Slashdot Mirror
Designer Jon McCann: "More Optimistic About GNOME Than In a Long Time"
An anonymous reader writes "In an extensive interview with derStandard.at, GNOME designer Jon McCann shares his thoughts about all the criticism GNOME 3 currently faces and why he doesn't think at all that GNOME is in a crisis. He also talks about the current plans for GNOME OS and explains why he thinks that Linux distributions should rethink their purpose."
I hate to be the bearer of bad news, but Gnome 3 has issues, and the criticisim is legitimate. Why does it always have to look like that? At this point Windows 8 looks easier to customize than Gnome 3.
yeah I kind of thought the same....we have a critisism that says "the gnome leadership doesn't listen to it's users" and it's users saying "wtf, I can't select the font size???", etc, etc, etc.
then you have an article by one of the gnome team says he's "super confident" about the project.
doesn't that kind of explain everything, in perfect clarity.....and prove the point beyond doubt that the gnome leadership don't in fact listen....
Gnome 3 is a complete mess. and it's UI is not easier to use or more intuitive, its just trendy and "different" It is 5 years behind Gnome 2.x in usability and polish. A lot of the criticism for Gnome 3 is justified. The problem is knowing how the Gnome team works, they will ignore everyone and do what they want.
I have tried several times to use it and every time the same parts fall down. Luckily some smart people are picking up the abandoned 2.x line and forked it. So linux will continue to have a useable desktop instead of the wierd social experiments that are Gnome 3 and Unity.
Do not look at laser with remaining good eye.
Well OK here are the main points from the interview
1) He wants to move towards distributing Gnome more directly in particular to Windows.
2) He likes the fact that Gnome has clear direction. In his mind the crisis was when Gnome 2 started wrapping up and the Gnome developers didn't know what to do. While for the developers Gnome 3 has been full of direction
3) The Canonical divorce is continuing and Ubuntu will not be the testing platform going forward. Gnome OS is coming somewhat out of the desire to have a stable place to test Gnome.
4) He really believes the diversity of the open source eco system makes it impossible to support software.
5) He believes that the Gnome community is responding to the criticism they can extract, i.e. the constructive criticism. For example changes to the UI file movement and getting rid of the "copy and paste" applied to files.
The GNOME people aren't listening. They thought when they had something people were happy and comfortable with that they "lacked direction" so they got together and decided on a direction. So once they got to a point where people were happy and comfortable, they somehow thought it meant it was time for change.
I think this is where the problem has begun. In my mind there are few acceptable reasons for change:
1. A brick wall has been hit while going one direction and the previous goal is not achievable
2. A crisis of compatibility or usability has occurred where the current way of doing things is no longer acceptable, applicable or useful
3. People are moving away from GNOME because something better has their attention
4. People are moving away from GNOME because the development team isn't responding to them
There may be more, but those are just the first few that occur to me. Of those only #4 is applicable and that is only because they decided to change and not listen to the people using GNOME. They caused #4 and persist in it.
GNOME developers are completely out of touch. They created change for the sake of change and that is a very bad reason for change when people are depending on keeping things as they are.
The article/interview parallels what GNOME has done with Mac OSX and Windows. Mac OSX changes were... not completely necessary but also not completely alienating to the user from an interface standpoint. Microsoft's changes are perfect examples of end user rejection and how the users affect the marketplace. Shame on you, GNOME team, for not noticing this. No one has accepted Vista. Windows 7 has been accepted because there is no more Windows XP. And Windows 8? Developers are shying away from developing for it. Microsoft at least acknowledges that it is screwing up and has reversed some of the things that have offended developers with regard to Windows 8. We see none of that from GNOME... yet...
Or hellz, XP.
Make it look and work as close as possible, out of the box. No dicking around, no "Yehbut, we can improve it just a little bit here, maybe a dab there, a sprinkling over that wa- ah, we'll fix that in the next version".
My wife will use it. My mother will use it. My employer might even take a look at it.
Stop with your new paradigm fantasies. The desktop isn't broke (until Windows 8). Quit trying to fix it.
If you were blocking sigs, you wouldn't have to read this.
For years, using Linux was something for the advanced users who enjoyed having to dig deep on occasion to make it all work. Then along came Ubuntu and turned the Linux desktop into something that was REALLY a lot easier to install, use and maintain then Windows and yes, even Mac. Can either of those two run from a CD with full access to the HD if something goes tits up? Do either of them have a full desktop available with browser and everything else during install?
Finally, Linux the desktop. WORKED.
And then the Gnome 3 team said "nah, it doesn't, we know better how you should work". And they released a badly tested, badly thought out and badly documented product way to early and with no training to get people used to the new interface or any motivation for wanting to get used to the new interface.
It is like me forcing you to sit upside down on the toilet, with no training or handy handholds all for the pleasure of crapping on your face. Whatever secret benefit it might give, you are not going to be in the mood to find out right?
