name. Don't bastardize an existing, working software base in pursuit of some sort of acclaim you'll never have.

Ubuntu, GNOME, Microsoft, and Mozilla all think they're getting in on the ground floor of a new technology base (tablets and phones) and that they're going to be big successes. The problem is that they're several years _late_ to the party. The people working on these things have lost any sensibility they had in their previous endeavors. With the advent of smartphones and tablets they have all lost their heads.

Last edited by BearOso (2013-06-01 03:32:55)

sixsixfive · 2013-06-01 16:59:43

paweuj wrote:

- Porting everything to QT

After I read this post (especially the Transmission Systray Icon part) I would fully understand if they would switch or fork something. Anyway!, we will see what the future brings and if Xfce@GTK3 sucks there is still Razor-qt, LXDE or Enlightenment tongue

paolo321 · 2013-06-01 17:09:29

paweuj wrote:

- No clear direction of development, no consistent HOLISTIC VISION
(if exists why don't share with us? ) development is driven by users requests, making it fragmented and reductionistic.
- Hermetic team, no new members inflow
- Focus on efficient, not resources consuming framework, but mainly for old devices
- No self-motivating, team building stuff
- No broader cooperation

I almost don't understand your point of view. Xfce has a slow development because there are not many devs, so they focus on small and gradual changes.
btw, I just like it that way. When I update I find new features, but I never feel "confused".
If somebody has skills to help develop, I think they'll like contributions :-)
They've done nice things following even suggestions from users, like thunar tabs, and they'll always add new features to applications, keeping them up-to-date but without "weird new things".

- Porting everything to QT
- Design also for modern, small devices with different user interfaces
- Write more about your development, share your ideas, attract more developers to be involved
- Create a roadmap with your VISION and brainstorming feature!

I'm not a developer, but imho this is outside xfce "focus".. Why port everything to qt when there are other qt lightweight desktop managers? Why create a tablet-phone interface when I just use a pc on my desk? :-)

- Try to involve summer-code google students and other projects
- Use crowd funding

In my idea, being a "small project" does not mean being a "bad project".
Many users (like me) tried xfce when they stopped forcing themselves to use unity or gnome3 with their weird new UIs , then imho started liking Xfce "principles".

Anyway, if you have suggestions there's always someone reading them, sometimes devs post on forum so I think they read it! btw, there's also another thread where some small improvements are proposed...

Last edited by paolo321 (2013-06-01 17:10:46)

poltiser · 2013-06-01 18:31:26

I like XFCE... It is very simple DE but functional and sophisticated. The changes of last year with new thunar, connectivity and so on are impressive. I am not IT pro and I am doing well. I love direct access to files and folders from panel (only one panel by the way!) using Directory menu (set with *.* to show everything). I can connect all computers we have at home and phones using printer and common files... and weather applet, fantastic!
Only problem to correct I can comment on are configuration files. There are too many places with config files which turn back to default settings without warning. I have an impression that enthusiastic developers are more interested in creating new features than polishing old ones. I sympathize. But certainty is what I need working.
Anyway - thank you all for your work. Impressive!

paweuj · 2013-06-03 18:58:40

Thanks for the answers! I've bumped with the strong conservative community!
I like XFCE too, using it on my little netbook,
however on bigger laptop I've recently installed Linux Mint Olivia with amazing Cinnamon.
(gtk made it a little buggy - nothing is perfect)

I've read carefully your posts and came up with several findings:

Less changes - better
Less developers - better
Messy gtk - better
Work in less devices - better

I know, it's hard to have absolute values, that's why we have so many various nice things!

One more thing: a good project is not bad if it's small,
but if a good project is under the process of getting smaller - bad news i would say.

I also don't understand completely the fear that if we adjusted the framework also to mobile devices
it would push it in a shit like gnome, windows or Unity.
Is there just one path that we can go, inevitably reaching those crappy frameworks level?
I thought rather about a framework which would adjust to a device that works, having best desktop
interface on desktop and best possible on mobile dev. (not necessary mobile phone!)
So for instance instead of switching from XFCE to {gnome, unity} - which are shitty wherever they work,
just smoothly let XFCE switch to its mobile mode.

Summing up I would leave all the best features XFCE has now and add an adaptive mode for it.

