If you’ve been following this blog, you’ll know that a team has been working on updated designs for the Activities Overview. (Previous posts on this topic covered our initial motivations and design goals, as well as the results from some early exploratory research that we conducted.) This initiative has been the subject of significant activity over recent months, and we’re now at a point where we can share more details about what we’ve been doing.
My feelings on GNOME are very double. On the one hand, I love the fact that the GNOME team really seems to have a solid plan of what it wants, and it sticks to this plan to a fault. On the other hand, I just do not like this plan. It doesn’t mesh with what I want from a desktop computing experience. The fact I have to mess around with shaky extensions through a web interface to make GNOME halfway usable to me just isn’t a great user experience.
But that’s fine – this isn’t the Windows or macOS world where we have to take it or leave it. We have tons of other options to choose from (Cinnamon for me), exactly so the developers and users of GNOME can build what they want.
I’ve been a GNOME user for more than 20 years now. I remember when GNOME 2 was released and people complained about it, but then it got better. I remember when GNOME 3 was released and people complained about it, but then it got better. Now GNOME 40 is going to be released and people will complain about it, but then it will get better.
Having said that, what they’ve showed with GNOME 40 is that some of the metaphors and design decisions of GNOME 3 were not right, which means that the people who complained were correct and the decision to keep pursuing those designs was wrong. Remember the whole brouhaha when GNOME 2 when full spatial mode and was opening a new window for every folder opened in Nautilus? They’ve kept that for years before reverting.
The main reason why people loved GNOME 2 so much was that Red Hat and Sun poured a lot of money on usability studies in order to have a good UI and UX, scientifically proven. GNOME 3 brought a lot of good underlying tech, but threw away all of those studies in favour of some new fancy design trends and hunches. That’s why people complained so much and that’s why Canonical had to come up with their own solution, since they couldn’t trust GNOME any longer. GNOME 40 is living proof of that.
I sincerely hope that GNOME 40 reverts that trend and the teams goes back to do usability studies and following up on them, which, thankfully, they seem to be doing. Can’t wait to try it out.
I’ve never understood the love affair with Gnome2. I found it awful and unpleasant to use. I much preferred Xfce in CDE mode during that era.
Once the 3D video drivers stabilized, Gnome3 really came into it’s own, and the UI/UX is more inline with what I want out of a DE. Believe me, I fully expected to hate it, but it grew on me once they refocused on being a keyboard-centric design.
“If I asked people what they wanted, they would have said a faster horse.” — Henry Ford
Doing things on a hunch is how evolution works. Maybe it works, maybe it doesn’t. 50/50 shot.
People complained because it was new, and they had to learn new things. People complain about the UI changes in new versions of Windows, and MacOS having the same basic UI since the mid-80s is the counter.
Canonical wanted to own the code for the entire stack which is why they started Mir and Unity. Wayland and Gnome3 were convenient excuses.
Sure, I’m not disputing that. What I’m saying is that you need to put your hunch to the test, put it in front of users and see how well that change performs. A/B testing and all that.
People also complained because some of the changes were really unfortunate. The requirement of 3D graphics drivers meant that you couldn’t use it on a VM any more. Finding an application on the menu took many clicks and you had to travel with your mouse from top left to bottom left to middle right to center. Madness.
It’s funny that you mention MacOS: people didn’t complain about MacOS X because it was the only one that didn’t switch its paradigm.
Canonical sure loves their NIH syndrome. The good is that because of Mir we got that kick that Wayland needed and because of Unity we got that kick that GNOME needed.
It seems to me that the primary purpose of GNOME40 changes is driven by a desire to introduce and integrate a better virtual desktop model. But, look at the mouse travel required (press Activities menu item at top-left of screen, then travel to bottom of screen to press the Overview (9-dots button) at the bottom right, then travel back up to the virtual desktop that you want. This is a workout and really unnecessary.
A better model is something like this GNOME3 extension (Workspaces to Dock [1]) which requires moving the mouse to extreme right of the screen, the virtual desktops pop out from the right and you mouse over the one you want to click and change virtual desktops. Minimal amount of mouse travel and a decent mental model. I’m obviously not a designer, but GNOME3 is usable for me with this reasonable addition. Unfortunately, the extension author has recently stopped maintaining after several great years.
[1] https://github.com/passingthru67/workspaces-to-dock
Good design is based on good scientific principles which begins to be more an issue of fact than opinion. People don’t like arbitrary change and too much choice can be a bad thing. What happens if you use one desktop at home and another at work? Changing isn’t a seamless one cick exercise. You have to start reading manuals and double check it works with the OS you’re using. The fact that in the real world it’s not as easy or seamless as it could or should be is one major reason which puts me off using Linux. Life is too short to have arguments about management and market research and usability and the costs of technical support issues every time a random committe changes direction.
Well said.