It is the same with Windows Metro. WHAT IS IT SUPPOSED TO FIX? What was missing in the classic desktop user design that is being fixed in by either Unity, Gnome 3 or Metro?
People are perfectly willing to change for a well known UI if there is a really good reason to do so. Who here still uses rotary dialing on their phone? Touch keys WERE a massive improvement, not just more accurate but also less stressful on your finger if had to dial a lot. The mobile phones and indeed the rise of OSX has shown that people are not stuck to the classic desktop, as MS thought judging by their early attempts at a phone OS.
But for the desktop, the desktop design, just works well enough. Gnome 3 made its introduction even worse by not being very well put together and doing it while things like Nautilus were still horrid pieces of buggy crash prone slow as molasses software. They then threw out all the good bits all the improvements others had made to make Gnome 2.0 workable and made something nobody wanted instead.
But all is good. Hello? You have been ditched left and right by distro's. Mint rose as a distro from nothing just because they offered people non-gnome3 despite their insistence of screwing up google searches.
The Iraqi minister of looking silly couldn't do a better job of dis-information. Gnome 3.0 has not been ditched by all users. Gnome 2.0 fork is NOT eating our lunch.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
The KDE team really fucked up when they went from 3 to 4, sor your innovation for the Gnome project is the best thing that's happened to KDE in a long time.
If someone is passing you on the right, you are an asshole for driving in the wrong lane.
Either Gnome 3 developers are delusional, or being paid by Apple to screw the open source desktops on purpuse. How, otherwise, did Gnome and Ubuntu fall from the top, while on the peak of success?
Also, I can understand Ubuntu because the leader drops a lot of $$ on it, but Gnome? I would have thought Gnome was a community project influenced by the community, but if delusional people (and I mean delusional because they state they target laptops, yet make an OS for tablets) is running the project, something must have gone wrong somewhere.
Subject says all XD
Rehdon
I agree w/ this. Guys like the Libre-Linux crowd and OBSD use GNOME 3 in the fallback mode. Most of the BSDs are yet to move there - they may just prefer KDE or other DEs. Debian has abandoned GNOME for LXDE, and Mint for Cinnamon. In the meantime, on the Qt side, for those who think KDE has too many bells & whistles or is too much of a resource hog, there is Razor-qt as well.
GNOME's purpose for existing was gone once
I prefer the Qt side's approach to this - KDE for a fully functional UX w/ different target implementations for laptops, tablets and phones on one hand, and Razor-qt for a lightweight version of the same thing. Particularly good is the projects' attempt to create various applications that can be used as building blocks, like the parts of Calligra Suite, and so on
Essentially we were designing GNOME2 to be the free software version of a desktop computer system.
Our main target for GNOME3 is laptop use, which I think is by far the overwhelming majority of computing use today - in the non-mobile space.
I look at things a bit differently. I walk around the conference and I'm absolutely amazed by the energy we are seeing in the GNOME community right now. I am more optimistic about GNOME than I've been in a long time.
We have fewer people testing GNOME outside of the active contributors. And there are a number of reasons for that, but that's also why we have these discussions around making GNOME more easily testable.
So to paraphrase: We changed everything, new paradigm baby. Us developers love it, but it turns out the users just don't like it anymore and we lost all our testers. So now we feel we need to make it easier for us to test GNOME since we have to do it ourselves.
"but money is the God of Algiers & Mahomet their prophet." - Rich. O'Bryen June 8th 1786
People should try it before they ignore it. It is not ill-suited to everyone. I think its a case of the gnome team thinking everyone works like they do. I use keyboard controls almost exclusively, with lots of windows open, mostly command lines. I start applications from a run box, or commandline, not menus. OSX came along and the spotlight/quicksilver method of starting apps was a big step forward, it would autocomplete the name of the application for me. Gnome 3 and unity are another step forward in that it will give me a nearest match if I mispell something, I can type either the visible name (like "files") or the application name "nautilus" and either works. Or natulius for that matter. Additionally its a single key press to start typing rather than two as in windows7 or osx. workspace key shortcuts haven't changed from gnome2 and the window tiling is sufficient, though usually I don't dock windows. I prefer gnome3 to unity for the shrinky window thing it does showing whats open (like osx).
If you mostly start apps from the commandline instead of menus or quickbars, gnome3 is for you so give it a try. They should have realized though that not everyone works that way and made it more flexible.
refactor the law, its bloated, confusing and unmaintainable.
This is one thing the ReactOS guys should do. Have different 'themes' from various Windows versions, all of which can be used for the OS. Windows 7, Vista, XP, 2000, NT 4.0, maybe even NT 3.5. Let the users select which one they want, and enable that during installation. Or even from the display panel.
My definition of operating systems is basically two things: It's a well defined user experience and it's a well defined developer experience. So those are the two interfaces we need to provide and everything else is an implementation detail - down to the hardware.
