My first fedora 11 beta run-in with not having dontzap set to false.
I have a dual twin setup. This already requires an xorg.conf file, because in the no-config running, xorg decides you want to only use one of the two screens it finds. I don't know why, I consider it a bug.
I configure it to have dual screen. When I login, my window manager dies on me. My gnome panel only works in my left screen, but anything I select pops up on the right screen, out of focus. I can move my mouse into the other screen but I cannot get any focus. So I cannot get to a terminal or any other application.
So I ctrl-alt-f2. Now my left monitor goes into standby mode and my right monitor just shows a blinking cursor. The same for all VC's except alt-f7 which brings me back to my broken X session.
I had to login via ssh to fix my machine, because I could not restart X. Indeed, i did not have sysrq enabled yet, that's disabled per default it seems.
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Paul
On 4/7/2009 4:31 PM, Paul Wouters wrote:
My first fedora 11 beta run-in with not having dontzap set to false.
I have a dual twin setup. This already requires an xorg.conf file, because in the no-config running, xorg decides you want to only use one of the two screens it finds. I don't know why, I consider it a bug.
I configure it to have dual screen. When I login, my window manager dies on me. My gnome panel only works in my left screen, but anything I select pops up on the right screen, out of focus. I can move my mouse into the other screen but I cannot get any focus. So I cannot get to a terminal or any other application.
So I ctrl-alt-f2. Now my left monitor goes into standby mode and my right monitor just shows a blinking cursor. The same for all VC's except alt-f7 which brings me back to my broken X session.
I had to login via ssh to fix my machine, because I could not restart X. Indeed, i did not have sysrq enabled yet, that's disabled per default it seems.
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this. It appears that you, and all others, have two options.
1: Live with it as default 'off'.
2: Change it to 'on' to suit yourself.
Great thing about Linux. You can do pretty much just as you wish.
Good luck.
On Tue, Apr 7, 2009 at 5:43 PM, David dgboles@comcast.net wrote:
On 4/7/2009 4:31 PM, Paul Wouters wrote:
My first fedora 11 beta run-in with not having dontzap set to false.
I have a dual twin setup. This already requires an xorg.conf file, because in the no-config running, xorg decides you want to only use one of the two screens it finds. I don't know why, I consider it a bug.
I configure it to have dual screen. When I login, my window manager dies on me. My gnome panel only works in my left screen, but anything I select pops up on the right screen, out of focus. I can move my mouse into the other screen but I cannot get any focus. So I cannot get to a terminal or any other application.
So I ctrl-alt-f2. Now my left monitor goes into standby mode and my right monitor just shows a blinking cursor. The same for all VC's except alt-f7 which brings me back to my broken X session.
I had to login via ssh to fix my machine, because I could not restart X. Indeed, i did not have sysrq enabled yet, that's disabled per default it seems.
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this. It appears that you, and all others, have two options.
Indeed. It's pretty clear that any argument to change this new default is doomed to failure. This thread is purely noise at this point.
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
Kevin Kofler
On Wed, 2009-04-08 at 08:39 +0200, Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
Kevin Kofler
+1
I haven't ever triggered it "accidently" till date.
regards,
Ankur
/me tried to stay aways from this discussion, but now I bite
On 08.04.2009 08:55, Ankur Sinha wrote:
On Wed, 2009-04-08 at 08:39 +0200, Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about. It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
I haven't ever triggered it "accidently" till date.
I have. Multiple times actually. And I don't even know how emaxs is spelled or what it actually is.
Any yes, I missed it myself already now and then since I updated to rawhide. But I'm nevertheless glad it's gone.
But whatever: Mailing lists are not a good place to voting about "enabling or disabling Ctrl+Alt+BkSp". Only two of the several reasons:
- subscribers to lists like fedora-devel are quite different from ordinary users
- we never know how many people agree or disgree but don't reply
So we will never get any reliable data by a discussion or voting like this. Even a voting by all those that are registered in FAS would not help, as the problems would be similar.
CU knurd
On Wed, Apr 8, 2009 at 2:12 AM, Thorsten Leemhuis fedora@leemhuis.info wrote:
/me tried to stay aways from this discussion, but now I bite
On 08.04.2009 08:55, Ankur Sinha wrote:
On Wed, 2009-04-08 at 08:39 +0200, Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about. It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
I haven't ever triggered it "accidently" till date.
I have. Multiple times actually. And I don't even know how emaxs is spelled or what it actually is.
Any yes, I missed it myself already now and then since I updated to rawhide. But I'm nevertheless glad it's gone.
But whatever: Mailing lists are not a good place to voting about "enabling or disabling Ctrl+Alt+BkSp". Only two of the several reasons:
- subscribers to lists like fedora-devel are quite different from ordinary
users
- we never know how many people agree or disgree but don't reply
What you do have an idea of is the percentage of the people who care enough to be here in the first place. Ignoring something based on what an unknown number of people who probably don't care either way is weird at best.
At this point, my understanding is that those that care don't have a say -- so get over it.
On 04/08/2009 01:12 AM, Thorsten Leemhuis wrote:
- we never know how many people agree or disgree but don't reply
I didn't want to reply but... what the heck -- I care and would prefer to have Ctrl+Alt+BkSp back.
On 04/12/2009 09:37 PM, Dariusz J. Garbowski wrote:
On 04/08/2009 01:12 AM, Thorsten Leemhuis wrote:
- we never know how many people agree or disgree but don't reply
I didn't want to reply but... what the heck -- I care and would prefer to have Ctrl+Alt+BkSp back.
http://who-t.blogspot.com/2009/04/zapping-server.html
In Fedora 11, you'll be able to turn it on just like any other keyboard hotkey combination.
~spot
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway tcallawa@redhat.com wrote:
On 04/12/2009 09:37 PM, Dariusz J. Garbowski wrote:
On 04/08/2009 01:12 AM, Thorsten Leemhuis wrote:
- we never know how many people agree or disgree but don't reply
I didn't want to reply but... what the heck -- I care and would prefer to have Ctrl+Alt+BkSp back.
http://who-t.blogspot.com/2009/04/zapping-server.html
In Fedora 11, you'll be able to turn it on just like any other keyboard hotkey combination.
We live on a ship of fools.
An excerpt from the Fedora foundations page:
Like any friends, we occasionally disagree on details, but we believe in finding an acceptable consensus to serve the interests of advancing free software. We believe in a strong partnership between Red Hat and our enormous volunteer community, since they both provide essential contributions that help the Fedora Project succeed.
Why is this principle being ignored with the ctrl-alt-bksp issue?
I think it would be better to have a hotkey combination to turn it *off* and just leave it enabled by default as it has been and as it should be (no, Peter Hutterer, you are completely wrong).
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway tcallawa@redhat.com wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Assume that the server never supported zapping in the past. Now we add a feature that immediately and without asking terminates your session, shuts down all applications, logs you out, brings down your wireless network in the process, shuts down your VPN and generally makes the computer giggle at you.
That is one way to look at it. You could also see it as adding a cool innovative new feature which allows inexperienced users to easily recover from a bad X state.
Experienced users will benefit from zapping. So make it accessible to them.
No. Inexperienced users will benefit from zapping. So it should be enabled by default.
Could you imagine if this was actually put up for a vote?
"Stay with default" option would be hands-down winner.
Good thing we will never see a vote on this, because we have to keep the x.org egos inflated. They could never possibly make a brain dead lemming like decision.
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Orcan
Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Many people in thread here and other places like the freedesktop mailing list and in http://lwn.net/Articles/327141/ have claimed to accidentally tripped this key combo. It is not that hard.
Rahul
On Mon, Apr 13, 2009 at 2:41 AM, Rahul Sundaram wrote:
Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Many people in thread here and other places like the freedesktop mailing list and in http://lwn.net/Articles/327141/ have claimed to accidentally tripped this key combo. It is not that hard.
Rahul
I really don't know what to say. It feels like the probability of burning my hair on the stove while sticking my foot in the freezer.
