I have to admit I have not been a huge fan of some of the design choices in the past or the way Gnome responded to the community. With that said, the new releases work very well for me and it really does work well for my day to day use. Congratulations Gnome team on making each release progressively better!
If you do not change - some people will complain, if you do change - other people will complain. Some people will always complain. This is open source product and if enough designers/developers will feel it is going into wrong direction, I am pretty sure it will be forked. And for people who like gnome2 experience - there is mate desktop. So it looks like most cases are covered.
I don't think that's meant as a justification for anything, just a statement of fact. Open source projects are heavily dependent on electronic communication (mailing lists, forums, etc.) for everything from feature planning to user feedback, but what all of those mediums have in common is that people are 100x more likely to post that they hate than that they like something. So, it really is true that no matter what you do or don't do, seemingly everyone on the Internet is going to hate it, when in fact that's just a very vocal minority.
Point being, even GNOME 3 (or Unity...) levels of vocal online user hate don't necessarily mean that a meaningful percentage of the user base actually dislikes the changes. Feedback should always be read and reflected on, just not followed slavishly. If you just try to avoid Internet hate from users, you'll never change anything, and even that philosophy will garner you a continuous stream of heated forum posts about how nothing ever changes.
It's kind of like reading Yelp reviews. In the aggregate, they help some places stand out from the crowd, but they're still not any sort of objective judgement on the place being reviewed. You always have to think about where the reviewer is coming from and take that into account. For example, dumbasses who go to some incredibly popular and trendy restaraunt at 7:30pm on Friday without reservations, and then post negative reviews because they had to wait too long for a table. Well, duh. In the software world, the equivalent would be complaints about a piece of software not being just like X (where X could be a previous version, or some competitor), when in fact your entire goal is to create something unlike X. If someone wants X and you're making Y, then have to take "Y sucks because it's not X!" complaints with a massive grain of salt. It still doesn't mean you ignore those complaints (maybe Y was a bad idea?), but you wouldn't be doing anyone any favors by doing the most common Open Source thing, which would be to either say, "oh well, I guess we'll stop making Y and just make another X instead", or (even worse, and sadly even more common), "I know, we'll make everyone happy by making something that's both an X and a Y! We can just put a checkbox in the settings!"
I'm saying that when your arguments can be used to justify either imposing sharia law, pulling your child out of public school, or changing the body design of the next years 200cc Vespa - it's really just an argument for doing what you want when you want. Why the need to rationalize it? Why the need to insult the people who do object as anomalies of human behavior, an emergent result of complex constrained random interactions, or Haters?
People who want to do, and can do, don't have to justify anything.
If what you want is to do the right thing (rather than a more concrete goal) then instead of (pseudo)accepting objections to your decisions as inevitable and to be ignored, why not address them individually, and on their merits? Let's pretend that the world is complex, and there's something left for all of us to discover about human behavior and what's easier, harder, and more productive for people.
The GNOME Devs did the right thing though. They are open source so the community can fork anything they want and there are plenty of other really similar DEs. Ubuntu and GNOME can't compete with modern UIs without making something that's multi-platform oriented. I believe GNOME 3 and Unity actually have the right ideas when it comes to convergent interfaces, whereas Microsoft just basically gives you two UIs. Traditional interfaces are great for traditional computers, but to compete with the big dogs, which is, I think, what Canonical and GNOME (to a lesser extent) want to do, they need to create interfaces which can work on anything.
What will happen to GNOME if it does not successfully compete with the big dogs? Nothing. It does not compete with the big dogs now, has never done so, and - whatever convergent multiplatforming it may do in the future - is unlikely ever to do so. And yet it is still a viable project.
This isn't the beverage or toilet paper industry. Niche products - especially ones like GNOME that don't have to "sell" or turn a profit - have a demonstrated way of living on.
There may be good arguments for what GNOME has been doing, but the need to compete (with largely imaginary competitors imho) is not one of them.
>the need to compete (with largely imaginary competitors imho) is not one of them.
But it is though. If this is your philosophy I would love to see some of your crappy projects. Believe it or not, when you design an interface you're competing with the other people who are doing the same thing.
I'm sorry you dislike GNOME, but your comment adds absolutely nothing to the discussion.
If I was going to make a search engine I would make it with google in mind. If I was going to make a social network I would make it with facebook in mind. I don't care what twisted way you want to frame your argument, it's incorrect. GNOME 3 and Unity do compete with Microsoft whether you their competition is a real threat or not.
Do you not think Apple competes with Microsoft? Do you not think internet explorer competes with chrome?
> Believe it or not, when you design an interface you're competing with the other people who are doing the same thing.
Sure, and when you're designing a $100k sports car you're competing with the people who are designing a $20k utility truck.
> If I was going to make a search engine I would make it with google in mind. If I was going to make a social network I would make it with facebook in mind.
You'd probably make a shitty search engine/social network then. There is no reasonable way you could beat Facebook/Google at their own game by "having them in mind" as your main design rationale.
