OK, this talk about this thing called Wayland had me doing some research. Like is this something that will be of concern at some point???
Well a bit of an eye-opener. Bye, bye, miss American X11.... (Recently saw some fun utubes on McLean :) ).
So where is Xfce, that I have run now for a lot of years (maybe around 10?)?
I found one discussion back in Mar '21 about plans for Xfce 4.18 and supporting Wayland. F35 uses 4.16.
Does anyone know current efforts on Wayland (that seems to be where Linux desktops have to go or die) for Xfce?
One cute comment I saw that Xfce will have to be called 'E', as it is no longer 'fc' and once it jettisons X, well...
Thanks
Joe Zeff writes:
On 1/8/22 9:03 PM, Robert Moskowitz wrote:
One cute comment I saw that Xfce will have to be called 'E', as it is no longer 'fc' and once it jettisons X, well...
Xfce hasn't stood for anything in particular for many years now.
Which is precisely the reason why xfce is a breath of fresh air. It just works. It does what a usable desktop should do: give me desktop icons; a menu bar; a tray; a pager; and some dockable icons. That's it.
Its profile is low enough not to attract interest from self-appointed UI experts whose mission is to frak up and turn a usable desktop environment into whatever they imagined while under the influence.
On 1/10/22 06:52, Sam Varshavchik wrote:
Joe Zeff writes:
On 1/8/22 9:03 PM, Robert Moskowitz wrote:
One cute comment I saw that Xfce will have to be called 'E', as it is no longer 'fc' and once it jettisons X, well...
Xfce hasn't stood for anything in particular for many years now.
Which is precisely the reason why xfce is a breath of fresh air. It just works. It does what a usable desktop should do: give me desktop icons; a menu bar; a tray; a pager; and some dockable icons. That's it.
Its profile is low enough not to attract interest from self-appointed UI experts whose mission is to frak up and turn a usable desktop environment into whatever they imagined while under the influence.
https://wiki.xfce.org/releng/wayland_roadmap
Though I was told on the Xfce user list that there is no current work on this roadmap.
Robert Moskowitz writes:
https://wiki.xfce.org/releng/wayland_roadmap
Though I was told on the Xfce user list that there is no current work on this roadmap.
I am not surprised. It's hard for me to come up with a value-added benefit from Wayland that's visible to the end user. Oh, I'm sure there's a list, somewhere, of all the wonderful advantages of Wayland. But what exactly does it do, better, that the end user can immediately hold in their hand, and inspect?
I don't know the answer to this question. I am not involved with the xfce folks, but I am involved in a very related pet project, that also involves X innards. I'm pretty certain I understand why they don't seem to place much priority on Wayland. I can't quite see, myself, exactly why I will want to sink a massive amount of time into pretty much reimplementing what works perfectly adequately in X, in a completely different way with Wayland, and what that will allow me to do that I cannot do (very well) with X.
Plus, X11 has very, very useful built-in advantages. It is very easy to artificially constrain X's performance: tunnel X11 over ssh over a wifi connection. If you're doing low level development using X11 primitives, this setup makes any inefficiencies and bottlenecks in your use of X11 become very, very easy to see and troubleshoot.
Setting aside the fact that you can't do tunneling with Wayland (AFAIK), complaints about poor Gnome/Wayland performance, on older, less powerful hardware, happen quite often. This probably isn't because of any inherent defects in low-level layers, closer to the hardware, but rather to something dumb happening higher up in the protocol stack, but completely missed by someone who was developing it on a high-end, 64-core threadripper. I forecast rocky road ahead for Gnome/Wayland, with constant complaints about bad performance and stability. That's just the way it is.
On Mon, 10 Jan 2022 at 07:52, Sam Varshavchik mrsam@courier-mta.com wrote:
Joe Zeff writes:
On 1/8/22 9:03 PM, Robert Moskowitz wrote:
One cute comment I saw that Xfce will have to be called 'E', as it is
no
longer 'fc' and once it jettisons X, well...
Xfce hasn't stood for anything in particular for many years now.
Which is precisely the reason why xfce is a breath of fresh air. It just works. It does what a usable desktop should do: give me desktop icons; a menu bar; a tray; a pager; and some dockable icons. That's it.
Its profile is low enough not to attract interest from self-appointed UI experts whose mission is to frak up and turn a usable desktop environment into whatever they imagined while under the influence.
I don't think that is a realistic description of the effort that goes into DE designs. Linux developers have to have a target class of users in mind, and also try to make it easy for users to migrate from Apple and Microsoft DE's. Years ago at work, we had users whose primary system was Apple or Windows and needed to use software running on a mini-super. Our Xterminals were not popular with that group, so we introduced a few NeXT systems and found that both groups of users were happy to work with the NeXT systems.
There is actually a lot of research behind Microsoft and Apple user interface designs, including focus groups and testing. In my field (remote sensing) I encounter users trained in agronomy or marine biology whose prior exposure to computing was their smart phone and game consoles. If you suggest they copy and paste the text of an error message into an online forum post, you get a screen capture. They have never encountered a terminal and many have difficulty filling in text boxes in GUI applications -- they want a pull-down menu where they can choose an item.
One of the "mission critical" applications from NASA is a set of command-line programs that were developed on SGI IRIX64 and now run on linux and macOS. NASA added a capability for the application to dump an XML file with details of the command-line options and then used these to add a menu item with a section for each program to an existing Java application. The user fills in forms and the Java application constructs a command line which can be run locally on linux or passed to a remote server. This allows many users who would never get started if a command-line was the only way to run the programs. Once started, they want to run the same process on 100's to 1000's of files, and will be willing to make the step to batch processing. Many younger users learned Python in university and now do batch processing with a Python IDE.
For many users, this fell apart when they were suddenly forced to work remotely after relying on using systems configured by others at their workplace.
Many posts in this forum are seeking help with problems involving desktop configuration, which shows that there is a big gap between UI design and design for fixing broken systems. Many of the younger users will simply reinstall the OS and apps when something breaks. Apple and Microsoft both provide cloud storage that preserves user files and settings across installs, so for many users this actually works as long as you are willing to live with a basic configuration and minimal tweaks, but if the breakage is a reproducible bug they get stuck in an endless loop until Dr. Google leads them to the fix.
On Mon, 10 Jan 2022 10:01:22 -0400 George N. White III wrote:
Linux developers have to have a target class of users in mind
Yes, a completely imaginary class of users they invent in their head to justify whatever insanity they decide they want to implement, so when people object they can say, "but this is what the users want".
On 1/10/22 09:31, Tom Horsley wrote:
On Mon, 10 Jan 2022 10:01:22 -0400 George N. White III wrote:
Linux developers have to have a target class of users in mind
Yes, a completely imaginary class of users they invent in their head to justify whatever insanity they decide they want to implement, so when people object they can say, "but this is what the users want".
Focus groups are strange beasts. I have been on a few over the years.
One of the big challenges is getting a good cross-section of users to get decent advise.