Anyway, I opened a poll at fedoraforum about this: http://www.fedoraforum.org/forum/showthread.php?t=219559
I do wonder what will come out. Of course this will not reflect the true statistics of all Fedora userbase. It never does. But it might give an idea.
Orcan
Orcan Ogetbil wrote:
I really don't know what to say. It feels like the probability of burning my hair on the stove while sticking my foot in the freezer.
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally. It doesn't have to be a majority necessarily. I know atleast one such case personally where a friend of mine wanted to press control alt del and since the delete key was close to the backspace key, he accidentally pressed it. When (you are new to Linux ) you do it accidentally, the experience is apparently quite disturbing since you often lose data and there is no indication that it is a expected behaviour of the system rather than a crash.
Ctl-alt-del is a well know Windows shortcut for rebooting but atleast prompts before doing it (same thing happens in GNOME as well btw). I personally think that the SUSE patch is a decent compromise and it looks like fedora-setup-keyboard is going to make it a easy toggle in Fedora as well.
Anyway, I opened a poll at fedoraforum about this: http://www.fedoraforum.org/forum/showthread.php?t=219559
I do wonder what will come out. Of course this will not reflect the true statistics of all Fedora userbase. It never does. But it might give an idea.
You might want to avoid putting in your own judgement in a neutral survey. Leading questions often distort the nature of the answer.
Rahul
Rahul Sundaram wrote:
Orcan Ogetbil wrote:
I really don't know what to say. It feels like the probability of burning my hair on the stove while sticking my foot in the freezer.
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally. It doesn't have to be a majority necessarily. I know atleast one such case personally where a friend of mine wanted to press control alt del and since the delete key was close to the backspace key, he accidentally pressed it. When (you are new to Linux ) you do it accidentally, the experience is apparently quite disturbing since you often lose data and there is no indication that it is a expected behaviour of the system rather than a crash.
Let me ask this, how many times after that first "learning experience" did your friend hit that key combo by accident again? Let me ask another question, does he use EMACS?(had to throw that in so you could say "stop with the conspiracy" one more time.) If he doesn't use emacs, the likelihood of him doing it again is about the same as him tripping over the power cord and pulling it from the wall. While possible, its not probable.
Ctl-alt-del is a well know Windows shortcut for rebooting but atleast prompts before doing it (same thing happens in GNOME as well btw). I personally think that the SUSE patch is a decent compromise and it looks like fedora-setup-keyboard is going to make it a easy toggle in Fedora as well.
Anyway, I opened a poll at fedoraforum about this: http://www.fedoraforum.org/forum/showthread.php?t=219559
I do wonder what will come out. Of course this will not reflect the true statistics of all Fedora userbase. It never does. But it might give an idea.
You might want to avoid putting in your own judgement in a neutral survey. Leading questions often distort the nature of the answer.
Rahul
The poll isn't neutral at all. It is clear how the poll creator feels about this change.
What this comes down to is, I lose a feature because some people pay no attention to detail, or should that be, use emacs and pay no attention to detail? It's a conspiracy!
TK I mash keys without thinking, thank you for saving me 009
TK009 wrote:
Let me ask this, how many times after that first "learning experience" did your friend hit that key combo by accident again?
I have no idea. He doesn't use Linux that often anyway these days.
Let me ask another question, does he use EMACS?(had to throw that in so you could say "stop with the conspiracy" one more time.)
Nothing to do with Emacs at all.
If he doesn't use emacs, the likelihood of him doing it again is about the same as him tripping over the power cord and pulling it from the wall. While possible, its not probable.
Even in other threads in the mailing list, people have said they have done so, repeatedly. I have read comments in other places as well. Apparently, it is not just possible but happening to people as well. Analysing one particular case isn't going to reveal much.
Rahul
TK009 wrote:
Let me ask this, how many times after that first "learning experience" did your friend hit that key combo by accident again?
I have no idea. He doesn't use Linux that often anyway these days.
Let me ask another question, does he use EMACS?(had to throw that in so you could say "stop with the conspiracy" one more time.)
Nothing to do with Emacs at all.
If he doesn't use emacs, the likelihood of him doing it again is about the same as him tripping over the power cord and pulling it from the wall. While possible, its not probable.
Even in other threads in the mailing list, people have said they have done so, repeatedly. I have read comments in other places as well. Apparently, it is not just possible but happening to people as well. Analysing one particular case isn't going to reveal much.
Rahul
On Mon, Apr 13, 2009 at 3:02 AM, Rahul Sundaram sundaram@fedoraproject.org wrote:
Orcan Ogetbil wrote:
I really don't know what to say. It feels like the probability of burning my hair on the stove while sticking my foot in the freezer.
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally. It doesn't have to be a majority necessarily. I know atleast one such case personally where a friend of mine wanted to press control alt del and since the delete key was close to the backspace key, he accidentally pressed it. When (you are new to Linux ) you do it accidentally, the experience is apparently quite disturbing since you often lose data and there is no indication that it is a expected behaviour of the system rather than a crash.
I'm confused here. He was attempting to restart the entire machine, and instead restarted the X server, and this is considered a significant problem?
Ctl-alt-del is a well know Windows shortcut for rebooting but atleast prompts before doing it (same thing happens in GNOME as well btw). I personally think that the SUSE patch is a decent compromise and it looks like fedora-setup-keyboard is going to make it a easy toggle in Fedora as well.
First of all, I'm not sure why what Windows does is really important. I think as a general rule for using machine, one shouldn't generally randomly press buttons -- on purpose or by accident -- and expect it to be all okay.
That said... when did Ctrl+Alt+Delete stop rebooting? I just tried it and was surprised to see a dialog box instead. I remember using that key combo way back since RH 7.3, I just rarely needed it.
Arthur Pemberton wrote:
I'm confused here. He was attempting to restart the entire machine, and instead restarted the X server, and this is considered a significant problem?
Yes because control alt del calls up the task manager in Windows in recent versions and even initiating a reboot, you get a prompt for confirmation and applications get time to save data.
That said... when did Ctrl+Alt+Delete stop rebooting? I just tried it and was surprised to see a dialog box instead. I remember using that key combo way back since RH 7.3, I just rarely needed it.
I don't recall the specifics but it has been that way for a long time now.
Rahul
Rahul Sundaram wrote:
Arthur Pemberton wrote:
I'm confused here. He was attempting to restart the entire machine, and instead restarted the X server, and this is considered a significant problem?
Yes because control alt del calls up the task manager in Windows in recent versions and even initiating a reboot, you get a prompt for confirmation and applications get time to save data.
well there we have it! another who wants fedora to behave like it's written by m$, and i thought the redmond theme was as close as fedora would get to windows :(
That said... when did Ctrl+Alt+Delete stop rebooting? I just tried it and was surprised to see a dialog box instead. I remember using that key combo way back since RH 7.3, I just rarely needed it.
I don't recall the specifics but it has been that way for a long time now.
Rahul
psmith wrote:
Rahul Sundaram wrote:
Arthur Pemberton wrote:
I'm confused here. He was attempting to restart the entire machine, and instead restarted the X server, and this is considered a significant problem?
Yes because control alt del calls up the task manager in Windows in recent versions and even initiating a reboot, you get a prompt for confirmation and applications get time to save data.
well there we have it! another who wants fedora to behave like it's written by m$, and i thought the redmond theme was as close as fedora would get to windows :(
I was explaining to Arthur the difference between zapping X in Linux vs rebooting in Windows. That difference in user experience explains why it is important to not equate the two things together. This has nothing to do with wanting Fedora to be like Windows.
Rahul
Arthur Pemberton wrote:
I'm confused here. He was attempting to restart the entire machine, and instead restarted the X server, and this is considered a significant problem?
Yes because control alt del calls up the task manager in Windows in recent versions and even initiating a reboot, you get a prompt for confirmation and applications get time to save data.
That said... when did Ctrl+Alt+Delete stop rebooting? I just tried it and was surprised to see a dialog box instead. I remember using that key combo way back since RH 7.3, I just rarely needed it.
