GNOME 3.0 Laptop Change Frustrates Some Users

[deanjo]

So far only Mac OS X gets it right: On lid close, suspend to RAM and as backup in case of battery failure also suspend to disk.

OS X also allows you to use the system with the lid closed by design. It is called clamshell mode.


http://support.apple.com/kb/ht3131

[Nevertime]

Maybe its because I don't use gnome but I like the idea of major linux gui forcusing on user friendlyness for inexpirianced computer users or people who simply don't want to ever have to ever bother configuring options.

That said I don't totaly buy into this notion that having options has to damage user friendlyness. Just stick it under an advanced tab if its that confusing.

The other issue is as you strip options away, people have to exercute more complex methods when they do want to make changes.

Macs are only any good if you want to limit yourself to interacting with your computer rigidly via the steve jobs method. The design of the gui's are nice and well thought out as long as your prepared to adapt yourself to how they think a gui should be. If you want to make the gui work for you its an issue. Steve jobs owns your mac and its his desktop you have to adapting to.

If gnomes trying to be the mac gui of linux then I'm sure it'll have its place. But its unlikley to drag me off kde.

[squirrl]

If the distributions don't listen what makes you think developers will?

I found it hard to believe Patrick Volkerding used Slackware 13 or 13.1 with KDE4 and KOffice which lacked a spell checker. Just an example of ignoring users.

vi/aspell I know.

Kubuntu should have stayed with KDE 3.5.10. Mark Shuttleworth can afford to hire a developer to maintain it. Another example of ignoring your users.

SUSE obviously listened for a while, giving a user the option of choice. But ultimately caved in by dropping 3.5.10 to some build system; unsupported.

The point here is enjoy 2.6 gnome while you can.


The distributions sank atlantis, too userfriendly

[Ulukai]

Well it would be dconf-editor for Gnome 3.

As far as everybody else goes... your all nuts. Closing the lid to suspend is the way it should be and it's impossible to get right otherwise without confusing the average end user. Whether or not there is a check mark to enable or disable the feature is immaterial... there remains a lot of heuristics left to make sure that the system behaves correctly when the laptop lid is closed and it's not something that anybody has gotten correct so far.

Without the option working correctly and being fully supported then there is no point to having that option. It just confuses the issue.

99% of the reason to NOT have your laptop suspend automatically has to do with crappy graphics drivers and the general shittiness of X. So instead of forcing the user to work around shitty drivers in Linux... the drivers need to be fixed.

That is the whole point of this. Fix the fucking graphics drivers.

For too long Linux has depended on users screwing around with their systems trying this or that configuration or different software options to work around broken crap like X crashing on suspend. That has to end _NOW_ if Linux has any hope of being a non-Geek desktop OS.

It has far less to do with making it easy to use for morons and much more to do with not shipping a OS that is broken by default. Nobody wants to put up with this crap anymore.

You can't call 90% of the mass thinking the same nuts, it's the other 10% who's nuts. Gnome devs should make their idea default, but at least leave the option for users to change it if they want to. After all, they are developping it for the end-users so their wishes should be their primary focus. As demonstrated, there are plenty of reasons to blank the screen when closing the lid. I do it all the time with Gnome 2 and _NEVER_ had any problems with overheating, display etc.

It's funny, I would rather consider an OS that doesn't resume after suspend to be broken... There are much more issues with resuming than with blanking screens.

[bitu-derr]

I wonder which it would actually make more sense to fork -- Gnome 2, or Unity itself?

Forking Unity would reflect changes like dbus vs bonobo, and perhaps support for panel "applets" could be restored to Unity, the launcher dock could be more customizable (up to and including making it look once again like a 24px wide "panel" that could be on any side of the screen).

One question I would have here would be who would own any fork of Unity, thinking in terms of the Canonical contributor agreement.

[bitu-derr]

FYI, tried posting this earlier, but for anyone who is interested and likes the Gnome 2 experience, there is at least one fork of Gnome 2 now, it is called EXDE, I am not sure how active it is. It seems to have only been started this past January.

The website has a FAQ, a roadmap, et al:
http://exde.org/

I'm not involved with EXDE, I just would like to see some semblance of my current UI maintained, whatever the most fruitful path for that outcome may be.

[Djhg2000]

The more I read of this thread the more I realize... they're right.
We need to force users to suspend. It's the only way to make driver developers care.

It took years for my old Inspiron 6000 to get proper suspend support and I don't want to go through all of those times of accidentally loosing data again with my DV7.

So what do I do?
Well one option would be to shut down the machine every time I'm going somewhere, like I used to. Or I could work on fixing it.
Either way would make little difference since while I do know a lot more about the internal workings of Linux now than 5 years ago, I'm still just a lonely power user/small scale hobby developer with a single recent laptop.

By forcing everyone to use it (technically you can avoid it, but you know you will try it just once to see if it actually ended up working that way) a lot more people will complain about either the behavior of Gnome or how suspend is broken.