One 'classic' that I was involved with is back around '94 Ford's Windstar focus group said a driver-side passenger was not needed. The Chrysler Caravan focus group said this was one of the top items needed in the ng.
This was one of the features that buried the Windstar for the '96 model year. Chrysler had a great ad of a dolphin jumping through a Caravan down in the Keys.
All to how to create a focus group. But then, you have to be willing to listen and have the resources to create said focus group.
So we get what we get in Linux UI. "You want something different? Code it up and show us."
Sigh.
On 1/10/22 06:31, Tom Horsley wrote:
On Mon, 10 Jan 2022 10:01:22 -0400 George N. White III wrote:
Linux developers have to have a target class of users in mind
Yes, a completely imaginary class of users they invent in their head to justify whatever insanity they decide they want to implement, so when people object they can say, "but this is what the users want".
Except that it works very well for a lot of people. I have introduced many people to Linux with Gnome and I don't get complaints. I even had one woman ask if I could put it on her Macbook (I didn't try that). When I first saw the new Gnome demos a few years ago, I was very skeptical, but I gave it a try and I do like it. It was a little rough at first, but it has been improving rapidly and it's very good now.
Maybe it's not for you or maybe you haven't given it a decent try, but either way, that's not an excuse for you to put down the developers and be very rude about it repeatedly. And that goes for some others that have commented in this and similar threads.
On 1/10/22 16:16, Samuel Sieb wrote:
On 1/10/22 06:31, Tom Horsley wrote:
On Mon, 10 Jan 2022 10:01:22 -0400 George N. White III wrote:
Linux developers have to have a target class of users in mind
Yes, a completely imaginary class of users they invent in their head to justify whatever insanity they decide they want to implement, so when people object they can say, "but this is what the users want".
Except that it works very well for a lot of people. I have introduced many people to Linux with Gnome and I don't get complaints. I even had one woman ask if I could put it on her Macbook (I didn't try that). When I first saw the new Gnome demos a few years ago, I was very skeptical, but I gave it a try and I do like it. It was a little rough at first, but it has been improving rapidly and it's very good now.
Maybe it's not for you or maybe you haven't given it a decent try, but either way, that's not an excuse for you to put down the developers and be very rude about it repeatedly. And that goes for some others that have commented in this and similar threads.
And in all fairness, I jettisoned Gnome after some time with v3 and how slow it was making my old systems. I had started working with Xfce on Fedora-arm on 32bit Allwinner chips. So I made the switch on my main notebook and have not even looked at Gnome since.
I have often said to friends, that you choose which distro you want to spend time with and anything else is pulled from surplus cycles.
I did not like the slowness and complexity of Gnome3 and have no idea which way it has gone since those first Fedora versions that used it. At times I complain about Xfce, but most are minor or I have to learn a new tweak.
Those of you that have the ability or need to work with multiple UIs and ver of Linux, hats off to you!
On 1/10/22 13:58, Robert Moskowitz wrote:
And in all fairness, I jettisoned Gnome after some time with v3 and how slow it was making my old systems. I had started working with Xfce on Fedora-arm on 32bit Allwinner chips. So I made the switch on my main notebook and have not even looked at Gnome since.
That's fine, Gnome really isn't intended for those systems anyway. Gnome-shell requires a reasonable 3d-capable video card to run well. I have a bunch of old P4 computers in a school for students to use. I set them up with Mate which works great there.
On Mon, 2022-01-10 at 14:07 -0800, Samuel Sieb wrote:
Gnome really isn't intended for those systems anyway. Gnome-shell requires a reasonable 3d-capable video card to run well. I have a bunch of old P4 computers in a school for students to use. I set them up with Mate which works great there.
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
If I was doing fancy graphics in the programs that I use, then I'd consider spending an outrageous amount of money on the graphics card. But not just for the sake of the desktop interface.
On 1/10/22 22:10, Tim via users wrote:
On Mon, 2022-01-10 at 14:07 -0800, Samuel Sieb wrote:
Gnome really isn't intended for those systems anyway. Gnome-shell requires a reasonable 3d-capable video card to run well. I have a bunch of old P4 computers in a school for students to use. I set them up with Mate which works great there.
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
To prepare for apps that need beefy graphics cards. If the apps potentially are going to need it, use it also.
If I was doing fancy graphics in the programs that I use, then I'd consider spending an outrageous amount of money on the graphics card. But not just for the sake of the desktop interface.
So you go to a UI that is built on the assumption that there will be no beefy graphics apps. That is why I realized I did not need where Gnome was heading. I do not do heavy graphics like my oldest son that is a mechanical engineer and has to use all those beefy 3D tools practically all the time he is at his system.
On 1/10/22 19:10, Tim via users wrote:
On Mon, 2022-01-10 at 14:07 -0800, Samuel Sieb wrote:
Gnome really isn't intended for those systems anyway. Gnome-shell requires a reasonable 3d-capable video card to run well. I have a bunch of old P4 computers in a school for students to use. I set them up with Mate which works great there.
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
If I was doing fancy graphics in the programs that I use, then I'd consider spending an outrageous amount of money on the graphics card. But not just for the sake of the desktop interface.
Whatever graphics card you have, whether integrated or discrete, will work, unless you have a really old computer. Like I mentioned in earlier, I run Mate on some old P4s from 2005. That's fine, they can't hold enough RAM for much else anyway.
I don't understand why the few of you in this thread feel the need to exaggerate so badly in order to put down Gnome. I get that you don't like it for whatever reason, but you haven't actually given any reason other than it's different and you don't like it. Your comments about the graphics requirements are just hyperbole. The only current usage where it might matter is in a VM, but then why are you running a Gnome desktop in a VM? I even setup a vnc login for my son to use temporarily and that also used Gnome and worked reasonably well.
Tim:
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
Samuel Sieb:
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
Yeah, right. Slowly opening menus, slowly spinning desktop cubes, hover and wait before continuing, splash screens and other animations that delay me doing something "make it work better"?
I remember when using computers on just a few megahertz, now that we're using gigahertz computers the computers should be a thousand times faster to use, but they're not.
And, sure, I can turn off the decorative crap. But it's still the using the same bloated behemoth underneath.
I don't understand why the few of you in this thread feel the need to exaggerate so badly in order to put down Gnome.
Oh, bullcrap. We're not exaggerating. We've used the various desktops, we've seen the silliness and ignored the Emperors New Clothes sales pitch. The excessive over-the-top Gnome defending against evidence to the contrary, and putting down of anybody who criticises it, really pisses me off.
Tim via users writes:
Tim:
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
Samuel Sieb:
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
Yeah, right. Slowly opening menus, slowly spinning desktop cubes, hover and wait before continuing, splash screens and other animations that delay me doing something "make it work better"?
I agree. Having said that: if Gnome wants to target the power user, with the latest high-end video hardware, the kind who follows the latest UI trends, then I see nothing wrong with that, with Gnome becoming a boutique, specialty UI that targets a specific userbase.