I don't recall the specifics but it has been that way for a long time now.
Rahul
On 4/13/2009 3:57 PM, Callum Lerwick wrote:
On Mon, 2009-04-13 at 13:32 +0530, Rahul Sundaram wrote:
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally.
So change it to ctl-alt-lshift-rshift-bs.
Every time I disable this /dev/null void.
Honest to $Diety. *All* of you. Change it to something. Anything. Or don't. But *please* shut the hell up about it. Be a man and suck in in. Deal with it and stop whining. Shessh.
David wrote:
On 4/13/2009 3:57 PM, Callum Lerwick wrote:
On Mon, 2009-04-13 at 13:32 +0530, Rahul Sundaram wrote:
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally.
So change it to ctl-alt-lshift-rshift-bs.
Every time I disable this /dev/null void.
Honest to $Diety. *All* of you. Change it to something. Anything. Or don't. But *please* shut the hell up about it. Be a man and suck in in. Deal with it and stop whining. Shessh.
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
Adam obviously has not given much thought to the ramifications this change has. Hey Adam, why don't you pretend you are sitting behind a help desk trying to help out newbies? Maybe it is *you* who needs to grow some grey matter.
Christopher Stone wrote:
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating.
I don't think I am exaggerating.
This is a pretty trivial issue.
It is not - It is Red Hat and X having abandoned a substantial and vital feature.
IMO, it's equivalent to them having removed the "emergency brakes". A step they should be ashamed of of.
Grow a sense of perspective.
Adam obviously has not given much thought to the ramifications this change has.
The #1 ramification will be people finding themselves in situations of having "to unplugging the power cord" to get out X when "X/their desktop runs wild/becomes unusable".
Great progress, great achievement!
Ralf
On Tue, Apr 14, 2009 at 6:20 PM, Christopher Stone chris.stone@gmail.com wrote:
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
Why do people at redhat and/or x.org seem to think the zap feature is trivial or unimportant or not useful? I hate to burst your bubbles, but despite all the cool things X can do, ctrl-alt-backspace (as simplistic and trivial as it is) is probably one of the most useful, cool, recognizable features in X. If you asked a newbie what is the coolest feature X has, they would probably say ctrl-alt-backspace. Im sure it makes everyone at x.org cringe.
Christopher Stone wrote:
On Tue, Apr 14, 2009 at 6:20 PM, Christopher Stone chris.stone@gmail.com wrote:
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
Why do people at redhat and/or x.org seem to think the zap feature is trivial or unimportant or not useful?
Probably the fact it's not being used "when things just work" and is only _necessary_ in situations of emergency/"when thinks go bizarre"
"Infrequently used" == "unnecessary"?
Some food for thought: How to quit X when X comes up but the window-manager fails to come up?
In recent days, this has happened several times to me on FC10 (reason yet unknown, probably somebody having messed up something related to nis/yp, nfs, autofs). My escape: ctl-alt-bs.
Ralf
On Tue, Apr 14, 2009 at 6:54 PM, Ralf Corsepius rc040203@freenet.de wrote:
Christopher Stone wrote:
On Tue, Apr 14, 2009 at 6:20 PM, Christopher Stone chris.stone@gmail.com wrote:
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
Why do people at redhat and/or x.org seem to think the zap feature is trivial or unimportant or not useful?
Probably the fact it's not being used "when things just work" and is only _necessary_ in situations of emergency/"when thinks go bizarre"
Perhaps, I'm sticking to my theory that telling an X programmer that "ctrl-alt-bksp is one of the coolest features in X" is the worst possible insult you can give to an X programmer. So they are disabling it by default. :P
On Tue, 2009-04-14 at 18:20 -0700, Christopher Stone wrote:
On Tue, Apr 14, 2009 at 4:11 PM, Adam Williamson awilliam@redhat.com wrote:
On Tue, 2009-04-14 at 18:14 +0200, Ralf Corsepius wrote:
We aren't whining.
We are fiercely fighting to prevent serious harm to Fedora's usability.
Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective.
Adam obviously has not given much thought to the ramifications this change has. Hey Adam, why don't you pretend you are sitting behind a help desk trying to help out newbies? Maybe it is *you* who needs to grow some grey matter.
Heh...that's funny.
awilliamson at forum.mandriva.com: Total posts: 25452 [4.15% of total / 17.23 posts per day]
AdamW at fedoraforum.org: Join Date: Dec 2008 Location: Vancouver, BC Posts: 385
not to mention countless posts on osnews, distrowatch and zillions of tiny blogs all over Teh Intarwebs, the vast majority of which were, precisely, helping out newbies.
I don't usually bother with arguments based on personal reputation, but you clearly have no idea what I've been doing for a living for a very long time. :)
On Wed, 2009-04-15 at 14:50 -0400, Paul Wouters wrote:
On Wed, 15 Apr 2009, Adam Williamson wrote:
Heh...that's funny.
you clearly have no idea what I've been doing for a living for a very long time. :)
Fedora QA Community Monkey
Ahh, so this is about job continuity[1]? :)
Paul [1] It's a joke. Sorry, couldn't resist.
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace. Total cost: about twenty seconds (time to reboot vs. time to restart X). Killing X kills all X apps in any case, so they're not going to lose any data rebooting that they wouldn't have lost anyway by doing ctrl-alt-backspace. Or are our newbies running emacs in virtual consoles now?
On Wed, 15 Apr 2009, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace. Total cost: about twenty seconds (time to reboot vs. time to restart X).
Oh. it is a secret project to test data loss for ext4!
Killing X kills all X apps in any case, so they're not going to lose any data rebooting that they wouldn't have lost anyway by doing ctrl-alt-backspace.
UNEXPECTED FILE INCONSISTENTY, please run fsck manually
We'll see if the newbies can find /dev/mapper/VolGroup00-LogVol00 as their disk and see if the figure out the "-y" option in time for F12 :)
Paul ps. This is slightly less of a joke. It's the real difference of having or not having ctlr-alt-bksp (or alt-sysrq-o, or was it alt-sysrq-r? and did I have to wait for it to print 'OK' before hitting alt-sysrq-b again? oh wait sysctl.conf had this disabled.)
Paul Wouters wrote:
On Wed, 15 Apr 2009, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace. Total cost: about twenty seconds (time to reboot vs. time to restart X).
Oh. it is a secret project to test data loss for ext4!
Killing X kills all X apps in any case, so they're not going to lose any data rebooting that they wouldn't have lost anyway by doing ctrl-alt-backspace.
UNEXPECTED FILE INCONSISTENTY, please run fsck manually
I've been doing my damnedest to not follow this thread, but ... neither ctrl-alt-backspace nor a reboot should result in a corrupted filesytsem or an unclean fs shutdown....
-Eric
Eric Sandeen wrote:
I've been doing my damnedest to not follow this thread, but ... neither ctrl-alt-backspace nor a reboot should result in a corrupted filesytsem or an unclean fs shutdown....
If their system does not respond to attempts to reboot from the menu (or in some cases even if it would because they don't even bother trying), users will just hit the reset button or pull the plug, and in that case the file system has no time to clean up anything.
Kevin Kofler
On Thu, 2009-04-16 at 00:03 +0200, Kevin Kofler wrote:
If their system does not respond to attempts to reboot from the menu (or in some cases even if it would because they don't even bother trying), users will just hit the reset button or pull the plug, and in that case the file system has no time to clean up anything.
Which still shouldn't cause a need for an interactive fsck session. That's why the journal is there.
Jesse Keating wrote:
On Thu, 2009-04-16 at 00:03 +0200, Kevin Kofler wrote:
If their system does not respond to attempts to reboot from the menu (or in some cases even if it would because they don't even bother trying), users will just hit the reset button or pull the plug, and in that case the file system has no time to clean up anything.
Which still shouldn't cause a need for an interactive fsck session. That's why the journal is there.
Thank you.