So, it's about raising the awareness of the suspend issues, not whether it's a sensible default. While this does indeed seem selfish from the point of view of the users who never suspend their laptop automatically, it will bring better drivers and probably a higher quality feel since one less thing is broken.

Oh and on a side note; nobody is stupid because of their preferences.

[kazetsukai]

The more I read of this thread the more I realize... they're right.
We need to force users to suspend. It's the only way to make driver developers care.

While you have a point, the ones that end up losing here are the users, especially new users. See below.

By forcing everyone to use it (technically you can avoid it, but you know you will try it just once to see if it actually ended up working that way) a lot more people will complain about either the behavior of Gnome or how suspend is broken.

Come on, you can't be serious? Power users and people with many years of Linux under their belt, maybe. New users? New users are going to blame "Linux", rather than "Gnome" or "The driver", period.

So, it's about raising the awareness of the suspend issues, not whether it's a sensible default. While this does indeed seem selfish from the point of view of the users who never suspend their laptop automatically, it will bring better drivers and probably a higher quality feel since one less thing is broken.

You're assuming that Gnome alone will drive driver development, regardless of the facts:

• It is not the responsibility or goal of Gnome to do so!
• Driver developers that have these kinds of problematic suspend code are either hobbyists who don't really care about that functionality, or part of a company which develops their own drivers and did not prioritize the feature. In either case, there is more motivtion for them to come back and say "Don't use Gnome" than actually fix the problem.
• Gnome is more or less the only ones doing this.

Observation: When KDE4 comes along and starts putting OpenGL 3 desktop effects into KWin, people scream bloody murder. When Gnome3 mandates suspend on lid close regardless of the fact that suspend is known not to work on several mobile platforms, the same people say its cool? Really? When closing the lid is something almost every laptop user is going to /have/ to do at one point in the lifetime of their laptop?

The pushing factor behind driver development of drivers for workstation and consumer PCs is based on one thing: Market presence. If noone used Linux, no or person would develop drivers for it. Companies and hobbyists don't have a clear count, but they do know that millions of people use Linux on the desktop. Every user makes a part of that invisible count, be it new users, experienced users, or developers. Forcing buggy behavior on users will scare some away from the platform, and is definitely the wrong approach. The right thing to do is to have bug tracking in place for upstream issues like this.

[Djhg2000]

Come on, you can't be serious? Power users and people with many years of Linux under their belt, maybe. New users? New users are going to blame "Linux", rather than "Gnome" or "The driver", period.

True. I don't really think it's the best way to do it either, but I think I understand why they do it which is what must of us have been asking in this thread.

You're assuming that Gnome alone will drive driver development, regardless of the facts:

• It is not the responsibility or goal of Gnome to do so!
• Driver developers that have these kinds of problematic suspend code are either hobbyists who don't really care about that functionality, or part of a company which develops their own drivers and did not prioritize the feature. In either case, there is more motivtion for them to come back and say "Don't use Gnome" than actually fix the problem.
• Gnome is more or less the only ones doing this.

Well, KDE4 did, and Gnome has a possibly even larger user base.
I think this is meant to put pressure on the distro devs as it will probably take a while after release until it's actually implemented in Ubuntu, Fedora, openSUSE and so on, not to mention being the default option.
New users most likely won't try Gnome 3 until it becomes just that; the default desktop environment.

Observation: When KDE4 comes along and starts putting OpenGL 3 desktop effects into KWin, people scream bloody murder. When Gnome3 mandates suspend on lid close regardless of the fact that suspend is known not to work on several mobile platforms, the same people say its cool? Really? When closing the lid is something almost every laptop user is going to /have/ to do at one point in the lifetime of their laptop?

I don't think it's cool, I think it's just a way to push development so that it doesn't take half of a laptop lifetime to get suspend support.

The pushing factor behind driver development of drivers for workstation and consumer PCs is based on one thing: Market presence. If noone used Linux, no or person would develop drivers for it. Companies and hobbyists don't have a clear count, but they do know that millions of people use Linux on the desktop. Every user makes a part of that invisible count, be it new users, experienced users, or developers. Forcing buggy behavior on users will scare some away from the platform, and is definitely the wrong approach. The right thing to do is to have bug tracking in place for upstream issues like this.

Very true, but would Windows ever get this far if people wouldn't tolerate a little buggy behavior?
In fact, I think that's about the only reason Windows is still used for things like browsing the web and writing papers instead of being reduced to a gaming platform.

[Alejandro Nova]

I have my KDE desktop configured to a) when on batteries, hibernate on lid close; b) when on AC power, DO NOTHING. I love to tune some radio station with Bangarang, close the notebook and leave it on. You are leaving me without my radio, and you are leaving me without my hibernation (I need HIBERNATION, not suspension).

I was thinking about testing GNOME 3.0. With this, guys, you close the door on my face.