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
On Tue, 2022-01-11 at 07:15 -0500, Sam Varshavchik wrote:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
As a KDE user, I'm holding off from Wayland until it actually works correctly. Too many bugs at the moment, my particular showstopper being the buggy session restore code.
poc
On 1/11/22 07:15, Sam Varshavchik wrote:
Tim via users writes:
Tim:
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
Samuel Sieb:
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
Yeah, right. Slowly opening menus, slowly spinning desktop cubes, hover and wait before continuing, splash screens and other animations that delay me doing something "make it work better"?
I agree. Having said that: if Gnome wants to target the power user, with the latest high-end video hardware, the kind who follows the latest UI trends, then I see nothing wrong with that, with Gnome becoming a boutique, specialty UI that targets a specific userbase.
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
Much of Fedora is user developed. So this need not be the case. It is clear from what I read over in Xfce-land that better tools are needed. Probably to even better migrate Gnome and KDE to Wayland.
This is a bigger change than systemd and firewalld. But we made those. We will be on Wayland across the board, it seems, in 5 years or less.
On Tue, 11 Jan 2022 at 09:20, Robert Moskowitz rgm@htt-consult.com wrote:
On 1/11/22 07:15, Sam Varshavchik wrote:
Tim via users writes:
Tim:
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
Samuel Sieb:
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
Yeah, right. Slowly opening menus, slowly spinning desktop cubes, hover and wait before continuing, splash screens and other animations that delay me doing something "make it work better"?
I agree. Having said that: if Gnome wants to target the power user, with the latest high-end video hardware, the kind who follows the latest UI trends, then I see nothing wrong with that, with Gnome becoming a boutique, specialty UI that targets a specific userbase.
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
Much of Fedora is user developed. So this need not be the case. It is clear from what I read over in Xfce-land that better tools are needed. Probably to even better migrate Gnome and KDE to Wayland.
This is a bigger change than systemd and firewalld. But we made those. We will be on Wayland across the board, it seems, in 5 years or less.
Wayland will push app devs to use big toolkits so they don't have to work with low-level API's. This will still be a lot of work for older apps that rely directly on X11, and may force UI changes. Non-Western developers are becoming a major force but have less experience with legacy X11 apps so may prefer developing new apps from scratch over migrating existing apps to Wayland. They may, however, be big contributors to Wayland toolkits as they encounter bugs and coverage gaps.
-- George N. White III
Robert Moskowitz writes:
On 1/11/22 07:15, Sam Varshavchik wrote:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
Much of Fedora is user developed. So this need not be the case. It is clear from what I read over in Xfce-land that better tools are needed. Probably to even better migrate Gnome and KDE to Wayland.
This is a bigger change than systemd and firewalld. But we made those. We will be on Wayland across the board, it seems, in 5 years or less.
I'm going to reserve the right to pick at the claim that "Fedora is user developed", if this means that what goes into Fedora depends on what users want. I do not recall a great clamoring for some of the things that you mention, and notable amount of opposition from the community. But that's a different topic. On the subject matter of time estimates I don't see this happening in 5 years, if at all. I'm going to go with 7-8 at the least. If XFCE does not support Wayland at that point, I guess that'll be the end of the XFCE spin.
On Tue, 11 Jan 2022 at 18:39, Sam Varshavchik mrsam@courier-mta.com wrote:
Robert Moskowitz writes:
On 1/11/22 07:15, Sam Varshavchik wrote:
Sadly, I expect that Fedora at some point will become exclusively Gnome
and
KDE, because only these stacks will support Wayland, in order to ditch
X,
and also target the same userbase. This won't happen anytime soon, but
it
will happen.
Much of Fedora is user developed. So this need not be the case. It is clear from what I read over in Xfce-land that better tools are needed. Probably to even better migrate Gnome and KDE to Wayland.
This is a bigger change than systemd and firewalld. But we made those.
We
will be on Wayland across the board, it seems, in 5 years or less.
I'm going to reserve the right to pick at the claim that "Fedora is user developed", if this means that what goes into Fedora depends on what users want.
"user developed" means stuff users want badly enough to contribute in the form of code, patches, bug reports, and documentation.
I do not recall a great clamoring for some of the things that you
mention, and notable amount of opposition from the community. But that's a different topic. On the subject matter of time estimates I don't see this happening in 5 years, if at all. I'm going to go with 7-8 at the least. If XFCE does not support Wayland at that point, I guess that'll be the end of the XFCE spin.
Or earlier if some new Wayland-based project that improves on XFCE (together with English translation of the docs from the author's native language!) gains traction.
George N. White III writes:
Much of Fedora is user developed. So this need not be the case. It is clear from what I read over in Xfce-land that better tools are needed. Probably to even better migrate Gnome and KDE to Wayland.
This is a bigger change than systemd and firewalld. But we made those.
We
will be on Wayland across the board, it seems, in 5 years or less.
I'm going to reserve the right to pick at the claim that "Fedora is user developed", if this means that what goes into Fedora depends on what users want.
"user developed" means stuff users want badly enough to contribute
in the form of code, patches, bug reports, and documentation.
I originally wrote a different snarky response to this, but thought better and deleted it. I really, really didn't want to go there. Let's just say that if I were to call for a show of hands: who, around here, actually contributed "code, patches, bug reports, and documentation" to the project in question? Well, that would've been one hell of a curve-ball over the batter's box.
In fact, I'll bet more people here know exactly what I was going to say, more people than the ones who contributed "code, patches, … and documentation". Not sure about the bug reports, though. However: the bug reports were met with hostile responses more often than not. So, I don't think they should count.
I do not recall a great clamoring for some of the things that you mention, and notable amount of opposition from the community. But that's a different topic. On the subject matter of time estimates I don't see this happening in 5 years, if at all. I'm going to go with 7-8 at the least. If XFCE does not support Wayland at that point, I guess that'll be the end of the XFCE spin.
Or earlier if some new Wayland-based project that improves on XFCE (together
with English translation of the docs from the author's native language!) gains
traction.
That's an interesting thought experiment: what would be the chances of the Wayland stake-holders going out and choosing to contribute to XFCE's migration to Wayland?
Tim,
Yeah, right. Slowly opening menus, slowly spinning desktop cubes, hover and wait before continuing, splash screens and other animations that delay me doing something "make it work better"?
Sam Varshavchik:
I agree. Having said that: if Gnome wants to target the power user, with the latest high-end video hardware, the kind who follows the latest UI trends, then I see nothing wrong with that, with Gnome becoming a boutique, specialty UI that targets a specific userbase.
It was possible to add those features without making the underlying system sluggish, it was done before. Now, you have a behemoth that you can turn off the whizzy bits, but still uses too much grunt to show the desktop in a basic manner.
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a ton of features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If they don't reimplement them, why would those people want to use it?
Tim via users writes:
Sam Varshavchik:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a ton of features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If they don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems. The stated problems with X, that were the purported drivers for Wayland -- I just can't find those problems, myself.
