If you want thin, When CDE was new I ran FBWM. I discovered Windowmaker and then FLUXBOX after that. I still use FLUXBOX today when I need a very thin, fast GUI that gets out of my way so I can get work done.

GNOME has gotten bloated, but there are now 5 flavors of GNOME with different degrees of bloat. If you want a thinner, faster one you can get it by looking. Old KDE was bloated as well, and now it is one of the fastest and does not waste nearly as much space. If those do not float your boat there are 50 other options out there. (Although SOME wull not run right on Wayland even with XWayland, so you might be restricted to X.ORG, XWindows, or TinyX to have running under them. I do not consider that a hardship.)

With all of the available options, anyone complaining that they cannot find or make one to suite them is just lazy or ignorant. Both of those can be treated with a high degree of success. (As opposed to dishonesty or stupidity, but I do not suspect anyone here of either of those.)

PS: some projects are suffering bloat, corruption, or mismanagement because they have been invaded by people supporting monolithic and non-free alternatives who are willing to either take over GNU/FOSS projects and convert or kill them. FSF, Debian, and others are under attack. Canonical was never strong in FOSS support int he first place. I an not aware of GNOME being under such attack, but I would not rule it out without talking to the GNOME old guard.

No I use KDE. I only tried Gnome awhile back to see if it would even run on my Core2 laptop. Point is if the video is doctored then the whole article is suspect so it's pointless to read. When images and videos are added they have to match the article. People will naturally gravitate to the images first. If the images suggest something then it makes the rest of the article automatically biased.

It's misleading at best. At worst it is little more than a blatant attack on Gnome for personal reasons from someone with something to prove.

You can't mix fake with real. Credibility disappears.

GNOME’s software is Free Software: all our code is available for download and can be freely modified and redistributed according to the respective licenses.

Good news is anyone is able to fix what they want to fix.

Yeah in my early tears of being a Linux newbie, I was constantly using GNOME this was 20 +some odd years ago before my eyelids had been pulled back after I got lost one Saturday evening when I was 13 or 14 years old when I was just so happened to find both the Xfce and LXDE variants Ubuntu. That is to say I had already been dabbling around with Debian and other distributions that came with either of that provided one or both Xfce or LXDE

I am not sure what is the purpose of the article.
The only thing it shows is that your system has configuration issues. I did run Fedora/Gnome in VM and never seen lag. And this was on the laptop (2015 Intel). I would suggest optimising you hardware.
Ideally have same distro with kde installed alongside (not one distro with two DEs). I am not a fan of Gnome but I don't like rants. Do you have any advise for Gnome devs?

Also:
suggests VM in spite of .

This disqualifies your review as honest.

No, if they did not MENTION if the testing ws done in a VM or native iron THAT would be dishonest. We often test something, and if the result is less than optimal that way recommend running things in a VM where they cannot harm anything outside of their container. That is not dishonest, just smart.

Although I do agree that some politicians and many criminals have an issue with the distinction.

Shrug, if one compares underpowered environment i.e. opengl slow disk to anything bare metal, the this is dishonest. In part OP admits it pointing to deprecaded opengl version.
Opengl 2.1?
Which distro stopped at opengl 2.0? Except vbox.
No this is crap, dishonest review. I am not a fan of Gnome but is is low quality journalism. One can also attempt to install most of the linux distros sporting any DE on i486 with 8mb ram and complain about compete failure of linux.

@Aeterna, there's no dishonesty here. The fact that GNOME disables window animations when OpenGL 3.0 is not present is just a fact, whether you're running in a VM or real machine. There are some cases where OpenGL 3.0 won't be avaiable, and running on a VM is just one example. This whole animation thing in GNOME is just bad and power hungry.

That's because you probably haven't read it until the end nor this thread. Gonna try once again: GNOME is an ecosystem that goes way beyond the desktop environment, and because it's spread all over Linux distros, it has serious implications, and unfortunately GNOME is made by amateurs that are also very arrogant. So this is a problem to all people who use Linux, regardless if you like/use GNOME desktop environment or not. If you don't care about any of this, fine, but it doesn't mean the subject is not relevant to the community.


This is anecdotal evidence. The fact you don't perceive any lag doesn't prove anything. That's why I provided numbers and videos -- the machine used was the same as in the previous articles, also tested on lightning fast Ryzen 7840HS with the same results. If you find there's any hardware/distro issue on my side (tip: there isn't), you can make a video of your environment showing that.

https://www.linuxquestions.org/quest...es-4175710602/
already closed

There's always that link "My Blog" over on the right-hand side of the "My LQ" menu. Would it have been OK to point to that?

https://felipec.wordpress.com/2023/0...e-still-sucks/

Note, this thread seemed to largely miss the point, which is odd. I took the time to read most of the initial article, then the above, which was linked the article, and first, most of the complaints have zero to do with Gnome running in a vm, they are about concrete failures and bugs and issues.

All this is largely due to Gnome and GTK being under the IBM-redhat corporations umbrella, all the negative issues with developers etc are a direct result of that culture.

It seems to me that many simply missed the forest for a few single trees here.

I personally have nothing good to say about the recent corporate owned redhat division, and particularly not about Gnome.

The comments in the above link from Alan Cox, Linus Thorvalds, trying to get Poettering to grasp the value of incremental development (which takes skill and discipline) vs the break everything then force it on your users model, for example, to me are shining examples of the core culture these issues come from.

I personally have to admit to never understanding why the redhat corporation felt having a full desktop was that important, since the main role of rhel os is as a server system, but I guess the desire to emulate windows or osx in as many ways as possible was simply too tempting, despite the fact the group is simply too small re dev manhours available compared to those two mega corps.