In my opinion windows 8, Unity and Gnome are not just too late.
In fact they are designed for devices still not popular on the market.
It's interesting how pc market will change in next two years,
however it can be roughly anticipated that with new, more efficient processors, graphic cards, more sustainable batteries -
hybrid devices become standard. Why? People use to work on PC, but mostly people do "consume"
a content than create new one. Furthermore hybrids are more collaborative, flexible and soon will be much cheaper.

Haswells (and their counterparts) gonna start more numerous emergence of hybrids -
their successors afterwards will end the pc era. You don't believe it, do you?

Last edited by paweuj (2013-06-03 19:04:29)

Sideburns · 2013-06-03 22:17:50

I think that there's a problem here that you're not understanding: in order to make a DE that will work well on tablets and other mobile devices, you have to design it around their size limitations and their use of a touch screen instead of a mouse or other pointing device. Most experienced computer users aren't willing to put up with that on a more fully-featured computer, such as a desktop or laptop. And, if you try to create a single DE that works well on all devices, you either have an immensely bloated resource hog, or a bewildering number of modules trying desperately to provide the same experience on different platforms. In either case, you also end up with a maintenance nightmare. To me, at least, Xfce is doing the smart thing by not trying to be all things to all users on all platforms. It was designed to work on desktops and laptops with limited resources, using as little CPU and RAM as possible and leaving the rest of it available for those programs doing the actual work, and up until now, that's what it's been. Personally, I hope it stays that way, because that's what I want, both on my desktop and my laptop. I also have a tablet, but its needs and requirements are different, and I doubt that it would be a good idea for Xfce to try to fill them because doing so would require it to turn away from its main goal.

MountainDewManiac · 2013-06-04 07:50:07

I think the public is in the process of being insulted, and unfairly so. It appears that certain DE (and OS, if you include Microsoft's Windows to the list, but I'd say that's more properly a DE issue as well; they're just integrated) creators think that we are not intelligent enough to have two devices that are similar in nature but have two different "DE"s. So they're trying to create ONE single DE which will look/act the same on a laptop/desktop (with a larger screen, mouse, keyboard, et cetera) as it does on a tablet or cell phone (with a small screen and no mouse or keyboard). The problem being that, in order to produce such a thing, it is unable to take full advantage of either device, will never really look/feel correct and "native" on either, and is often a RPitA on both.

If there's someone keeping track, somewhere, please mark me down as being thankful and apreciative to the folks behind Xfce for NOT insulting my intelligence by assuming that all of my devices need to look/act exactly the same.

-----

Pure speculation on my part, so I could be 100% mistaken (but I think not): The Xfce team doesn't start their development day by asking, "What can I change?" Instead, they start their development day by asking, "What can I improve?" IOW, if they make a change, they did so in order to improve something or to add a feature - NOT just so they can add another mark to the "Changed" column.

-----

I'd guess that when Xfce goes to Gtk3, it'll be done well. Notice they haven't rushed to do so? I think it'll be Gtk2 at least through the next release. I would assume they'll wait until making the change will provide a definite improvement either to the user or to the developers.

-----

[RAMBLE] I'm poor enough that if I could get by with only a crescant wrench, a couple of multi-tasking widgets, and a BFH in my toolbox, I would. But to be able to do the most jobs (and to do them well), I have found that having a wide range of quality tools that are designed to do a task well - as opposed to doing many tasks not so well - is not just recommended, it's a requirement. So as much as I like Xfce on my laptop/desktop, if I never see it appear on a cell phone, that's fine with me. Especially if it means that my laptop/desktop will always look/work well and be mouse-/keyboard-/me-friendly.[/RAMBLE]

Best regards,
MDM

andrzejr · 2013-06-04 19:16:29

My own comments, more senior developers may disagree.

Xfce is a mature project, it is working well and is useful to many people "as is". Development focuses on incremental improvements and maintaining the existing code (port to Gtk3 is underway etc). The team is small but sufficient for the task. If anyone is interested in doing such groundwork just do it - clone a git repo and fix/implement something useful.