But it's painfully obvious that every effort is being made to ditch X in favor of the Next Greatest Thing. It's a path well trodden by Gnome 3 and systemd. I'm just a realist here, and I see the handwriting on the wall.
The only thing that will keep X in business is a popular widget set or a desktop that does not get ported to Wayland. I'm keeping my fingers crossed that xfce will be enough to keep X around. Maybe not in Fedora, but in other distributions.
On Wed, 12 Jan 2022 at 18:48, Sam Varshavchik mrsam@courier-mta.com wrote:
Tim via users writes:
Sam Varshavchik:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a ton of features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If they don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems.
Read https://www.kernel.org/doc/ols/2004/ols2004v1-pages-227-238.pdf
This paper describes the problems Wayland is trying to solve.
The stated problems with X, that were the purported drivers for Wayland -- I just can't find those problems, myself.
Other people can. Just because X works for you doesn't mean Wayland doesn't make linux suitable for use cases where X fails.
But it's painfully obvious that every effort is being made to ditch X in favor of the Next Greatest Thing. It's a path well trodden by Gnome 3 and systemd. I'm just a realist here, and I see the handwriting on the wall.
Real money is being spent to develop Wayland because the need for better linux graphics justifies the effort.
The only thing that will keep X in business is a popular widget set or a desktop that does not get ported to Wayland. I'm keeping my fingers crossed that xfce will be enough to keep X around. Maybe not in Fedora, but in other distributions.
Most likely scenario is a new lightweight DE using Wayland, maybe stealing code from xfce or maybe completely written from scratch.
George N. White III writes:
On Wed, 12 Jan 2022 at 18:48, Sam Varshavchik <<URL:mailto:mrsam@courier- mta.com>mrsam@courier-mta.com> wrote:
Tim via users writes:
Sam Varshavchik:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a ton of features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If they don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems.
Read <URL:https://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdf>https://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdf
This paper describes the problems Wayland is trying to solve.
I skimmed through it. I didn't read anything that precludes an implementation as an X protocol revision. Ditch last century's baggage (colormaps, fonts); use RENDER exclusively; replace the archaic COMPOUND_TEXT with UTF-8; etc…
On Wed, 12 Jan 2022 at 23:18, Sam Varshavchik mrsam@courier-mta.com wrote:
George N. White III writes:
On Wed, 12 Jan 2022 at 18:48, Sam Varshavchik <<URL:mailto:
mrsam@courier-
mta.com>mrsam@courier-mta.com> wrote:
Tim via users writes:
Sam Varshavchik:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a
ton of
features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If
they
don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems.
Read URL:https://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdfhttps://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdf
This paper describes the problems Wayland is trying to solve.
I skimmed through it. I didn't read anything that precludes an implementation as an X protocol revision. Ditch last century's baggage (colormaps, fonts); use RENDER exclusively; replace the archaic COMPOUND_TEXT with UTF-8; etc…
The people who do the work get to choose how they do it. From my perspective, the biggest issue with Wayland has lack of support from Nvidia, which is really not the responsibility of Wayland developers.
George N. White III writes:
On Wed, 12 Jan 2022 at 23:18, Sam Varshavchik <<URL:mailto:mrsam@courier- mta.com>mrsam@courier-mta.com> wrote:
George N. White III writes:
On Wed, 12 Jan 2022 at 18:48, Sam Varshavchik
<<URL:mailto:URL:mailto:mrsam@courier-mrsam@courier-
URL:http://mta.commta.com><URL:mailto:mrsam@courier-
mta.com>mrsam@courier-mta.com> wrote:
Tim via users writes:
> Sam Varshavchik: > > Sadly, I expect that Fedora at some point will become exclusively > > Gnome and KDE, because only these stacks will support Wayland, in > > order to ditch X, and also target the same userbase. This won't > > happen anytime soon, but it will happen. > > I'm still not convinced Wayland's a great idea. The old X had a
ton of
> features that people wanted, but Wayland doesn't. *If* they > reimplement them, how's Wayland going to be different from X? If
they
> don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems.
Read <URL:URL:https://www.kernel.org/doc/ols/2004/ols2004v1- https://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdf<URL:https://www.kernel.org/doc/ols/2004/ols2004v1- https://www.kernel.org/doc/ols/2004/ols2004v1- pages-227-238.pdf
This paper describes the problems Wayland is trying to solve.
I skimmed through it. I didn't read anything that precludes an implementation as an X protocol revision. Ditch last century's baggage (colormaps, fonts); use RENDER exclusively; replace the archaic COMPOUND_TEXT with UTF-8; etc…
The people who do the work get to choose how they do it. From my perspective, the biggest issue with Wayland has lack of support from Nvidia, which is really not the responsibility of Wayland developers.
Oh, yeah. That's going to do that. Without a buy-in from Nvidia, Wayland's going nowhere.
On Wed, Jan 12, 2022 at 10:44:59PM -0400, George N. White III wrote:
On Wed, 12 Jan 2022 at 18:48, Sam Varshavchik mrsam@courier-mta.com wrote:
Tim via users writes:
Sam Varshavchik:
Sadly, I expect that Fedora at some point will become exclusively Gnome and KDE, because only these stacks will support Wayland, in order to ditch X, and also target the same userbase. This won't happen anytime soon, but it will happen.
I'm still not convinced Wayland's a great idea. The old X had a ton of features that people wanted, but Wayland doesn't. *If* they reimplement them, how's Wayland going to be different from X? If they don't reimplement them, why would those people want to use it?
I couldn't agree more. Wayland is a solution in search of problems.
Read https://www.kernel.org/doc/ols/2004/ols2004v1-pages-227-238.pdf
This paper describes the problems Wayland is trying to solve.
I think https://www.phoronix.com/scan.php?page=article&item=x_wayland_situation&... is a more accessable listing of the reasons wayland exists.
Granted it's from almost 9 years ago now, but it does give some interesting reasoning IMHO.
kevin
On Tue, 11 Jan 2022 at 01:57, Samuel Sieb samuel@sieb.net wrote:
On 1/10/22 19:10, Tim via users wrote:
On Mon, 2022-01-10 at 14:07 -0800, Samuel Sieb wrote:
Gnome really isn't intended for those systems anyway. Gnome-shell requires a reasonable 3d-capable video card to run well. I have a bunch of old P4 computers in a school for students to use. I set them up with Mate which works great there.
I find this just crazy. Why should just the desktop interface require a beefy graphics card? I use Mate because of that silliness.
It doesn't need to be beefy. Pretty much any video card made in the last 10 years should work. Since the capability is there, why not use it to make the desktop work better?
If I was doing fancy graphics in the programs that I use, then I'd consider spending an outra
geous amount of money on the graphics card.
But not just for the sake of the desktop interface.
Whatever graphics card you have, whether integrated or discrete, will work, unless you have a really old computer. Like I mentioned in earlier, I run Mate on some old P4s from 2005. That's fine, they can't hold enough RAM for much else anyway.
