I was wrong about GNOME 3, it might become one of the strongest DE contender than it ever was.

I was wrong. I won't apologize of course, we all know that GNOME 3 didn't exactly had an spectacular release. And I still hate the official tablet-like look--mind you, I still use GNOME Classic.

So what was I so wrong about? Why doesn't GNOME 3 die and burn in hell for eternity?

Well, if you have to ask. They did a lot of things right. I would say they provided an even more intuitive interface than KDE ever did.

How important is unification?

If you are some sort of neckbeard or wannabe zealot then chances are you are enjoying your Linux desktop in whatever window manager of your choosing while simultaneously assaulting other desktops environments and depriving them of their choices.

After using GNOME 3.12, something came to mind. They were doing a lot of things right. I think screenshots will do a better talking than me.

User Management is a bliss.

Evolution unifies really well with my google account

No brainer date and time interface

One of my favorites! I wish KDE had something like this

Automatic printer sharing detection. I was using Samba of course.

It's so easy it hurts.

Beautiful font rendering

And this wins the prize

I don't think many of you know but I was a GNOME 2 avid user. I loved it to death, and then KDE 4 came with all its stupid wobbly windows, but you know, it worked. It was usable, sure it had its ups and down but it didn't turn out so bad. 

GNOME 3 did a daring move though, one that could have ended the project itself. Yet, they started listening to feedback. I think that in a way I blame their user interface designers because they could have introduced more HUMANE interfaces than what they provided in the initial release. It felt like they forgot their old userbase existed and that's bad. The thing that "we don't know what we want" is a bullshit line, imho. Even if the end-user didn't know what she/he wants, you shouldn't have gone full retard. 

Of course, GNU/Linux is all about choice. Sadly, if we are going to CONQUER the desktop we need to choose a desktop environment to rule them all. If so, which would be your desktop environment of choice that the end-user would feel usable? Let's play a game. Choose one that needs the least of terminal interaction.

The Year of Linux Desktop. Is it coming anytime soon? I don't think so.

Before you start raging about how wrong I am, please listen to what I have to say then draw a conclusion.

I don't think Linux is ready to be a desktop. It's too fragmented, which is one of the issues for me. To me, fragmentation is only part of the problem. Right now we have too many flavors; it's hard for a developer to deliver a commercial application and is expected to support all types of distributions in which all of them has different types of libraries versions. It's like someone developing an application in a bleeding edge environment, the person uses the latest libraries and then find out that people using stable distributions such as Debian 7, CentOS, RHEL, etc can't because they are using a shiny libc6 2.19.

Yet. That's not the problem, not at all.

It's hard to make a person who doesn't use GNU/Linux to be comfortable. You can all be screaming about how open and great it is, how you feel great about leaving Windows, how you feel great using the terminal.

And that's the problem. For example, I don't need my father to know what is a package manager or what is apt-get or yum or how to configure NVIDIA drivers so he can play whatever he wants.

Why should an end-user worry about the terminal?
Why should an end-user worry about what is an stable/rolling/semi-rolling release?
Why should an end-user worry about package managers?
Why should an end-user worry about dependency hell?
Why should an end-user worry about about the latest libraries?

Everything I write here are opinions. They can be dumb, enraging, stupid, serious, fun, etc. I like writing my blog posts, even though I know people aren't reading at all. 

My point is that a "normal" person shouldn't give a fuck about any of those. You can argue that it will help him understand linux better. Have you ever seen someone browsing through Microsoft Windows registry to see how it works? Have you ever seen someone trying to figure out how everything is structured in C:/Windows to learn how Microsoft team organizes their shit?

No. A normal user shouldn't even use normal commands like "cd" or "ls" or "grep" or whatever.

Just a note: I'm not saying that a person CANNOT learn it because it's complicated. I'm saying that they shouldn't do it because his/her perspective will say that in Windows he doesn't do anything like that.

I think that GNOME 2 did a great job. I think that KDE4 does a great job. I think XFCE doesn't do a great job, it feels way too conservative.

I feel like the problem right now is that whoever is designing all these stuff--which by the way, thank you for your hard work--is just concentrating in the pretty visuals instead of having a balance between usable and functional.

The biggest challenge to those working on full fledged desktop environments is how can we prevent the user from opening the terminal? it's a challenge alright, an impossible one I might add. Why? Well, what's the first thing an end-user is going to find out when he/she starts troubleshooting on why his/her wireless is not working on the chosen distribution? A terminal will be involved. There's no "device manager" interface (if I'm not mistaken there used to be one).

It's not the same telling someone to go to Device Manager in Windows than telling someone "go open the terminal, type lspci, type modprobe whatever, type lshw, type apt-get whatever". Chances are that the person will go insane and come back to Windows because it's too much of a hassle.

I'm going to make an assumption and say many of you think it's okay for a person to put efforts in reading documentation and learning about shells and how they work. I don't think you are right.

If anything, a desktop environment should, in my opinion, be intuitive, straightforward. Sometimes we don't have that.

I would genuinely feel sorry if they are thrown to use distributions like Arch Linux and be forced to use xmonad, i3, openbox, etc because that's just not how it should be. Everything is about choice, yes, but how solid and stable is that choice?

There's been something on my mind for a while but giving support to Linux users is a bit complicated, especially if they have never been exposed to the terminal. Now I'm going to change the subject:

I want to advocate to the many GNU/Linux users out there to use stable releases. Let's take a note: Windows XP and Windows 7 have been hell successful. They have been receiving security updates and bug fixes most of the time. What they don't do however is upgrade the latest software. 

That's where it gets tricky. If I want a stable release then I will have to sacrifice updates on software, and I can't install new versions of that software unless I compile it myself, or download the binaries, or use a backports repo. In exchange, I get security updates.

But now I mentioned two things the end-user shouldn't know about: compilation and backports.

Let me know your thoughts. What you think of Linux as a Desktop? Will it ever make it?