There are millions of devices sold with Apple/MS software on them EVERY DAY. The goals these products have in mind (for Apple, for example, that includes converting users to stay in their ecosystem, use iTunes, iMessage, iCloud, etc.) have nothing to do with the problems KDE and Gnome aim to solve.
Unless you're saying that the goal of Gnome/KDE is to displace that? Because if so, for the past 10 years, we have been FURTHER away from that goal every single year.
There's a reason why LXDE and friends have been gaining in popularity. There is a demand for lightweight, straightforward environments that just work, even on super old hardware. Gnome and KDE are just relics of a very, very different past, and at this point their death is inexorable.
The one thing I dislike is that those of us who were using GNOME 2 lost our desktop environment for a long time. There should have been an attempt to keep the old DE going. I'm now using Mate on Linux Mint and am very happy.
It's important to understand that this isn't just "Gnome 2," but actually full Gnome 3 underneath with all the benefits that contains. That's why just continuing with Gnome 2 was a non-starter.
> That's why just continuing with Gnome 2 was a non-starter.
There wasn't enough time to maintain the existing Gnome 2, but there was enough time to create an entirely new desktop environment? I don't think that's possible.
The reason Gnome 2 was a non-starter was because developing a new DE is a lot of fun, and maintaining an old DE is not.
It's their choice, of course, but let's not pretend that abandoning Gnome 2 was anything other than a desire to work on something new.
Gnome 2.32 was released in September 2010. Gnome 3.8 was released in March 2013. That's 3 years that a decent Gnome 2 replacement didn't exist. You don't have to drop 2.x support during development on 3.x.
One thing that still bothers me about Adwaita is the huge amounts of blank/white space around some widgets like list items and enormous padding on buttons. I wish for a way to easily customize this.
It also quite undocumented so you spend a day or two googling around and breaks on random minor updates when the package manager overwrites your changes again. Sounds fun.
They will probably never document it because they don't think anybody should change it. On the other hand, I can't stand invisible borders and I always need to increase the border grab range. So just a good, highly-googleranked blogpost would be nice. I think the "correct" way is to copy the theme from /usr/share/themes to ~/.themes and make your changes there, then it won't break if the package manager updates them.
Evince's new layout is pretty bad. They removed the ability to see a document in fullscreen without any toolbar bothering you, which is probably one of the most important things in a PDF reader.
I really wish it was possible to tell GTK to respect the window manager's bar styling with client side decorations- I get pretty grumpy about the ui in a non gnome context.
Oddly, I hear this may work on Wayland, but not on X11...
I'm extremely excited for the Wayland support in 3.14. Being able to drag a window without it and the cursor getting out of sync really makes the interface feel that much better. It's equivalent to the difference in smoothness between Android and iOS.
> Being able to drag a window without it and the cursor getting out of sync really makes the interface feel that much better.
Could you please rephrase/clarify?
I have a similar issue on X11 since Gnome 3.6 or 3.8 (can't remember exactly) where when dragging windows, the window is moving a bit behind on where it should be relative to the cursor.
Feels smooth rather than crisp as it was previously, but it also feels like my CPU is playing catchup although it isn't.
> when dragging windows, the window is moving a bit behind on where it should be relative to the cursor
I meant precisely that. I believe it has to do with difficulties synchronizing the X Server with the compositor. Wayland eliminates this by making the compositor the server.
I remember this never used to happen with Gnome 3.4 and earlier.
I think it started happening when they announced this frame-perfect timing/rendering system. Solved a tearing problem but introduced a lag problem :) I guess you're right though, the lag problem should be solved with Wayland.
How do people use GNOME, KDE or Unity for more than a couple of months and not get extremely annoyed at the constant pointless changes? I just don't understand it.
I get pissed off everytime Firefox updates and things break, aren't in the same place, have to be reexamined, disabled, enabled, whatever. Using ever-changing sites like youtube is just a nightmare.
I've switched pretty much all of my workflow on Linux to terminal software. I'm confident that programs like mutt and emacs will be essentially identical 10 years from now. Struggling with/rediscovering interfaces all the time just isn't on my list of things to do. I would prefer if developers invested some thought into it and got it mostly right the first time.
38 comments
[ 0.18 ms ] story [ 72.3 ms ] threadhttps://fosdem.org/2014/schedule/event/gnome3_outreach/
It clarifies a lot of thing and is a bit of a post-mortem of all major switches that Gnome has ever done from a community perspective.
This is a thought-terminating justification for anything.
Point being, even GNOME 3 (or Unity...) levels of vocal online user hate don't necessarily mean that a meaningful percentage of the user base actually dislikes the changes. Feedback should always be read and reflected on, just not followed slavishly. If you just try to avoid Internet hate from users, you'll never change anything, and even that philosophy will garner you a continuous stream of heated forum posts about how nothing ever changes.