I don't understand why the few of you in this thread feel the need to exaggerate so badly in order to put down Gnome. I get that you don't like it for whatever reason, but you haven't actually given any reason other than it's different and you don't like it. Your comments about the graphics requirements are just hyperbole. The only current usage where it might matter is in a VM, but then why are you running a Gnome desktop in a VM? I even setup a vnc login for my son to use temporarily and that also used Gnome and worked reasonably well.
With commercial OS's you are pretty much stuck with this year's desktop, and each new version gets similar complaints from certain groups of users.
One advantage of linux is that you can choose a UI you like and stick with it for years. Complaining that Gnome is horrible only serves to trash the reputation of linux and linux users. What would be more helpful is some insight into use cases (e.g., ancient hardware) where other desktops have advantages over Gnome.
The diversity of linux is confusing to new users. We need to do a better job of explaining that linux allows people to develop UI's that are better suited to a particular use case that the alternatives. In the academic world, students start out on the distro and desktop that faculty are using, and often stick with that for years afterward because colleagues in their field are using the same configuration and workflows tuned to that environment. In economics this is called a "network benefit". When new users who need to run a linux application in my field ask me which the "best" linux distro, I tell them to check around at their institution to see what distros are popular. Some institutions have local mirrors of certain distros and documents for access to file servers, VPN, etc. for those distros. If you have peers using the same distro, they can often help you sort out glitches because they have seen them before.
George N. White III writes:
One advantage of linux is that you can choose a UI you like and stick with it for years. Complaining that Gnome is horrible only serves to trash the reputation of linux and linux users. What would be more helpful is some insight into use cases (e.g., ancient hardware) where other desktops have advantages over Gnome.
I think there's a middle ground where legitimate criticism lives. When Gnome jumped my shark I simply switched desktops and that was that. But I still think that explaining the reasons why is useful, and pretending that Gnome is not a completely different UI paradigm is not useful, and this is a legitimate point of discussion.
I use the Windows 10 generally. I briefly poked the tires on the current, default Gnome desktop in Ubuntu, last year. There were some claims that Windows is moving closer to Gnome's UI paradigms, but I just didn't see that. Perhaps that was referring to Windows 11; if so then I'm wrong on this, but I am convinced that average Windows users will have absolutely no clue what to do with the Gnome desktop, when they sit down in front of it. Is that a good thing?
On 1/11/22 05:18, Sam Varshavchik wrote:
I use the Windows 10 generally. I briefly poked the tires on the current, default Gnome desktop in Ubuntu, last year. There were some claims that Windows is moving closer to Gnome's UI paradigms, but I just didn't see that. Perhaps that was referring to Windows 11; if so then I'm wrong on this, but I am convinced that average Windows users will have absolutely no clue what to do with the Gnome desktop, when they sit down in front of it. Is that a good thing?
Have you missed my comments about giving "average" users, mostly school teachers, but a few others as well, laptops installed with the default Gnome desktop? I give them a couple of pointers, like how to access the overview, and they don't have any problem with it. Like I said earlier, one person even asked if I could put it on their macbook.
I am a technical user. I've been using Linux since Redhat 5. I've used many different window managers and I was skeptical about the big Gnome 3 change, but I tried it out. I think I switched from KDE at that time. I'm *very happy* with how Gnome works. I'm not an "apologist" for Gnome, I'm just really sick of people making such disgusting comments about it and the people developing it (and other software in the stack as well). Just because you happen to not like it for whatever reasons doesn't mean that it's not good for many other people. No one is forcing you to use it.
On Tue, Jan 11, 2022 at 04:46:16PM -0800, Samuel Sieb wrote:
[ .... ]
I am a technical user. I've been using Linux since Redhat 5. I've used many different window managers and I was skeptical about the big Gnome 3 change, but I tried it out. I think I switched from KDE at that time. I'm *very happy* with how Gnome works. I'm not an "apologist" for Gnome, I'm just really sick of people making such disgusting comments about it and the people developing it (and other software in the stack as well). Just because you happen to not like it for whatever reasons doesn't mean that it's not good for many other people. No one is forcing you to use it.
+1
And yes: I'm certainly not a Fedora aficionado. There are lots of things I don't like on this OS, or on other Linux systems. I won't go into details, for one simple reason: I'm free to use something else, if I don't like what I have. Plus: admittedly I never was a convinced Linux user. The reason I started Linux around 2000 first time was the fact I didn't have - or better: know - anything else that came close to my interests for software, which were - still are - Unix-like systems.
But I find it sometimes puzzling, annoying, sometimes even boring, when people who really could know better, complain that Linux isn't the *nix-like, light-weight or whatever system that it probably was some decades ago. Especially in a situation where there are other software options out there.
Actually I'm even sort of grateful that Linux changed the way it did over the last few years, up to the point when I more or less had to look around for - and in the end find - some different from a modern Linux system.
One last thing. There are people out there who seem to like ad hominem attacks in threads like this one. Could you guys please stop that? As Grandma' reportedly once put it: "If you can't say it nicely, don't say anything at all."
Regards, Wolfgang
On Tue, 2022-01-11 at 08:49 -0400, George N. White III wrote:
Complaining that Gnome is horrible only serves to trash the reputation of linux and linux users.
It wouldn't get the complaints if it weren't...
And for a tangential example, it wasn't the people complaining about the crappiness of Windows that were trashing its reputation, Microsoft trashed its own reputation by inflicting their bad decisions on everyone (even those who don't use their systems).
On Wed, 12 Jan 2022 at 05:30, Tim via users users@lists.fedoraproject.org wrote:
On Tue, 2022-01-11 at 08:49 -0400, George N. White III wrote:
Complaining that Gnome is horrible only serves to trash the reputation of linux and linux users.
It wouldn't get the complaints if it weren't...
And for a tangential example, it wasn't the people complaining about the crappiness of Windows that were trashing its reputation, Microsoft trashed its own reputation by inflicting their bad decisions on everyone (even those who don't use their systems).
There are constructive complaints that point out specific things that should be changed, but too many vague "Gnome is horrible" complaints. Describing why Xfce, etc are better for some use cases may encourage someone to start work on a new DE. That someone may be an experienced wayland dev or someone just starting out and looking for a project.
Robert Moskowitz writes:
I did not like the slowness and complexity of Gnome3 and have no idea which way it has gone since those first Fedora versions that used it. At times I complain about Xfce, but most are minor or I have to learn a new tweak.
I found the Xfce community to be very friendly and receptive. One time I stumbled across an xfwm bug. I did put a bit of elbow grease into a reproducer, but it was fixed quickly after I reported it.
In contrast, my break with Gnome happened about a year after a release that suddenly decided that my MP3 player turned into an MTP device, instead of the usb-storage it always was. The reported bug gathered dust without a single response, for a year, when I discovered by accident that XFCE had no issues mounting the same player as usb-storage, while the Gnome desktop next to it insisted it was an MTP device.