-Eric
2009/4/16 Jesse Keating jkeating@redhat.com:
On Thu, 2009-04-16 at 00:03 +0200, Kevin Kofler wrote:
Which still shouldn't cause a need for an interactive fsck session. That's why the journal is there
No it does require fsck sometimes. I had some physical error on one of my ext2 partitions. Every time there was a unclean shutdown there would be a drdy error and I would have to do fsck -y /dev/sdax.
2009/4/16 Kevin Kofler kevin.kofler@chello.at:
Ext2 is not journaled. Do you mean ext3
Ahh, my bad. I was using ext2. I recently upgraded to ext3 and it runs fine.
On Wed, Apr 15, 2009 at 11:59 AM, Adam Williamson awilliam@redhat.com wrote:
On Wed, 2009-04-15 at 14:50 -0400, Paul Wouters wrote:
On Wed, 15 Apr 2009, Adam Williamson wrote:
Heh...that's funny.
you clearly have no idea what I've been doing for a living for a very long time. :)
Fedora QA Community Monkey
Ahh, so this is about job continuity[1]? :)
Paul [1] It's a joke. Sorry, couldn't resist.
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace. Total cost: about twenty seconds (time to reboot vs. time to restart X). Killing X kills all X apps in any case, so they're not going to lose any data rebooting that they wouldn't have lost anyway by doing ctrl-alt-backspace. Or are our newbies running emacs in virtual consoles now?
You're correct. I guess we will just have to live with the fact that X got a little bit lamer. I don't fully understand why, perhaps it's an ego thing, perhaps X strives to be more like windows, perhaps people are just clueless and dumb or maybe a little bit of all of the above. Maybe Microsoft is paying the X guys to disable the feature by default, I don't know? I guess no one will ever know the *real* reason why the defaults were changed.
Adam Williamson wrote:
On Wed, 2009-04-15 at 14:50 -0400, Paul Wouters wrote:
On Wed, 15 Apr 2009, Adam Williamson wrote:
Heh...that's funny. you clearly have no idea what I've been doing for a living for a very long time. :) Fedora QA Community Monkey
Ahh, so this is about job continuity[1]? :)
Paul [1] It's a joke. Sorry, couldn't resist.
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
Very helpful ....
On Thu, 2009-04-16 at 06:20 +0200, Ralf Corsepius wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
On Wed, Apr 15, 2009 at 10:09 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2009-04-16 at 06:20 +0200, Ralf Corsepius wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
I like your logic. It's easy to change defaults, so therefore it makes sense to have bad defaults.
On Wed, 2009-04-15 at 22:50 -0700, Christopher Stone wrote:
On Wed, Apr 15, 2009 at 10:09 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2009-04-16 at 06:20 +0200, Ralf Corsepius wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
I like your logic. It's easy to change defaults, so therefore it makes sense to have bad defaults.
You're putting words in my mouth. I have no particular opinion either way. All I've ever said is that it isn't a big deal. Put it this way - the time lost by the possible drawbacks of *either* approach is extremely unlikely *ever* to add up to the amount of time and energy smart people have wasted in the seventeen thousand threads about this.
Adam Williamson wrote:
On Wed, 2009-04-15 at 22:50 -0700, Christopher Stone wrote:
On Wed, Apr 15, 2009 at 10:09 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2009-04-16 at 06:20 +0200, Ralf Corsepius wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
I like your logic. It's easy to change defaults, so therefore it makes sense to have bad defaults.
ACK. Push around users by changing defaults which force them to increasingly customize the distro.
=> Usabiitly regression.
You're putting words in my mouth. I have no particular opinion either way. All I've ever said is that it isn't a big deal.
Wrong, you are defending RH's decision not to revert this "ctl-alt-bs" insanity, i.e. you have taken a position.
Put it this way - the time lost by the possible drawbacks of *either* approach is extremely unlikely *ever* to add up to the amount of time and energy smart people have wasted in the seventeen thousand threads about this.
You @RH guys could easily have avoided these threads - It was solely your decision to ignore the community.
On Thu, 2009-04-16 at 09:28 +0200, Ralf Corsepius wrote:
Adam Williamson wrote:
On Wed, 2009-04-15 at 22:50 -0700, Christopher Stone wrote:
On Wed, Apr 15, 2009 at 10:09 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2009-04-16 at 06:20 +0200, Ralf Corsepius wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
You are presuming a newbie on a single seat/single user system.
Total cost: about twenty seconds (time to reboot vs. time to restart X).
In a corporite environment, BIOS passwords or similar will prevent them from rebooting. A service tech/sys-admin will have to come by.
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
I like your logic. It's easy to change defaults, so therefore it makes sense to have bad defaults.
ACK. Push around users by changing defaults which force them to increasingly customize the distro.
=> Usabiitly regression.
You're putting words in my mouth. I have no particular opinion either way. All I've ever said is that it isn't a big deal.
Wrong, you are defending RH's decision not to revert this "ctl-alt-bs" insanity, i.e. you have taken a position.
Put it this way - the time lost by the possible drawbacks of *either* approach is extremely unlikely *ever* to add up to the amount of time and energy smart people have wasted in the seventeen thousand threads about this.
You @RH guys could easily have avoided these threads - It was solely your decision to ignore the community.
No we ignored you and one other guy.
I'm sure every decision you make ever is for the good of the community, just as the X.org developers believe this was for the good of the community.
My only worry, is that nobody thought of the children.
Dave.
Dave Airlie wrote:
On Thu, 2009-04-16 at 09:28 +0200, Ralf Corsepius wrote:
You @RH guys could easily have avoided these threads - It was solely your decision to ignore the community.
No we ignored you and one other guy.
Announcement in your radio: "There's a ghostdriver on interstate Fedora" You: "One? Hundreds!"
I'm sure every decision you make ever is for the good of the community, just as the X.org developers believe this was for the good of the community.
I don't care what the X.org guys do - I care about what the RH people who maintain the X-packages in Fedora do: Turn Fedora into a "Ubuntu cult" and turn Linux into a poor single-user/single-seat Windows imitation.
In case I wasn't clear: The "ctl-alt-bs" issue only is the tip of the iceberg. Much of what RH has pushed into fedora is from same category.
... turning away in disgust.
On Thu, 2009-04-16 at 09:28 +0200, Ralf Corsepius wrote:
You're putting words in my mouth. I have no particular opinion either way. All I've ever said is that it isn't a big deal.
Wrong, you are defending RH's decision not to revert this "ctl-alt-bs" insanity, i.e. you have taken a position.
IT'S A CONSPIRACY!!! IT'S A CONSPIRACY!!!
No, I'm not. I said:
"Oh, come on, quit exaggerating. This is a pretty trivial issue. Grow a sense of perspective."
That's not defending either position, just contending that the stature of the issue as a whole is not very high.
Adam Williamson wrote:
On Thu, 2009-04-16 at 09:28 +0200, Ralf Corsepius wrote:
You're putting words in my mouth. I have no particular opinion either way. All I've ever said is that it isn't a big deal.
Wrong, you are defending RH's decision not to revert this "ctl-alt-bs" insanity, i.e. you have taken a position.
IT'S A CONSPIRACY!!! IT'S A CONSPIRACY!!!
No, it's simply insanity, lack of knowledge and poor management.
That's not defending either position, just contending that the stature of the issue as a whole is not very high.
Wait, until you'll be facing the consequences of this change - This change will hit not only newbies (they are used to reboot), it will hit the experienced users and sysadmins.
If they feel sufficiently pissed, they will quit and leave Fedora alone.
On Thu, 2009-04-16 at 19:33 +0200, Ralf Corsepius wrote:
If they feel sufficiently pissed, they will quit and leave Fedora alone.
People threaten that all the time, and yet they never seem to follow through.
On 2009/04/16 10:44 (GMT-0700) Jesse Keating composed:
On Thu, 2009-04-16 at 19:33 +0200, Ralf Corsepius wrote:
If they feel sufficiently pissed, they will quit and leave Fedora alone.
People threaten that all the time, and yet they never seem to follow through.