Indeed, interest in Xfce is getting smaller but that's a part of a larger problem affecting all DEs. Partly caused by all more mainstream DEs almost simultaneously dumping their codebases and starting over (that has temporarily helped Xfce but in the long term it is a bad for all Unix DEs). Partly by people moving to the web and mobile devices - which is fine for Xfce, unlike for commercial projects which must expand to survive.

The biggest problem with Xfce *I* see is its development pipeline - long times between releases, no pre-build stable/devel/nightly binary packages etc. This means that some of our contributions (code, translations etc) have to wait more than a year for shipping. That's rather demotivating. Nick is doing what he can but he's still only one person.

As for some specific comments/suggestions:
- port to Qt - not a goal*.
- port to non-X11 platforms - not a goal*.
- UI paradigm change (mobile devices) - not a goal*.
- no direction, "holistic vision" etc - not true, the goal is to keep developing a lightweight, traditional, featureful X11 DE.
- poor PR - partially true. Most active communication channels are this forum, IRC channels and personal blogs/social sites.
- crowd funding - can't speak for everyone but people mostly join the project for fun/learning/social experience. I don't thing money would help much beyond paying some equipment/maintenance bills.
- mentoring, GSoc - always good thing to do but (1) needs additional effort, (2, just an observation) useful contributions usually come from people who received little mentoring.

*) We are not against deep changes (there were some discussions about Wayland in the past) but such development would require a deep rewrite and is better carried out on the side under a different name to avoid conflicts with the existing project. In practice, big changes require big effort so they need good justification and someone working really hard on them.

Sideburns · 2013-06-04 20:06:21

There are two mistakes that I'm glad that the Xfce devs haven't made, unlike those working on what I'll call "the Alaskan DE" to avoid controversy. One is that they don't think they're the only people who know what's right for the project and that the opinions of "mere users" don't matter. I used to use that DE, before version 3 came out and tried to use their support forum a few times. About the only way you could get a reply to a post was to say something about the devs that was less than adoration. Doing that would get you a large number of flaming replies from people telling you that the devs did not follow the forum, wouldn't offer any help and were only interested in the opinions of other devs. If you wanted to have any input to the process, you had to be actively working on the project and contributing code. (Why all of the people there were so hysterical about it when they knew that the devs didn't care is something I never bothered asking.) Here, of course, there's at least one regular contributor who lists himself as a "core developer." They listen, they reply, they help.

Second, unlike that other DE, the people working on Xfce don't assume that there's One True Way for things to be done. They don't decide for the users how many panels there MUST be, where they MUST be located or how windows are assigned to desktops, just to name a few examples. They also don't force people to create third-party extensions to return functionality that they've removed because they don't use it and therefore nobody else needs it. In short, the Alaskan devs live and work in an ivory tower, removed from the people who actually use what they produce and indifferent to their needs while the Xfce devs live in The Real World and care about what makes Linux users want to install and run their DE.

And yet, that DE is still very popular, and has more users than Xfce does. Why? Probably for the same reason that Windows has such a large market share: most PC computers come with Windows pre-installed and many buyers wouldn't know how to replace it with Linux, even if they wanted to. In the Linux world, a fair number of distros are built around the Alaskan DE and come with it by default. Users, especially inexperienced ones, tend to stick with whatever's provided because they presume that whoever put the distro together knows what they're doing and that they picked the best DE available. (Really new users, of course, probably aren't even aware that Linux has more than one DE, but that's beside the point.) As an example, my sister uses Ubuntu, because for the most part, It Just Works, and that's all she cares about. When Ubuntu switched to Unity, she stayed with it and tried to learn it. After a year of fighting it, she admitted that Unity and Parkinson's don't get along very well and let me install Xfce. If and when we need to do a real re-installation, we'll just use Xubuntu instead.

The important thing to remember is that Linux is all about choice. That's why there are so many distros, so many DEs, so many editors and so on. If you don't like one, try another and see if you like it better.

paweuj · 2013-06-07 10:07:11

I just wanted to say XFCE was quite mature and not many things to improve in its desktop functionality. So it can expand to more devices by reconfiguration according to the environment it would work. I mean it would smoothly switch interface when we connect to device keyboard or mouse or different screen. It doesn't require to keep much of bloating stuff in ram memory, because it would be dynamically loaded and unloaded according to needs, without loosing the user's session. So it's different concept than others DEs. It has nothing to do with insulting people's intelligence.
Cheers.