It's kind of like reading Yelp reviews. In the aggregate, they help some places stand out from the crowd, but they're still not any sort of objective judgement on the place being reviewed. You always have to think about where the reviewer is coming from and take that into account. For example, dumbasses who go to some incredibly popular and trendy restaraunt at 7:30pm on Friday without reservations, and then post negative reviews because they had to wait too long for a table. Well, duh. In the software world, the equivalent would be complaints about a piece of software not being just like X (where X could be a previous version, or some competitor), when in fact your entire goal is to create something unlike X. If someone wants X and you're making Y, then have to take "Y sucks because it's not X!" complaints with a massive grain of salt. It still doesn't mean you ignore those complaints (maybe Y was a bad idea?), but you wouldn't be doing anyone any favors by doing the most common Open Source thing, which would be to either say, "oh well, I guess we'll stop making Y and just make another X instead", or (even worse, and sadly even more common), "I know, we'll make everyone happy by making something that's both an X and a Y! We can just put a checkbox in the settings!"
People who want to do, and can do, don't have to justify anything.
If what you want is to do the right thing (rather than a more concrete goal) then instead of (pseudo)accepting objections to your decisions as inevitable and to be ignored, why not address them individually, and on their merits? Let's pretend that the world is complex, and there's something left for all of us to discover about human behavior and what's easier, harder, and more productive for people.
This isn't the beverage or toilet paper industry. Niche products - especially ones like GNOME that don't have to "sell" or turn a profit - have a demonstrated way of living on.
There may be good arguments for what GNOME has been doing, but the need to compete (with largely imaginary competitors imho) is not one of them.
But it is though. If this is your philosophy I would love to see some of your crappy projects. Believe it or not, when you design an interface you're competing with the other people who are doing the same thing.
I'm sorry you dislike GNOME, but your comment adds absolutely nothing to the discussion.
If I was going to make a search engine I would make it with google in mind. If I was going to make a social network I would make it with facebook in mind. I don't care what twisted way you want to frame your argument, it's incorrect. GNOME 3 and Unity do compete with Microsoft whether you their competition is a real threat or not.
Do you not think Apple competes with Microsoft? Do you not think internet explorer competes with chrome?
Sure, and when you're designing a $100k sports car you're competing with the people who are designing a $20k utility truck.
> If I was going to make a search engine I would make it with google in mind. If I was going to make a social network I would make it with facebook in mind.
You'd probably make a shitty search engine/social network then. There is no reasonable way you could beat Facebook/Google at their own game by "having them in mind" as your main design rationale.
There are millions of devices sold with Apple/MS software on them EVERY DAY. The goals these products have in mind (for Apple, for example, that includes converting users to stay in their ecosystem, use iTunes, iMessage, iCloud, etc.) have nothing to do with the problems KDE and Gnome aim to solve. Unless you're saying that the goal of Gnome/KDE is to displace that? Because if so, for the past 10 years, we have been FURTHER away from that goal every single year.
There's a reason why LXDE and friends have been gaining in popularity. There is a demand for lightweight, straightforward environments that just work, even on super old hardware. Gnome and KDE are just relics of a very, very different past, and at this point their death is inexorable.
Source: I am a former Gnome contributor.
Gnome 3 is, on the other hand, a very good alternative for those that hated Gnome 2, which is the crowd I fall in.
It's important to understand that this isn't just "Gnome 2," but actually full Gnome 3 underneath with all the benefits that contains. That's why just continuing with Gnome 2 was a non-starter.
There wasn't enough time to maintain the existing Gnome 2, but there was enough time to create an entirely new desktop environment? I don't think that's possible.
The reason Gnome 2 was a non-starter was because developing a new DE is a lot of fun, and maintaining an old DE is not.
It's their choice, of course, but let's not pretend that abandoning Gnome 2 was anything other than a desire to work on something new.
-Taskbar cannot be resized (!)
-Systray widgets cannot be added, moved or removed
-The entire concept of a "panel" is gone.
Gnome 2 was extremely flexible. Gnome 3's "Classic Mode" is more like Microsoft Windows, although even Windows lets you resize the taskbar!
https://download.gnome.org/core/3.13/3.13.3/NEWS
https://download.gnome.org/apps/3.13/3.13.3/NEWS
Oddly, I hear this may work on Wayland, but not on X11...
https://extensions.gnome.org/extension/584/taskbar/
Could you please rephrase/clarify?
I have a similar issue on X11 since Gnome 3.6 or 3.8 (can't remember exactly) where when dragging windows, the window is moving a bit behind on where it should be relative to the cursor.
Feels smooth rather than crisp as it was previously, but it also feels like my CPU is playing catchup although it isn't.
I hope your comment was sarcasm! :)
I meant precisely that. I believe it has to do with difficulties synchronizing the X Server with the compositor. Wayland eliminates this by making the compositor the server.
Nothing in my comment was sarcasm...
I think it started happening when they announced this frame-perfect timing/rendering system. Solved a tearing problem but introduced a lag problem :) I guess you're right though, the lag problem should be solved with Wayland.
I get pissed off everytime Firefox updates and things break, aren't in the same place, have to be reexamined, disabled, enabled, whatever. Using ever-changing sites like youtube is just a nightmare.
I've switched pretty much all of my workflow on Linux to terminal software. I'm confident that programs like mutt and emacs will be essentially identical 10 years from now. Struggling with/rediscovering interfaces all the time just isn't on my list of things to do. I would prefer if developers invested some thought into it and got it mostly right the first time.