How exactly do you gather data on who does or does not leave after threatening to leave?
On Thu, 2009-04-16 at 15:40 -0400, Felix Miata wrote:
On 2009/04/16 10:44 (GMT-0700) Jesse Keating composed:
On Thu, 2009-04-16 at 19:33 +0200, Ralf Corsepius wrote:
If they feel sufficiently pissed, they will quit and leave Fedora alone.
People threaten that all the time, and yet they never seem to follow through.
How exactly do you gather data on who does or does not leave after threatening to leave?
The bit where they keep posting to the forum or mailing list in question is usually a bit of a giveaway.
(You might know the two or three people I'm thinking of, here...:>)
On Friday 17 April 2009 00:50:05 Adam Williamson wrote:
On Thu, 2009-04-16 at 15:40 -0400, Felix Miata wrote:
(You might know the two or three people I'm thinking of, here...:>)
Either drop this "two or three people" point or add me to the list too (so it will be four).
To anyone who still hopes the change will be reverted: If You haven't figured it out Yourselves: The change *will not be reverted*. Also please note that the people You're arguing with are usually (always?) just random people who aren't the ones who made the decision. And no, You will not find anyone @RH who will agree that previews behavior was better, far from saying that the change should be reverted.
Suren
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
On Thu, 16 Apr 2009, Jesse Keating wrote:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
In a survey of 10 RH employees you will find between 10 and 40 different opinions. sometimes more if you don't ask some of them to confine their comments to a limited amount of time.
-sv
Seth Vidal wrote:
On Thu, 16 Apr 2009, Jesse Keating wrote:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
In a survey of 10 RH employees you will find between 10 and 40 different opinions. sometimes more if you don't ask some of them to confine their comments to a limited amount of time.
Classical situation to call "the management", let them decide and to name "someone to be in charge/responsible".
Unfortunately, this current FESCO has refused to decide.
On Fri, 17 Apr 2009 08:02:25 +0200 Ralf Corsepius rc040203@freenet.de wrote:
Seth Vidal wrote:
On Thu, 16 Apr 2009, Jesse Keating wrote:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
In a survey of 10 RH employees you will find between 10 and 40 different opinions. sometimes more if you don't ask some of them to confine their comments to a limited amount of time.
Classical situation to call "the management", let them decide and to name "someone to be in charge/responsible".
Unfortunately, this current FESCO has refused to decide.
No. FESCo did decide not to force maintainers to revert this change.
Speaking only for myself, I think such micro-management sets a bad precedent and should be avoided except in extraordinary situations, which I don't think this is.
kevin
On Friday 17 April 2009 02:55:51 Jesse Keating wrote:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
In fact that was exactly what I was saying. Of course there are (many) Red Hat employees who liked the previews default (and also those who like the new default), but no one from the former group appeared in this discussion.
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
How sad a statement that truly is.
Jesse, I'm a big fan of the all work that you do (it doesn't go unnoticed) and likely this comment has come off the back of getting annoyed of the amount of discussion there has been on the issue but if you really think that arguing about changes to default behaviour in Fedora is a waste of time then I'm not sure what to say.
The dissent, for the most part, shows how passionate people are about these changes. It is most certainly not a waste of time.
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead? Absolutely nothing? How ridiculous. Let's just make a bunch of extremely bad decisions and keep bad defaults because actually trying to convince someone that the defaults or decisions people make are bad is pointless.
I think we should remove the "Friends" section from the Fedora Foundations page. Because community input and discussion is a waste of time according to Jesse Keating.
On Fri, Apr 17, 2009 at 08:43:28AM -0700, Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead? Absolutely nothing? How ridiculous. Let's just make a bunch of extremely bad decisions and keep bad defaults because actually trying to convince someone that the defaults or decisions people make are bad is pointless.
The people to convince are the maintainers of the code in question, and in the general case sprawling threads on distribution mailing lists aren't a good way to do that. Some worthwhile ideas came out of this discussion and were implemented - however, at this point it's just repeating the same arguments over and over again, and argument ad nauseum is not the most convincing approach.
One thing that should be borne in mind is that the Fedora X maintainers are a subset of the upstream X maintainers. Convincing Fedora to change the default is likely to be an equivalent task to convincing upstream to change the default. The only real way that any divergence could develop would be for the Fedora maintainers to be overruled.
My understanding is that if FESCO is unwilling to do this then the issue could (in principle) be raised to the board. If that doesn't lead anywhere then your issue becomes one with the overall governance of the project, and that's best discussed somewhere other than on a technical development mailing list.
In summary: The technical issues raised are well understood by the maintainers. Unless there's any further insight into those issues then this is not the mailing list to continue the discussions. It's simply no longer a development issue.
* Matthew Garrett mjg@redhat.com [20090417 18:52]:
On Fri, Apr 17, 2009 at 08:43:28AM -0700, Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead? Absolutely nothing? How ridiculous. Let's just make a bunch of extremely bad decisions and keep bad defaults because actually trying to convince someone that the defaults or decisions people make are bad is pointless.
The people to convince are the maintainers of the code in question, and in the general case sprawling threads on distribution mailing lists aren't a good way to do that. Some worthwhile ideas came out of this discussion and were implemented - however, at this point it's just repeating the same arguments over and over again, and argument ad nauseum is not the most convincing approach.
While I am not a representative figure of the user population, I had a case today where:
* Thunar made a "grab" and I could not click to change focus away. To all intents and purposes, I was in the situation that the contingency for preserving C-A-Bs as enabled by default is holding up as the primary reason.
* Switching to vc/2 gave a blank screen
* My sshd setup is to only allow pub-key auth coming in to my laptop and I did not have the right keys on another system yet set up (that's now rectified), so ssh in to kill Thunar didn't work
I was doing some pretty intricate stuff for several customers in parallel, and shooting the session through C-A-Bs was out of the question *even though it was available as an option to me*.
I ended up logging in on the console "black screen", because I knew there was a login there, and typing blind, I killed thunar off to recover my session and the hours worth of work I had sitting in my session.
So - from my perspective, being one of those hate-object "@RH" people, while C-A-Bs was available, my choice was not to use it because the work was more valuable than the 3-4 minutes to assess the situation and to try a few ways to recover it.
C-A-Bs is akin to Alt-SysRq-C if you have SysRq's enabled. The default setting should be "off" so that people by accident can't trigger it. If you need it - you get to enable it yourself. That is a sensible default for a debug function.
In all else you wrote Matthew, I agree with you.
(And while I have not yet filed a BZ against Thunar, that is forthcoming shortly.)
-- /Anders
On Fri, Apr 17, 2009 at 1:38 PM, Anders Rayner-Karlsson anders@trudheim.co.uk wrote:
- Matthew Garrett mjg@redhat.com [20090417 18:52]:
[ snip ]
So - from my perspective, being one of those hate-object "@RH" people, while C-A-Bs was available, my choice was not to use it because the work was more valuable than the 3-4 minutes to assess the situation and to try a few ways to recover it.
Having that choice was good though, wasn't it? In the same way you hadn't copied your keys to another machine, you maybe hadn't added the new zap option to xorg.conf, so you no longer have that option.
And I don't think a new user would be expected to type blind in a console... so they would have had to hit the reboot button, or the shutdown button if they don't have a reboot button.
For me my desktop has as many important services running as I have apps -- so restarting X is not as costly as restarting the machine.
* Arthur Pemberton pemboa@gmail.com [20090417 21:15]:
On Fri, Apr 17, 2009 at 1:38 PM, Anders Rayner-Karlsson anders@trudheim.co.uk wrote:
- Matthew Garrett mjg@redhat.com [20090417 18:52]:
[ snip ]
So - from my perspective, being one of those hate-object "@RH" people, while C-A-Bs was available, my choice was not to use it because the work was more valuable than the 3-4 minutes to assess the situation and to try a few ways to recover it.
Having that choice was good though, wasn't it? In the same way you hadn't copied your keys to another machine, you maybe hadn't added the new zap option to xorg.conf, so you no longer have that option.