Korrode · 2013-06-14 04:02:00

Most of the points the OP claims as problems, I disagree that they're problems.

For the vast most part, i like Xfce just the way it is. From development philosophy to release cycle. it suits me very well.

The only suggestion from the OP I would support is porting to Qt.

Also, regarding andrzejr's mention of Wayland, I am in favor of having Wayland support as a high priority. I will be looking to move to Wayland as soon as is viable. (To my understanding; Qt would potentially speed and ease Wayland support, however that's not the only reason I would be in favour of a move to Qt.)

pressanykey · 2013-06-16 20:48:49

All I know is that this latest Xfce 4.10 release just works and is stable. Kudos to all that make that happen!!

Thunar and right clicking the desktop is fixed (at least in the Xubuntu 13.04 distro) without any changes to gvfs or networking files.

And Xfce coexists well with other desktops and I haven't experienced any mood swings with Xfce cpu and memory usage.

Not surprisingly, I like Xfce and its looks and find myself connecting to an Xfce session more often than not for its stability.

So, sorry I have nothing to add to the controversy... Except to not add anything..

Jristz · 2013-06-17 05:01:02

The only way that Xfce going to die if no releases are maked
or even if no git pull are done
and even if distros drop xfce THIS is what going to make xfce (or anyother desktop) death

inother words, make more xfce releases going to help
and dont forget Nick: documment all the option or varialbes, like that that can make tab show always, we not want hiden option or oscure undocumented features

ruziel · 2013-06-20 18:33:35

As someone who only moved from Windows to Linux 6 months ago & who loves using Manjaro Linux with Xfce DE, I believe that the rumours of Xfce's impending death are greatly over-estimated! The first distro I tried was Mint & while I initially liked Cinnamon, it soon proved too buggy for me to rely on to run my business day-to-day. I then tried Mate for a while, before giving Xfce ago. I have since come to really appreciate it's stability, as well as it's intuitive, attractive, light & simple interface. Personally I hope that the development team keep going with their steady & well thought out development cycle, which is not prone to flashiness or making changes for the sake of it or worse, changes that the majority of it's user base don't want. For those of us who want to get things done on their desktops & notebooks, it's an excellent choice. Great work team & a big thank you to you all. Ruziel.

fixitmanarizona · 2013-08-01 12:58:39

"Crowd funding?" c'mon. Like the joke that's the Ubuntu Phone at the moment?

MountainDewManiac · 2013-08-02 18:04:22

fixitmanarizona wrote:

"Crowd funding?" c'mon. Like the joke that's the Ubuntu Phone at the moment?

You have, perhaps, heard of Kickstart? I believe that several projects have received over a million dollars (US) in funding.

Not saying that it'd work in this case, IDK - but the idea has been known to work. I just wish it had a different name, lol; it reminds me of "mob intelligence" (which often approaches a zero-state).

Regards,
MDM

paweuj · 2013-08-05 11:25:28

fixitmanarizona wrote:

"Crowd funding?" c'mon. Like the joke that's the Ubuntu Phone at the moment?

They apparently overestimated the demand from corporate clients.
Companies prefer to buy something already existing in the market, which is well supported and evaluated.
It means they made a mistake in raising funds, which doesn't condemn the whole idea of crowd funding.

In fact it's a great tool for open source projects, but a project must be properly tailored to people's needs.
I think they should have offered two sizes of devices: 4.5'' and 6.66''. I would buy the bigger one wink

the_file · 2013-08-08 04:44:01

paweuj wrote:

I just wanted to say XFCE was quite mature and not many things to improve in its desktop functionality. So it can expand to more devices by reconfiguration according to the environment it would work. I mean it would smoothly switch interface when we connect to device keyboard or mouse or different screen. It doesn't require to keep much of bloating stuff in ram memory, because it would be dynamically loaded and unloaded according to needs, without loosing the user's session. So it's different concept than others DEs. It has nothing to do with insulting people's intelligence.
Cheers.