I agree that people don’t like arbitrary change, but that’s no reason to avoid evolution. If it’s done correctly, with usability in mind, there should be no issues. I’ve had many non-technical people praise GNOME 2’s interface and how easy it was to use. Having said that, after a couple of tweaks, even my mom is using GNOME 3 and support calls dropped to almost zero.
We mustn’t forget that these arbitrary changes are not a GNU/Linux only thing. Microsoft Windows and Microsoft Office changed a lot in the last decade, so the argument that changing to a Free desktop because people already know the old proprietary interface is void.
Sodki,
Well, all these discussions need to be prefaced with a disclaimer that aesthetics are extremely subjective, but even then there’s something to be said about WIMP desktops having reached a pinnacle a long time ago when GUIs were being developed through intensive user studies and methodical improvements. Business being what it is, they cannot stop and have to continually justify more and more changes to no end. For a lot of us, this process is not improving the experience
No kidding, the whole metro/flat artistic movement and the elimination of things like titlebars and toolbars were really pervasive and IMHO detrimental to usability. UIs have become so minimalistic and access to features hidden that it’s become very difficult to provide tech support. The solid frames of reference GUIs used to provide are gone. At least with most linux desktops you can tune things back, but with windows they killed a lot of the customization and to be perfectly honest there are times when I struggle to see where one window ends and the other begins because everything is flat and the borders are mostly gone, ugh… I can appreciate the value of art, but I don’t want my desktop to be an art piece, you know?
I get the feeling that it’s less art and more fashion/trends. If you go from a contemporary OS UI to one from the last generation, the difference is noticeable. I think this is the same effect occurring as with cars, phones, sunglasses. New designs are easier to market and sell and people get motivated to move on to the latest and greatest.
Well, all these discussions need to be prefaced with a disclaimer that aesthetics are extremely subjective
Aesthetics yes, usability no. You can set up usability tests and objectively come to a conclusion if a certain design or change is better or worse. This is what was missing when GNOME 3 started, and we have the proof now: they did usability studies for GNOME 40 and based on that they’ve changed the UX.
I agree with all of the above follow on comments. While aesthetics and function are largely seperate areas there is some overlap. It gets a bit hazy when you get into perception exactly which side of the line something can fall. In an ideal world “UX” types who have young buck syndrome and are meddling so they can meddle and marketing types would put their energy into this not concentrate on simply faffing about or peddling more stuff just because.
People deal with it, and it kind of is.
I switch between Gnome3, MacOS, iOS, and iPadOS seemlessly. Then there is Windows and Android which I can still find my way around.
Linux is a lot easier to live with then the other big two. Your lack of willingness to learn new things is the impediment, and “the real world is not easy” is an excuse, which is fine. Don’t blame the tools, and take responsibility for your opinions.
I simply have other priorities and my bandwidth is taken up plus a large amount of been there done that. I make my decsions for my reasons which suit me not yours. Please don’t assume.
Gnome as a project has a history of bullying their users and third-party developers. Use it if you have no problems with that, just don’t whine when they again deprecate the whole desktop and workflow you’ve got used to. This has already a problem in Gnome2 days, albeit at a smaller scale – rejecting minor features, not fixing bugs. The biggest issue was the spatial mode in Nautilus, but they were still sensible enough to backtrack on it before Gnome2 has got popularized by Ubuntu.
To be clear – I don’t mind and indeed don’t care about about Gnome3. What I don’t *like* is that Gnome devs went to great lengths to kill Gnome2, which was a de-facto Linux desktop at the time and has just managed to get its first consumer market deployments. Ten years later Linux desktop is still nowhere near to what it once was.
They could have easily renamed their v3 stuff so that it didn’t conflict with v2 and let other folks step in and maintain the old desktop. Whatever their reason was – arrogance, getting rid of competition, stupidity – they chose not to. Luckily for the *existing* users, Linux is not Windows or MacOS – there are always other choices like Xfce, Cinnamon, KDE or indeed Mate (kudos to them for fixing what Gnome devs have broken).
ndrw,
Yeah, I think they did it this way because they didn’t want to compete with their own successful desktop environment, but I also felt what they did hurt linux. This was taking place at the same time microsoft was shoving unpopular metro changes in windows too. It would have been nice to point to linux as a platform where user interests always come first, yet this was clearly not true with gnome. One may or may not like gnome3 desktop, but they way they handled things leaves a bad image.
I had been using gnome 2, but I rotate every few years. I tried cinnamon but at the time it wasn’t there for me yet. More recently it’s been Xfce and KDE. Xfce is a great no frills option. I’ve encountered a couple bugs but overall its consistency makes it a good choice for me. I’d recommend KDE for classic windows users who are trying linux for the first time since they’ve got a similar look and feel.
No comment on gnome 4 / 40 since I haven’t seen it.