Having the choice didn't make a blind bit of difference in this instance as I was determined to keep the work. If I in F11 want Zap capabilities, I can enable it. I'll probably leave it unmapped, so I am encouraged to file defects against misbehaving applications instead.
And I don't think a new user would be expected to type blind in a console... so they would have had to hit the reboot button, or the shutdown button if they don't have a reboot button.
Either way, they'd lose the content of their session. If X really is scrogged, hitting the power-button for a controlled shutdown so that the hardware is reset to sane state is not a bad idea *for the new users with limited skills*.
If you are savvy enough to know about Zap, you're savvy enough to switch it on, just like with SysRq's.
For me my desktop has as many important services running as I have apps -- so restarting X is not as costly as restarting the machine.
I'm glad for you, that you have an X-session so lightweight that you can Zap it without losing any work. I long for the days when I had so little to do on my desktop that I could make such a statement. ;)
-- /Anders
On 04/17/2009 08:38 PM, Anders Rayner-Karlsson wrote:
While I am not a representative figure of the user population, I had a case today where:
Yes, but a new Linux user will not be able to do what you describe. They will try all things they can come up with, pressing the keyboard blindly etc., and finally press the reset button, or power cycle the computer.
I can not see that this is a good way to solve a problem where a few (!) persons accidentally have pressed ctrl-alt-backspace in the past.
Ctrl-alt-backspace *is* useful, and should, IMHO, be *on* by default, as it always has been. If upstream thinks otherwise, Fedora should be able to set it to on, as, as seen in this thread, and apparently in a lot of others, a lot of users are *against* removing this feature.
C-A-Bs is akin to Alt-SysRq-C if you have SysRq's enabled. The default setting should be "off" so that people by accident can't trigger it. If you need it - you get to enable it yourself. That is a sensible default for a debug function.
How often does people actually accidentally press ctrl-alt-backspace? I have *never* done it.
I do not consider ctrl-alt-backspace a debug function. Sadly programs still hangs X once in a while, and having a keystroke available to press to kill X, when all you need to do is killing X, is good thing to have, and it should, IMHO, be available till we have a X that does not hang as often as it still does. Why having to reboot the machine when you actually only need to restart X? It does not make any sense, at least not to me.
/Lars
On Saturday 18 April 2009 00:36:48 Lars E. Pettersson wrote:
Why having to reboot the machine when you actually only need to restart X? It does not make any sense, at least not to me.
Of course it shouldn't make sense to anyone. 1. Kill X All the applications which are running on *that* X are detached. Most of them close right away. Some (the ones which are better designed) save state and exit only after that. Not that there are many programs in the latter group and in fact I don't know which programs do that, but at least they have all the chance to do it. 2. Reset No program has any chance to save anything - no matter how good they are. And the kernel is no exception.
So it's as simple as this: Kill X - lose data Reset - lose at least the same data + have a *big* chance of FS problems (and yes sometimes even completely hosed system).
<sarcasm>Which one is better? - You know, it's just a matter of taste.</sarcasm>
Suren
* Lars E. Pettersson lars@homer.se [20090417 21:37]:
On 04/17/2009 08:38 PM, Anders Rayner-Karlsson wrote:
While I am not a representative figure of the user population, I had a case today where:
Yes, but a new Linux user will not be able to do what you describe. They will try all things they can come up with, pressing the keyboard blindly etc., and finally press the reset button, or power cycle the computer.
You are not convincing me that Zap should be enabled by default with that statement. Maybe I am coloured by experience though.
I can not see that this is a good way to solve a problem where a few (!) persons accidentally have pressed ctrl-alt-backspace in the past.
The counter-point is that it is hard to see a good or viable reason for having a on-by-default ability to kill your session, your running applications and lose data, without confirmation or warning.
Ctrl-alt-backspace *is* useful, and should, IMHO, be *on* by default, as it always has been. If upstream thinks otherwise, Fedora should be able to set it to on, as, as seen in this thread, and apparently in a lot of others, a lot of users are *against* removing this feature.
I am not arguing that it is useful or not. I have use for SysRq's and find them useful, but I am not arguing for them to be on by default. In any event, flogging a stinking horse carcass repeatedly won't get the carcass winning the next Grand National. ;)
From one perspective, a safe default while still allowing skilled
users to switch on "poweruser" features is a better starting point from a usability argument.
"Once you know how to start a fire, you can proceed to singe your eyebrows off with it."
C-A-Bs is akin to Alt-SysRq-C if you have SysRq's enabled. The default setting should be "off" so that people by accident can't trigger it. If you need it - you get to enable it yourself. That is a sensible default for a debug function.
How often does people actually accidentally press ctrl-alt-backspace? I have *never* done it.
By same argument, taking a rather tongue in cheek attitude, how often does countries with nuclear ICBM's launch them by accident?
While you certainly can present an argument based on that you have _never_ done this by accident, I'd like to point you at http://en.wikipedia.org/wiki/Argument_from_ignorance for reasons why this is, IMHO, not a sustainable platform to argue from.
I do not consider ctrl-alt-backspace a debug function. Sadly programs still hangs X once in a while, and having a keystroke available to press to kill X, when all you need to do is killing X, is good thing to have, and it should, IMHO, be available till we have a X that does not hang as often as it still does. Why having to reboot the machine when you actually only need to restart X? It does not make any sense, at least not to me.
While I disagree with your classification of the C-A-Bs functionality, lets just agree to disagree on that, I agree that some programs still cause issues with X (having today encountered it first hand).
As has been explained *extensively* already, you will still have the ability to do what you desire - but you will need to (trivially) enable this yourself in future. As also has been posited already, this has been blown way out of proportion, quite possibly based on erroneous assumptions.
-- /Anders
On 04/17/2009 10:32 PM, Anders Rayner-Karlsson wrote:
- Lars E. Pettersson lars@homer.se [20090417 21:37]:
How often does people actually accidentally press ctrl-alt-backspace? I have *never* done it.
By same argument, taking a rather tongue in cheek attitude, how often does countries with nuclear ICBM's launch them by accident?
It is not argument as such. If the decision has been made to remove this functionality, the decision has to be based on something. In this case it seem to be based on that people accidentally can press this key-stroke combination and loose data. If this is the case, it is important to know how often this actually happens.
If you look at real life. In our kitchens we have knifes. You can accidentally injure yourself quite badly with a knife. Should we just because of this zap all knife blades to make them safe?
I.e. we have made a decision that knifes are good to have, accidents do happen, but the benefit from having knifes are greater than the consequences of the accidents that can happen. So we keep our knifes.
This analogy converted to this discussion says that OK some, a few, may accidentally press ctrl-alt-backspace, but at the same time this particular key-stroke is very handy under those circumstances when X behaves badly. By removing the functionality you also adds another cost, the cost of extra data loss and other problems, i.e. with file systems, as user, when ctrl-alt-backspace does not work, finally will press the reset button or power cycle their computer. As I see it the benefit from having ctrl-alt-backspace is greater than the cost of a few loosing data.
I.e. how often this problem happens, people accidentally pressing ctrl-alt-backspace, is a valid question in this discussion. Does it happen more often than X crashes? Or less? This is an important parameter in the decision to keep, or not keep, the functionality.
While you certainly can present an argument based on that you have _never_ done this by accident, I'd like to point you at http://en.wikipedia.org/wiki/Argument_from_ignorance for reasons why this is, IMHO, not a sustainable platform to argue from.
Sigh! I just mentioned that *I* had never accidentally pressed this key combination during all those years that I have been using Linux, and therefore find it strange that this has become such a big issue that some wants to remove this functionality. It was *NOT* ment to be a platform to argue from, such an argument would be plainly stupid. If I during all my years have never accidentally pressed this combination, how often does it happen to others? I.e. for me X crashes have happened way more often then me accidentally pressing ctrl-alt-backspace. How is it for others?
English is not my first language, I hope you get what I am getting at this time...