George N. White III writes:
I don't think that is a realistic description of the effort that goes into DE designs. Linux developers have to have a target class of users in mind, and also try to make it easy for users to migrate from Apple and Microsoft DE's.
I agree. Now, it would be interesting to ask someone, who worked on a Windows or an Apple desktop for years, to sit down in front of Gnome 3; and ask them to do something simple. I'm curious to see what the results are.
I'm going to add a disclaimer: I haven't looked into the state of the Gnome world in a number of years. Perhaps things have moved closer to the pre- Gnome 3 state. But I doubt it. The day job uses Ubuntu, and a year ago I was due for a new laptop, so they shipped me a brand new one, with Ubuntu 20. I took a very, very brief look at Gnome (before installing the XFCE desktop), and it was pretty much a collection of very unique UI concepts that I remembered from the initial days of Gnome 3. I had to run a marathon sprint with my mouse to do anything; first move the mouse to one corner to open the activities page, then move the mouse to the other side of the screen to the right icon, then finally click it.
Before risking carpal tunnel I quickly installed the xfce desktop and created plain, garden variety, permanent app shortcuts on the desktop.
There is actually a lot of research behind Microsoft and Apple user interface designs, including focus groups and testing. In my field (remote sensing) I
So, can someone point to me at the results of this research, in Windows, or MacOS, that looks like Gnome 3 and its apparently default workflow?
On 1/10/22 14:48, Sam Varshavchik wrote:
George N. White III writes:
I don't think that is a realistic description of the effort that goes into DE designs. Linux developers have to have a target class of users in mind, and also try to make it easy for users to migrate from Apple and Microsoft DE's.
I agree. Now, it would be interesting to ask someone, who worked on a Windows or an Apple desktop for years, to sit down in front of Gnome 3; and ask them to do something simple. I'm curious to see what the results are.
I have done this many times. A couple of quick explanations about how it works and generally no further issues.
I'm going to add a disclaimer: I haven't looked into the state of the Gnome world in a number of years. Perhaps things have moved closer to the pre-Gnome 3 state. But I doubt it. The day job uses Ubuntu, and a year ago I was due for a new laptop, so they shipped me a brand new one, with Ubuntu 20. I took a very, very brief look at Gnome (before installing the XFCE desktop), and it was pretty much a collection of very unique UI concepts that I remembered from the initial days of Gnome 3. I had to run a marathon sprint with my mouse to do anything; first move the mouse to one corner to open the activities page, then move the mouse to the other side of the screen to the right icon, then finally click it.
Why are you using the mouse? Using the keyboard is *way* faster. :-)
There is actually a lot of research behind Microsoft and Apple user interface designs, including focus groups and testing. In my field (remote sensing) I
So, can someone point to me at the results of this research, in Windows, or MacOS, that looks like Gnome 3 and its apparently default workflow?
The very funny thing I've been noticing is that Windows and MacOS have become much closer to Gnome over the last few years, except not done quite as well. Mac is ok, it would probably be good if I spent the time to figure out more keyboard shortcuts, but I don't use it enough to be worth it. However it's much better than windows with its apparent indecision between old-style and new-style.
Samuel Sieb writes:
I'm going to add a disclaimer: I haven't looked into the state of the Gnome world in a number of years. Perhaps things have moved closer to the pre- Gnome 3 state. But I doubt it. The day job uses Ubuntu, and a year ago I was due for a new laptop, so they shipped me a brand new one, with Ubuntu 20. I took a very, very brief look at Gnome (before installing the XFCE desktop), and it was pretty much a collection of very unique UI concepts that I remembered from the initial days of Gnome 3. I had to run a marathon sprint with my mouse to do anything; first move the mouse to one corner to open the activities page, then move the mouse to the other side of the screen to the right icon, then finally click it.
Why are you using the mouse? Using the keyboard is *way* faster. :-)
Oh, and how would a user discover what those useful keyboard shortcuts are?
At this point, the basic UI elements are instinctively known to everyone. Everyone knows what an icon does, how menus look like, etc…
Does a majority people really know, instinctively, what keyboard shortcuts to use in Gnome?
I am a big fan of keyboard shortcuts. I spend a lot of time in emacs.
However, strangely, during my brief exposure to the latest, default Gnome desktop in Ubuntu I had no idea, whatsoever, that there were keyboard shortcuts available to me. How could I know that?
On 1/10/22 15:50, Sam Varshavchik wrote:
Samuel Sieb writes:
I'm going to add a disclaimer: I haven't looked into the state of the Gnome world in a number of years. Perhaps things have moved closer to the pre-Gnome 3 state. But I doubt it. The day job uses Ubuntu, and a year ago I was due for a new laptop, so they shipped me a brand new one, with Ubuntu 20. I took a very, very brief look at Gnome (before installing the XFCE desktop), and it was pretty much a collection of very unique UI concepts that I remembered from the initial days of Gnome 3. I had to run a marathon sprint with my mouse to do anything; first move the mouse to one corner to open the activities page, then move the mouse to the other side of the screen to the right icon, then finally click it.
Why are you using the mouse? Using the keyboard is *way* faster. :-)
Oh, and how would a user discover what those useful keyboard shortcuts are?
At this point, the basic UI elements are instinctively known to everyone. Everyone knows what an icon does, how menus look like, etc…
Does a majority people really know, instinctively, what keyboard shortcuts to use in Gnome?
I am a big fan of keyboard shortcuts. I spend a lot of time in emacs.
However, strangely, during my brief exposure to the latest, default Gnome desktop in Ubuntu I had no idea, whatsoever, that there were keyboard shortcuts available to me. How could I know that?
The same way you find instructions for any system or application. Look in the manual. There's a pretty decent help application for Gnome and for a new user, there's a brief introductory app that starts the first time they login.
Samuel Sieb writes:
On 1/10/22 15:50, Sam Varshavchik wrote:
Oh, and how would a user discover what those useful keyboard shortcuts are?
At this point, the basic UI elements are instinctively known to everyone. Everyone knows what an icon does, how menus look like, etc…
Does a majority people really know, instinctively, what keyboard shortcuts to use in Gnome?
I am a big fan of keyboard shortcuts. I spend a lot of time in emacs.
However, strangely, during my brief exposure to the latest, default Gnome desktop in Ubuntu I had no idea, whatsoever, that there were keyboard shortcuts available to me. How could I know that?
The same way you find instructions for any system or application. Look in the manual. There's a pretty decent help application for Gnome and for a new user, there's a brief introductory app that starts the first time they login.
Fair enough.
But it was easier for me to use XFCE right off the bat, it wasn't necessary to find its documentation and read XFCE's manual. I don't think there is one, actually.
Someone who's already used traditional desktop environments, with desktop shortcut icons, a taskbar (on top or the bottom), with something that looks like a "Start" menu, a tray, a pager, and a few other familiar UI icons – someone like that should be able to hit the ground running with XFCE.