You don't seem to understand a simple fact, XFCE is a DESKTOP environment, NOT tablet or whatever device environment. You CAN'T have it running on both a desktop AND a "device" without severely crippling the other. Which is why I am saying if you want xfce on a tablet, you need to completely fork it, or else the desktop version is going to be changed into a tablet UI piece of shit like gnome 3 is.

paweuj · 2013-08-08 06:41:58

the_file wrote:

paweuj wrote:
I just wanted to say XFCE was quite mature and not many things to improve in its desktop functionality. So it can expand to more devices by reconfiguration according to the environment it would work. I mean it would smoothly switch interface when we connect to device keyboard or mouse or different screen. It doesn't require to keep much of bloating stuff in ram memory, because it would be dynamically loaded and unloaded according to needs, without loosing the user's session. So it's different concept than others DEs. It has nothing to do with insulting people's intelligence.
Cheers.
You don't seem to understand a simple fact, XFCE is a DESKTOP environment, NOT tablet or whatever device environment. You CAN'T have it running on both a desktop AND a "device" without severely crippling the other. Which is why I am saying if you want xfce on a tablet, you need to completely fork it, or else the desktop version is going to be changed into a tablet UI piece of shit like gnome 3 is.

Here everyone has understood that fact. XFCE is a desktop environment obviously. I do believe I CAN, though. I had a concept of a DE that loads only the modules that are necessary for a certain environment. I mean, it adjusts and automatically configures to the environment that it is launched. That's why it will not be crippling. Much different than Canonical's idea.

Sideburns · 2013-08-08 06:56:21

One of Xfce's big selling points has always been its small footprint, both in disk space and RAM. Making it able to detect that kind of thing and adapt on-the-fly would just add pointless bloat. You'd be much better off porting it to a version designed for use on a tablet, but if you did, would it still be Xfce?

paweuj · 2013-08-08 07:03:34

Sideburns wrote:

One of Xfce's big selling points has always been its small footprint, both in disk space and RAM. Making it able to detect that kind of thing and adapt on-the-fly would just add pointless bloat. You'd be much better off porting it to a version designed for use on a tablet, but if you did, would it still be Xfce?

Disk space is not a bottleneck anymore, especially if the space demanded in such modification wouldn't be much higher. RAM will not be affected at all, as there is a selective load approach.

Xfce Forum

Sub domains

#1 2013-05-29 08:59:56

Why XFCE gona die if no action taken

#2 2013-05-29 17:19:15

Re: Why XFCE gona die if no action taken

#3 2013-05-30 02:11:30

Re: Why XFCE gona die if no action taken

#4 2013-05-30 05:15:03

Re: Why XFCE gona die if no action taken

#5 2013-06-01 03:31:17

Re: Why XFCE gona die if no action taken

#6 2013-06-01 16:59:43

Re: Why XFCE gona die if no action taken

#7 2013-06-01 17:09:29

Re: Why XFCE gona die if no action taken

#8 2013-06-01 18:31:26

Re: Why XFCE gona die if no action taken

#9 2013-06-03 18:58:40

Re: Why XFCE gona die if no action taken

#10 2013-06-03 22:17:50

Re: Why XFCE gona die if no action taken

#11 2013-06-04 07:50:07

Re: Why XFCE gona die if no action taken

#12 2013-06-04 19:16:29

Re: Why XFCE gona die if no action taken

#13 2013-06-04 20:06:21

Re: Why XFCE gona die if no action taken

#14 2013-06-07 10:07:11

Re: Why XFCE gona die if no action taken

#15 2013-06-14 04:02:00

Re: Why XFCE gona die if no action taken

#16 2013-06-16 20:48:49

Re: Why XFCE gona die if no action taken

#17 2013-06-17 05:01:02

Re: Why XFCE gona die if no action taken

#18 2013-06-20 18:33:35

Re: Why XFCE gona die if no action taken

#19 2013-08-01 12:58:39

Re: Why XFCE gona die if no action taken

#20 2013-08-02 18:04:22

Re: Why XFCE gona die if no action taken

#21 2013-08-05 11:25:28

Re: Why XFCE gona die if no action taken

#22 2013-08-08 04:44:01

Re: Why XFCE gona die if no action taken

#23 2013-08-08 06:41:58

Re: Why XFCE gona die if no action taken

#24 2013-08-08 06:56:21

Re: Why XFCE gona die if no action taken

#25 2013-08-08 07:03:34

Re: Why XFCE gona die if no action taken

Board footer