As has been explained *extensively* already, you will still have the ability to do what you desire - but you will need to (trivially) enable this yourself in future. As also has been posited already, this has been blown way out of proportion, quite possibly based on erroneous assumptions.
What I will do or not do, or what the consequences will be for me personally is actually totally un-important for this discussion. I will just enable the functionality and live on as before.
The important thing is that this removal can create problems for new user of Linux. Why should we learn them to restart the whole machine when it is not needed, and even may create extra data loss, and even problems with the file system. I actually can not understand this. Should we not make life easy for the new ones?
/Lars
On Fri, Apr 17, 2009 at 11:29 PM, Lars E. Pettersson lars@homer.se wrote:
On 04/17/2009 10:32 PM, Anders Rayner-Karlsson wrote:
- Lars E. Pettersson lars@homer.se [20090417 21:37]:
How often does people actually accidentally press ctrl-alt-backspace? I have *never* done it.
By same argument, taking a rather tongue in cheek attitude, how often does countries with nuclear ICBM's launch them by accident?
It is not argument as such. If the decision has been made to remove this functionality, the decision has to be based on something. In this case it seem to be based on that people accidentally can press this key-stroke combination and loose data. If this is the case, it is important to know how often this actually happens.
If you look at real life. In our kitchens we have knifes. You can accidentally injure yourself quite badly with a knife. Should we just because of this zap all knife blades to make them safe?
I.e. we have made a decision that knifes are good to have, accidents do happen, but the benefit from having knifes are greater than the consequences of the accidents that can happen. So we keep our knifes.
This analogy converted to this discussion says that OK some, a few, may accidentally press ctrl-alt-backspace, but at the same time this particular key-stroke is very handy under those circumstances when X behaves badly. By removing the functionality you also adds another cost, the cost of extra data loss and other problems, i.e. with file systems, as user, when ctrl-alt-backspace does not work, finally will press the reset button or power cycle their computer. As I see it the benefit from having ctrl-alt-backspace is greater than the cost of a few loosing data.
I.e. how often this problem happens, people accidentally pressing ctrl-alt-backspace, is a valid question in this discussion. Does it happen more often than X crashes? Or less? This is an important parameter in the decision to keep, or not keep, the functionality.
While you certainly can present an argument based on that you have _never_ done this by accident, I'd like to point you at http://en.wikipedia.org/wiki/Argument_from_ignorance for reasons why this is, IMHO, not a sustainable platform to argue from.
Sigh! I just mentioned that *I* had never accidentally pressed this key combination during all those years that I have been using Linux, and therefore find it strange that this has become such a big issue that some wants to remove this functionality. It was *NOT* ment to be a platform to argue from, such an argument would be plainly stupid. If I during all my years have never accidentally pressed this combination, how often does it happen to others? I.e. for me X crashes have happened way more often then me accidentally pressing ctrl-alt-backspace. How is it for others?
English is not my first language, I hope you get what I am getting at this time...
As has been explained *extensively* already, you will still have the ability to do what you desire - but you will need to (trivially) enable this yourself in future. As also has been posited already, this has been blown way out of proportion, quite possibly based on erroneous assumptions.
What I will do or not do, or what the consequences will be for me personally is actually totally un-important for this discussion. I will just enable the functionality and live on as before.
The important thing is that this removal can create problems for new user of Linux. Why should we learn them to restart the whole machine when it is not needed, and even may create extra data loss, and even problems with the file system. I actually can not understand this. Should we not make life easy for the new ones?
/Lars
Lars E. Pettersson lars@homer.se http://www.sm6rpz.se/
-- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
If i'm right this is the second flame war about the DontZap "feature"... After reading roughly 100 posts (some quite interesting) in this flame war and some of the previous flame war i'm still not convinced that this new default for dontzap is the right one. The one -- and only -- reason that i see popping up all the time is emacs being stupid enough to have a ctr+alt+bs key binding in it as well.
I would really like to see the reasoning behind this new default for dontzap.. i wonder if there is any reasoning behind it at all.
The way i see it there are a few options to get past this issue. 1. (the red hat way) take it the way it is now or use another distribution 2. restore dontzap to it's "old" value.. it worked fine for many years so why change it if it isn't even broken 3. spam emacs for having a ctrl+alt+bs key binding and demand that they use some other key mix
Option 1 is what RH will probably do because they are to stubborn to revert it even though the majority of the people will vote for the old behavior if they dare to keep a vote. Option 2 and 3 is what has to be done if you ask me and as long as it hasn't been done redhat should patch those packages and contact the ones that need to be contacted.
About how the decision was made to change DontZap. Xorg is a big package and millions of users (hundreds of distributions) use it because it's just the "(un)official" standard. now it's very strange to see how ONE person (or that's the idea that i get behind this change) is somehow allowed to make a decision for millions of people! That just seems wrong to me. Meritocracy shouldn't be allowed in a project that can potentially have a huge impact and if that is allowed they should probably be forked to a group that does listen to the users (must be Democracy!!). Or all switch to DirectFB. Xorg should only be allowed to IMPROVE xorg, to support more stuff and not to delete features (or at least not features that everyone will notice in a negative way if deleted) unless approved by the biggest distributions that use Xorg. (and those distributions should all have polls with the same questions to ask the user base).
The way things are going now (with Xorg and Gnome!!) is still (just barely) going fine but i suspect that forks are going to arise if more is going to change that shouldn't be changed (if you ask the users).
I wonder what will happen if i make a application that listens to the CTR+ALT+DEL command and does something.. will gnome then get patched because another app is using the key mix?
And something different. Wouldn't it be a good idea to make a linux standards site decided by the users where the users decide how they like to have things. The majority then decides a standard instead of just a few people with a idea. With a site like that i don't mean the freedesktop standards!! Those are of course fine if the majority agrees with them.
Btw. Meritocracy seems dangerously close to ignorance and arrogance...
Just my opinion about DontZap. Mark.
On Fri, Apr 17, 2009 at 3:21 PM, Mark markg85@gmail.com wrote:
If i'm right this is the second flame war about the DontZap "feature"... After reading roughly 100 posts (some quite interesting) in this flame war and some of the previous flame war i'm still not convinced that this new default for dontzap is the right one.
Probably close to 500 posts or so now, but the people receiving kickbacks from Microsoft want you to think only one or two people are against this change.
The one -- and only -- reason that i see popping up all the time is emacs being stupid enough to have a ctr+alt+bs key binding in it as well.
There is an identical key binding in emacs which performs the same operation which is ctrl-alt-` or something like this, there is a post in one of the threads which mentions it.
I would really like to see the reasoning behind this new default for dontzap.. i wonder if there is any reasoning behind it at all.
This is the key. Not only will it be impossible to change, but the reasoning for the change is being kept a tightly held secret. Every single thread on every mailing list (including x.org!) has about a 9 to 1 ratio of people against making the change versus people in favor of it. The decision was made by less than a handful of people on IRC with no logs.
The way i see it there are a few options to get past this issue.
- (the red hat way) take it the way it is now or use another distribution
- restore dontzap to it's "old" value.. it worked fine for many years
so why change it if it isn't even broken 3. spam emacs for having a ctrl+alt+bs key binding and demand that they use some other key mix
Option 1 is what RH will probably do because they are to stubborn to revert it even though the majority of the people will vote for the old behavior if they dare to keep a vote.
There will never be a vote, because a vote would overwhelmingly decide to restore the original defaults. For some reason, this cannot be allowed to happen, and we have to be stuck with Microsoft like defaults.
Option 2 and 3 is what has to be done if you ask me and as long as it hasn't been done redhat should patch those packages and contact the ones that need to be contacted.
No, need to spam emacs, they already have made another key-binding, so it's not an issue. RH will not restore the defaults to sane values no matter how vociferously the community complains.