On 1/10/22 17:38, Sam Varshavchik wrote:
Someone who's already used traditional desktop environments, with desktop shortcut icons, a taskbar (on top or the bottom), with something that looks like a "Start" menu, a tray, a pager, and a few other familiar UI icons – someone like that should be able to hit the ground running with XFCE.
Of course, if you're going to use the exact same interactions, then you probably don't need instructions. But if you always keep everything the same as it has always been, then where is the chance for improvement? Or do you think Windows 95 was the ultimate desktop interface and there can never be anything better? That's basically what you're describing. :-)
Sam Varshavchik:
Someone who's already used traditional desktop environments, with desktop shortcut icons, a taskbar (on top or the bottom), with something that looks like a "Start" menu, a tray, a pager, and a few other familiar UI icons – someone like that should be able to hit the ground running with XFCE.
Samuel Sieb:
Of course, if you're going to use the exact same interactions, then you probably don't need instructions.
If a graphical user interface needs instructions to be usable, it's failed in its creation. The whole point of having such an interface is that you can see what's on offer, and how to use it.
But if you always keep everything the same as it has always been, then where is the chance for improvement? Or do you think Windows 95 was the ultimate desktop interface and there can never be anything better? That's basically what you're describing. :-)
Don't ever hold up Windows as an example of good design! But the idea, which was *NOT* Windows invented, of organised menus, applications in individual windows, and a taskbar to control them, is one of the most productive interfaces.
Back to Gnome, I'd like to know just who thought redesigning the interface of a desktop computer to use the design of a touchscreen device, was a good idea. Very few desktop systems have touchscreens. Page after page of virtually randomly assorted icons (alphabetically sorting them by their wierd names) is useless. Terrible interface for multitasking.
On 1/10/22 22:19, Tim via users wrote:
Back to Gnome, I'd like to know just who thought redesigning the interface of a desktop computer to use the design of a touchscreen device, was a good idea. Very few desktop systems have touchscreens. Page after page of virtually randomly assorted icons (alphabetically sorting them by their wierd names) is useless. Terrible interface for multitasking.
Supposedly because that is what the new generation is trained on and expects.
Not like me that learned on a 55baud teletype...
On 1/10/22 19:19, Tim via users wrote:
Sam Varshavchik:
Someone who's already used traditional desktop environments, with desktop shortcut icons, a taskbar (on top or the bottom), with something that looks like a "Start" menu, a tray, a pager, and a few other familiar UI icons – someone like that should be able to hit the ground running with XFCE.
Samuel Sieb:
Of course, if you're going to use the exact same interactions, then you probably don't need instructions.
If a graphical user interface needs instructions to be usable, it's failed in its creation. The whole point of having such an interface is that you can see what's on offer, and how to use it.
Are you trying to tell me that if I gave a system running Mate to someone that has never used a computer, they would somehow automatically know how to use it?
But if you always keep everything the same as it has always been, then where is the chance for improvement? Or do you think Windows 95 was the ultimate desktop interface and there can never be anything better? That's basically what you're describing. :-)
Don't ever hold up Windows as an example of good design! But the idea, which was *NOT* Windows invented, of organised menus, applications in individual windows, and a taskbar to control them, is one of the most productive interfaces.
Do you have any evidence for that? But besides, Gnome still has most of that. I have no need for a menu of applications or a taskbar. I have more screen space and quicker access without them.
Back to Gnome, I'd like to know just who thought redesigning the interface of a desktop computer to use the design of a touchscreen device, was a good idea. Very few desktop systems have touchscreens. Page after page of virtually randomly assorted icons (alphabetically sorting them by their wierd names) is useless. Terrible interface for multitasking.
Are you aware that desktop systems are becoming a minority now. I had to convince my work to let me have a desktop instead of a laptop for software development. Almost everyone, in all the departments, uses a laptop now. The next generation is most familiar with phones and tablets. Even my wife prefers to use her phone to do everything instead of using a computer or laptop, which drives me a little nuts.
That list of applications can be arranged if you want. You can make folders and you can move the icons around. But I don't even use any of that. It's much faster to start applications with the keyboard. Press the Logo key, type a few characters, hit enter. With Mate, I would have to click on the applications menu, guess which category it would be under, carefully move the mouse to get the right menu, then click to start it. You're free to use whatever DE you want, but stop being so negative about Gnome. It's not helpful and what are you saying about all the people that actually like it and are very productive with it?
Tim:
If a graphical user interface needs instructions to be usable, it's failed in its creation. The whole point of having such an interface is that you can see what's on offer, and how to use it.
Samuel Sieb:
Are you trying to tell me that if I gave a system running Mate to someone that has never used a computer, they would somehow automatically know how to use it?
Do you think it'd be hard to work out? At some stage we all were first users. I never found it hard to work out how to use such an interface.
Phones and tablets, on the other hand, required a lot of hunting around all over the place.
Don't ever hold up Windows as an example of good design! But the idea, which was *NOT* Windows invented, of organised menus, applications in individual windows, and a taskbar to control them, is one of the most productive interfaces.
Do you have any evidence for that? But besides, Gnome still has most of that. I have no need for a menu of applications or a taskbar. I have more screen space and quicker access without them.
Windows never had organised menus, it was always all dumped into a disorganised clutter. I don't recall who came up with the taskbar first. And I saw applications in individual windows long before I saw it in Windows.
Are you aware that desktop systems are becoming a minority now. I had to convince my work to let me have a desktop instead of a laptop for software development.
That may be, but I've yet to encounter any touch screen laptops. They're a rarity. Putting a touch screen style interface on a non- touch screen is just stupidity.
That list of applications can be arranged if you want. You can make folders and you can move the icons around. But I don't even use any of that. It's much faster to start applications with the keyboard. Press the Logo key, type a few characters, hit enter.
I've never found that faster. And for one thing, you've got to know what to type. It doesn't help that many applications have lunatic names.
With Mate, I would have to click on the applications menu, guess which category it would be under, carefully move the mouse to get the right menu, then click to start it.
Seriously? You'd have to *guess* that office software might be in an office category? Categorised menus would be the *only* way that some people might find some program that they've never used, or heard of, before. It's either that, or just go around randomly running every application you find on your icon screen. Yeah, that's a really intelligent interface design, not.
And you're not forced to do that, either. You can dump a gazillion icons on the desktop if you want to. Or shortcuts on task bars.
Since most applications tend to be mouse operated, anyway, it's less painful to be swapping between mouse and keyboard and just keep using the same interface device.
You're free to use whatever DE you want, but stop being so negative about Gnome. It's not helpful and what are you saying about all the people that actually like it and are very productive with it?
Stop being a Gnome apologist. Current Gnome *is* a bad idea, Gnome defenders are the new flat-earthers. I'm quite sick of people who can't take criticism and keep la-la-laing with their fingers in the ears any time someone points out the crappiness of something, defending it's horrors to the death as the new way.
On 1/11/22 04:53, Tim via users wrote:
Tim:
If a graphical user interface needs instructions to be usable, it's failed in its creation. The whole point of having such an interface is that you can see what's on offer, and how to use it.