Upthread there was the somewhat disingenuous statement that: it's free software, you can always fork it, which really is just an attempt to avoid the actual criticisms, one being the failure to handle bugs and issues in a reasonable and timely manner, so obviously if a bunch of paid devs can't handle or fix them, it would be unrealistic to assume a small group of forkers could deal with the code.

Speaking for myself only, I have totally given up on Gnome and KDE, I just got tired of them 'reinvisioning my desktop metaphor' without having the actual programming skill, manhours, or discipline, to actually carry that out in a seamless clean way, so I switched to i3 and xfce and am happy. Neither of those projects has any interest in reimagining without having the actual resources to do it my user experience so have been somewhat lovely to see evolve.

It's my inxi based research and understanding that as of now, KDE + Wayland is working quite well, in fact, talking to damentz recently of Liquorix kernel, he said he'd switched about a year or so ago and was finding it pretty good, and that fits with what i'm hearing, while gnome is having trouble with their wayland compositor (the corporate culture behind poor issue/bug report handling is hard to overcome, and isn't going to get better, as we saw with the recent gpl/rhel debacle, which is all anyone should need to see if they really want to understand what makes ibm-redhat tick today).

But I think spending a bit less time about the thing being a blog post, and a bit more actually reading it as a whole, probably would not have hurt.

I found the embedded links in particular interesting, the one above in particular, because they point to be historical nature of these issues, but are basically done in the present.

Luckily this is a basically free software ecosystem, with all its warts, so my solution is simply to avoid those 2 big desktops since both proved to me they can't be trusted with my desktop experience, so I don't bother with them except for testing the inxi desktop/wm detections etc in a vm, so they don't impact me, except the remarkably ugly gnome application interface, which I honestly thought was a bug when I first saw it because it was such an inferior design and implementation that I had a hard time believing anyone had actually done it on purpose. Now I try to not see it when it's there, but it always strikes me, maybe because i do gui web interface stuff as part of my job, as being a singularly inferior design decision, done by a group that clearly can't do ux/gui design at all, yet bumble and try anyway. That's a sort of professional surface appearance judgement by me, I'd never put something that ugly out and say its acceptable. Obviously that's sort of subjective, but I honestly don't think it is because there are standards out there, examples, norms, etc, and nobody is doing anything that ugly except gnome.

A long time ago a good programming teacher told me you can often judge the quality of a programs underlying code by the quality of its ux/gui, and I think that is very clearly the case here.

Anyway, to me, I don't care who wrote the original blog, I cared if it had information and links that were helpful to me, which it did, and those roughly confirmed my own understandings and experiences, particularly when it comes to dealing with projects that are basically owned by the redhat corporation.

I think the core concepts, which I follow somewhat religiously in all my professional and personal code and projects, is neither kde or gnome groups have the manpower to try to do the waterfall model of devlepment, but more important, it's a bad use of limited resources, incremental development, like we see with the linux kernel, openbsd, most of the non kde/gnome dekstops, uses available man hours very efficiently, and does not insult your user base by thrusting changes done for bad reasons with bad amateur usability people making decisions that impact more people than their own corporate culture internally.

I am increasingly finding that the best code and logic I do is done incrementally over many years, the only drawback there is as you get better over years, your old stuff needs to be upgraded to meet your new levels, which takes a lot of work, but that's the difference between incremental and waterfall, and it's why most free software projects should follow alan cox/linux thorvalds advice and take your users as having priority one, and the code should serve that goal, not the other way around. Poettering in that talk could not grasp that, because in my opinion he is not a good programmer or developer or project lead, gtk can't do it, not sure about modern qt.

As a good counterexample, xfce moved from gtk 2 to 3 and it was barely noticeable, the only change was those ugly gtk windows on gnome programs, which look and act broken, but I minimize my use of gnome and kde programs to avoid lockin to either culture, except a few key things. but oddly, kde integrates nicely, so they apparently have the skill to do that type of development, which fits with the reports of the better wayland experience on kde today than gnome, it suggests a better core project is doing the heavy lifting.

Xfce currently now is slowly moving towards wayland, but will not do so fully until they can have a clean transition, which is hard, but that's the difference. That's happening in 4.19 development version now. And they won't force something that doesn't work reliably on users, unlike what fedora is about to do with wayland gnome/kde.

But I do recognize that doing consistent incremental development is hard, and requires some care and focus and discipline, so I can easily see where projects that are paying people to do it will be less prone to doing that since those people care about their paychecks and next job, n ot the actual thing they are working on. Microsoft also had this problem, by the way, when I think it was vista was under development,they handed to core projects to groups of young developers, one was a new file system, which has never been released, never worked, and the other was vista. Both failed totally, and adults with experience had to come in to try to salvage the projects, which is why the vista release took so long. I think it was vista.

It's easy to see how these mistakes happen, particularly with corporate employees doing work because it's their job, but having no actual tie or passion for it beyond what they need to get their paychecks, so I'd tend to think these are more common errors in corporations than in real free open projects simply because the cost is too high, loss of users too severe, to make it a good idea.

I cannot speak to the origins of RHEL desktop, but when I was at the RHEL Summit Conference in about 2008 it was important because all of the RH employees there in development roles ran RHEL laptops and did development on them. They actually walked the walk in everything they did, and if it bugged them they tried to fix it before it could bug YOU!

I am not sure about IBM's RHEL support team, and no one there is answering (possibly stupid) questions right now.

1 members found this post helpful.