I don't know if you saw Lennart's talk the other day, he said that we should let the user experience and the developer experience drive everything in the lower level stacks.
When I saw this, I thought: Wow, someone who *gets* it! The OS provides the common ground for app developers and their users. Not only that, he is willing to even make that dev/user distinction, which I think is a crucial position that many FOSS projects try to shirk.
But he doesn't seem to get what those positions imply: In particular, that you need a single organization driving its vision down into the lower stacks, achieving
sufficient vertical integration to make the user and developer experiences sane and interesting. In order to really get somewhere with this idea, they will need to fork their entire stack and take charge of it the way Google has with Android.
He also doesn't make a distinction between system developer and app developer. IMO, this is also a common and necessary distinction that FOSS system devs tend to shirk. Blurring these distinctions does no one any real service, because it robs you of opportunities to keep asking the question, "Does X feature or Y implementation technique hinder or facilitate a user who wants to try programming... who might become an app developer for this platform someday?" Counting on your audio stack devs to design the next 'Garage Band' is kind of pointless -- it rarely works. You need to entice lots of people who would love programming but will never give a rats a** about tweaking system code.
What the lack of distinctions also does is create an attitude of indifference about system oddities and shortcomings... The old "If you don't like it fix it yourself" cop out (which I think Gnome has more of than any other DE). You attract almost no one that way.
All in all, the conceptual step displayed in this article may be good, but I think too little too late. He's discovering only part of what Jobs and Gates knew in the 1980s.
Re #4) I think he has a point. I used to support commercial software on Linux desktops and its much more difficult to get anywhere with the user unless they are the type of user that you can just tell them to drop to CLI. Even so, it was a GUI app and that is where Linux distros are the most chaotic.
Re #5) In the 8+ years I've been giving constructive criticism in forums like this and in bugtrackers, I've never seen a DE project implement a suggestion I've made. Mozilla and other projects have, including some proprietary products, but the DE projects IMHO are among the worst of them. They start by repudiating the defacto definition of an OS as including a GUI (to them the OS is a text environment that only provides programming interfaces, and their project is just a bit of high-calorie icing on top). From there it goes downhill.
From the article:
So, basically, they are targeting laptops (which are losing ground to tablets) and expensive machines, ignoring the truly vast numbers of cheap desktop PCs that exist in nearly every home at this point. For the inevitable automotive analogy, it's like they're making car paint that only looks good on Priuses and Teslas.
If they are de-optimizing for over half the installed hardware base it's unsurprising that they aren't satisfying end users.
Linux distributions should rethink their purpose.
Fuck you and fuck your arrogant decree that Linux distributions needs to match what you believe them to be. I'm going to make my Linux install exactly what I want. That's half the point of using an open source OS. And unsurprisingly it does not include Gnome 3 (other than a fork like Cinnamon) because its developers and "visionaries" don't give a shit about me, so I don't give a shit about them.
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs
The only thing that kept me on linux for the desktop as long as I stayed was how great the gnome 3 interface is... I personally think ubuntu's unity is an abhorrent, ugly, mess, and KDE is....well...kde. I feel like gnome 3 finally hit the nail on the head for what users want and need out of a desktop linux experience. When i first started using it I was flabbergasted at the bad feedback people had given it, because I thought it was fantastic, and still do.
I think the biggest problem is that users of desktop linux are overwhelmingly flat-out unwilling to accept any sort of change to their status-quo. Gnome 3 really is a better interface, and people need to stop hating and fear mongering about it.
That missing MS Windows "feature" of MS Windows of old, DLL hell, has finally come to linux with Gnome 3 where utterly stupid choices of names for components means it's gnome3 and no earlier if you want to use even a single recent gnome application. Google for attempting to run gimp2.8 on centos if you want more detail.
They already cloned Windows 7. It's called KDE4.
Seriously, though, I used KDE and loved it until 4 came out. It was unstable, bloated, and difficult to use. I switched to Gnome2 and loved it. Then a dist upgrade put Gnome 3 on my laptop and suddenly my laptop was less usable than with KDE4. I switched to Fedora's KDE spin, more so because of a need to run some commercial software that was much better supported under Red Hat RPM-based distros. I realized that in the time I was away, KDE4 slimmed down somewhat and was much more stable. More importantly, I was able to fully customize my eye-candy unlike Gnome or even Windows 7.
I still think the KDE group's philosophy of making KDE 4.1-4.4 a prolonged, nasty beta period was a terribly poor idea and I hope they seriously reconsider when looking forward to KDE5. However, now that the beta period is over and we now have KDE4 release candidates (4.6-4.8) I think it's worth a second look for a lot of *nix users.
GNOME designer Jon McCann shares his thoughts about all the criticism GNOME 3 currently faces and why he doesn't think at all.
There, FTTY.
systemd is Roko's Basilisk.