Samuel Sieb:
Are you trying to tell me that if I gave a system running Mate to someone that has never used a computer, they would somehow automatically know how to use it?
Do you think it'd be hard to work out? At some stage we all were first users. I never found it hard to work out how to use such an interface.
Phones and tablets, on the other hand, required a lot of hunting around all over the place.
My sons recognize that I am an old foggie that has to ask how things are done on my smartphone (currently Samsung Galaxy A01). ARGH!!!! :)
And trying to send a text message when the phone is busy 'helping' me with my spelling? And fixing all those Hebrew transliterations?
What we put up with!
Lots of years ago, my then current IT manager commented that this is probably the end of knowing the guts of what are in our desktop computers. At least my first CPS prof taught us about black boxes and choosing WHAT would be our black box; he helped build one of the first multi-user computers in the late '50s.
The current generation wizz along with their touch screens. It is understandable that the direction of the UI is for it to work well for their understanding. Unless they need something hefty to run AutoCAD or the such, they tend to work on tablets anyway and when they really need to type, they have their separate BT keyboard. This is where things are going. I am perfectly comfortable that the mainline UI is using these users as their focus. For the rest of us there is Xfce and the like.
Don't ever hold up Windows as an example of good design! But the idea, which was *NOT* Windows invented, of organised menus, applications in individual windows, and a taskbar to control them, is one of the most productive interfaces.
Do you have any evidence for that? But besides, Gnome still has most of that. I have no need for a menu of applications or a taskbar. I have more screen space and quicker access without them.
Windows never had organised menus, it was always all dumped into a disorganised clutter. I don't recall who came up with the taskbar first. And I saw applications in individual windows long before I saw it in Windows.
Are you aware that desktop systems are becoming a minority now. I had to convince my work to let me have a desktop instead of a laptop for software development.
That may be, but I've yet to encounter any touch screen laptops. They're a rarity. Putting a touch screen style interface on a non- touch screen is just stupidity.
That list of applications can be arranged if you want. You can make folders and you can move the icons around. But I don't even use any of that. It's much faster to start applications with the keyboard. Press the Logo key, type a few characters, hit enter.
I've never found that faster. And for one thing, you've got to know what to type. It doesn't help that many applications have lunatic names.
With Mate, I would have to click on the applications menu, guess which category it would be under, carefully move the mouse to get the right menu, then click to start it.
Seriously? You'd have to *guess* that office software might be in an office category? Categorised menus would be the *only* way that some people might find some program that they've never used, or heard of, before. It's either that, or just go around randomly running every application you find on your icon screen. Yeah, that's a really intelligent interface design, not.
And you're not forced to do that, either. You can dump a gazillion icons on the desktop if you want to. Or shortcuts on task bars.
Since most applications tend to be mouse operated, anyway, it's less painful to be swapping between mouse and keyboard and just keep using the same interface device.
You're free to use whatever DE you want, but stop being so negative about Gnome. It's not helpful and what are you saying about all the people that actually like it and are very productive with it?
Stop being a Gnome apologist. Current Gnome *is* a bad idea, Gnome defenders are the new flat-earthers. I'm quite sick of people who can't take criticism and keep la-la-laing with their fingers in the ears any time someone points out the crappiness of something, defending it's horrors to the death as the new way.
Samuel Sieb writes:
On 1/10/22 17:38, Sam Varshavchik wrote:
Someone who's already used traditional desktop environments, with desktop shortcut icons, a taskbar (on top or the bottom), with something that looks like a "Start" menu, a tray, a pager, and a few other familiar UI icons – someone like that should be able to hit the ground running with XFCE.
Of course, if you're going to use the exact same interactions, then you probably don't need instructions. But if you always keep everything the same as it has always been, then where is the chance for improvement? Or do you think Windows 95 was the ultimate desktop interface and there can never be anything better? That's basically what you're describing. :-)
A chance for what kind of improvement, specifically? If you believe you can show me some kind of an improvement, then show me. But then if I'm also told that I have to read the documentation and learn the right keyboard shortcuts to get anything accomplished, then I'm going to have a little bit of trouble getting convincened on what this end result is an improvement of.
This reminded me of one of the first "improvement"s that everyone experienced in the early days of Gnome 3: the mouse, and the scrollbar. Before Gnome 3 clicking on an open scrollbar area outside of the scrollbar's knob ended up scrolling the viewport exactly one page in the direction of the click. Everyone knew how that worked. But this was "improved" in Gnome 3. Now, doing the same warped the viewport to that absolute position. The former behavior, advance the viewport by one page, turned into a right mouse button click.
I was not the only one who called out this clusterfark. In response a claim was made that this is an "improvement" of some kind. But there was some struggle to explain exactly how that was an "improvement". There was some handwaving, citing to some mysterious UI studies cited, somewhere, that showed this to be the "right" UI. Finally an obscure setting was presented that restored the previous behavior. Here: do this, go back to your UI that worked this way for decades, have it your way, don't bother us because we're busy with more improvements.
I distinctly remember tweaking that knob even though I already used XFCE at that point, because XFCE is based on GTK. I just realized that late last year I reinstalled F34 from scratch, and I've been clicking my way on scrollbars without noticing anything amiss. I could be wrong, but I suspect that there was another improvement, at some point in the last 7-8 years, that improved on the original improvement.
Sake for the sake of change is not an improvement.
On 1/10/22 18:50, Sam Varshavchik wrote:
Samuel Sieb writes:
I'm going to add a disclaimer: I haven't looked into the state of the Gnome world in a number of years. Perhaps things have moved closer to the pre-Gnome 3 state. But I doubt it. The day job uses Ubuntu, and a year ago I was due for a new laptop, so they shipped me a brand new one, with Ubuntu 20. I took a very, very brief look at Gnome (before installing the XFCE desktop), and it was pretty much a collection of very unique UI concepts that I remembered from the initial days of Gnome 3. I had to run a marathon sprint with my mouse to do anything; first move the mouse to one corner to open the activities page, then move the mouse to the other side of the screen to the right icon, then finally click it.
Why are you using the mouse? Using the keyboard is *way* faster. :-)
Oh, and how would a user discover what those useful keyboard shortcuts are?
No. shortcuts are either remembered or sought after to ease the burden of hand movement. There is no visual roadmap.
At this point, the basic UI elements are instinctively known to everyone. Everyone knows what an icon does, how menus look like, etc…
Does a majority people really know, instinctively, what keyboard shortcuts to use in Gnome?
I am a big fan of keyboard shortcuts. I spend a lot of time in emacs.
However, strangely, during my brief exposure to the latest, default Gnome desktop in Ubuntu I had no idea, whatsoever, that there were keyboard shortcuts available to me. How could I know that?
users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
-
George N. White III -
Joe Zeff -
Kevin Fenzi -
Patrick O'Callaghan -
Robert Moskowitz -
Sam Varshavchik -
Samuel Sieb -
Tim -
Tom Horsley -
Wolfgang Pfeiffer