About how the decision was made to change DontZap. Xorg is a big package and millions of users (hundreds of distributions) use it because it's just the "(un)official" standard. now it's very strange to see how ONE person (or that's the idea that i get behind this change) is somehow allowed to make a decision for millions of people! That just seems wrong to me. Meritocracy shouldn't be allowed in a project that can potentially have a huge impact and if that is allowed they should probably be forked to a group that does listen to the users (must be Democracy!!). Or all switch to DirectFB. Xorg should only be allowed to IMPROVE xorg, to support more stuff and not to delete features (or at least not features that everyone will notice in a negative way if deleted) unless approved by the biggest distributions that use Xorg. (and those distributions should all have polls with the same questions to ask the user base).
It wont happen. Microsoft is probably spending less than ten grand paying off a few key developers to disable a feature by default. This is how the world works unfortunately. If the community wants to revert the default, they will have to come up with more money than Microsoft is paying them to disable it.
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
On Sat, Apr 18, 2009 at 10:18:45AM -0700, Christopher Stone wrote:
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
I got a t-shirt and some cake, but I was working on Ubuntu at the time. They don't love me as much now I'm working for Red Hat :(
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
You're going a little paranoid now i think. Also if you want to get it back the old way (like i do) you don't need to pay X.org 1 cent. Just fork it but think of the consequences if you do. X.org is huge.
I really wonder why this change isn't reverted yet.. there are so many opposed to this new dontzap default that some Xorg devs just don't seem or want to listen. O and btw i had my first Xorg dontzap issue as well. luckily a CTRL + ALT + F# followed by a killall appname solved it and x worked fine again.
On Sat, Apr 18, 2009 at 4:07 PM, Mark markg85@gmail.com wrote:
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
You're going a little paranoid now i think. Also if you want to get it back the old way (like i do) you don't need to pay X.org 1 cent. Just fork it but think of the consequences if you do. X.org is huge.
I really wonder why this change isn't reverted yet.. there are so many opposed to this new dontzap default that some Xorg devs just don't seem or want to listen. O and btw i had my first Xorg dontzap issue as well. luckily a CTRL + ALT + F# followed by a killall appname solved it and x worked fine again.
Increasingly the change sounds like a good idea just for the increasing motivation it will create to fix broken issues with X.
X shouldn't be freezing and the causes of X freezes can be fixed. Yet people will continue to mistakenly hit ctrl-alt-bs even if the software is made perfect. Yes, I think this is an unimportant change, but that doesn't mean that I don't also agree with it.
I'm sorry, people in this thread have talked about "windows mentality", but I think the real example of windows mentality is having a "crash" button on your keyboard because your system sometimes freezes.
Shall we add a wipe-dotfiles button next because sometimes they become corrupted and the desktop environment no longer lets you log in?
On Sat, Apr 18, 2009 at 5:43 PM, Gregory Maxwell gmaxwell@gmail.com wrote:
On Sat, Apr 18, 2009 at 4:07 PM, Mark markg85@gmail.com wrote:
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
You're going a little paranoid now i think. Also if you want to get it back the old way (like i do) you don't need to pay X.org 1 cent. Just fork it but think of the consequences if you do. X.org is huge.
I really wonder why this change isn't reverted yet.. there are so many opposed to this new dontzap default that some Xorg devs just don't seem or want to listen. O and btw i had my first Xorg dontzap issue as well. luckily a CTRL + ALT + F# followed by a killall appname solved it and x worked fine again.
Increasingly the change sounds like a good idea just for the increasing motivation it will create to fix broken issues with X.
X shouldn't be freezing and the causes of X freezes can be fixed. Yet people will continue to mistakenly hit ctrl-alt-bs even if the software is made perfect. Yes, I think this is an unimportant change, but that doesn't mean that I don't also agree with it.
I'm sorry, people in this thread have talked about "windows mentality", but I think the real example of windows mentality is having a "crash" button on your keyboard because your system sometimes freezes.
Shall we add a wipe-dotfiles button next because sometimes they become corrupted and the desktop environment no longer lets you log in?
Windows doesn't have a crash button. I'm not sure why you think having a button that will aid you in the event of a crash is a bad thing. However, I think you should have the choice to remove it, while is there by default for everyone else. Because in the event of a crash, you can choose not to use it, but I can't install it during the crash just so that I can then use it.
And I'm not sure what would be wrong with a tool that intelligently moved dot files if it allowed a non expert user to login and get assistance.
Gregory Maxwell wrote:
Shall we add a wipe-dotfiles button next because sometimes they become corrupted and the desktop environment no longer lets you log in?
Given how many people suggest to delete the entire .kde directory the first time somebody experiences even the slightest problem with KDE, you might actually find interested people...
(Hint: DO NOT delete your entire .kde directory if you have KDE issues. If you think it's a configuration issue, rename (don't delete!) the specific config files most likely to cause issues. In case of Plasma, that's plasmarc and plasma-appletsrc. In case of an application, that's the rc with the application's name in it. The same also goes for other (non-KDE) applications, always keep a backup (i.e. rename, don't delete) and only rename the files actually related to your problem.)
Kevin Kofler
* Christopher Stone chris.stone@gmail.com [20090418 22:47]: [snip]
It wont happen. Microsoft is probably spending less than ten grand paying off a few key developers to disable a feature by default. This is how the world works unfortunately. If the community wants to revert the default, they will have to come up with more money than Microsoft is paying them to disable it.
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
This is getting dargerously close to libel, and certainly is in the realm of slander. Both are criminal offenses. Please have a think about the ramifications and the consequenses of what you say. Just because it is a mailing list, you are not somehow excluded from the law.
Anders Rayner-Karlsson wrote:
- Christopher Stone chris.stone@gmail.com [20090418 22:47]:
[snip]
It wont happen. Microsoft is probably spending less than ten grand paying off a few key developers to disable a feature by default. This is how the world works unfortunately. If the community wants to revert the default, they will have to come up with more money than Microsoft is paying them to disable it.
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
This is getting dargerously close to libel, and certainly is in the realm of slander. Both are criminal offenses. Please have a think about the ramifications and the consequenses of what you say. Just because it is a mailing list, you are not somehow excluded from the law.
thank you mr. laywer for pointing that out to all of us, it was funny to read tho =)
On Sat, 2009-04-18 at 10:18 -0700, Christopher Stone wrote:
So, how much is Microsoft paying you guys? Anyone who has been posting to this thread who has been receiving kickback money, please let us know how much you are getting.
I didn't get anything :(.
On an entirely unrelated note, I hereby announce that tomorrow I'm retiring and going to live on my new private island in the Caribbean.
On Fri, Apr 17, 2009 at 11:29:02PM +0200, Lars E. Pettersson wrote:
On 04/17/2009 10:32 PM, Anders Rayner-Karlsson wrote:
- Lars E. Pettersson lars@homer.se [20090417 21:37]:
How often does people actually accidentally press ctrl-alt-backspace? I have *never* done it.
By same argument, taking a rather tongue in cheek attitude, how often does countries with nuclear ICBM's launch them by accident?
It is not argument as such. If the decision has been made to remove this functionality, the decision has to be based on something. In this case it seem to be based on that people accidentally can press this key-stroke combination and loose data. If this is the case, it is important to know how often this actually happens.
If you look at real life. In our kitchens we have knifes. You can accidentally injure yourself quite badly with a knife. Should we just because of this zap all knife blades to make them safe?
no, but most people keep them in a knife block and not in the same drawer as ladles, stirring spoons and whatnot.
I.e. we have made a decision that knifes are good to have, accidents do happen, but the benefit from having knifes are greater than the consequences of the accidents that can happen. So we keep our knifes.
and by moving the kives to a separate place we may have saved a few fingers over the history.
Also, it's important here that we're not talking about the butter knives where you can't hurt yourself unless you swallow them as a three-year old (like VT-switching, easy to recover). we're talking about the really sharp cooking knives, the onces where you don't know you cut yourself until you hear the scraping on the bone.
also, I think this knife analogy is stretching things a bit, but I like cooking (and it beats debugging grabs anytime), so we can continue :)
This analogy converted to this discussion says that OK some, a few, may accidentally press ctrl-alt-backspace, but at the same time this particular key-stroke is very handy under those circumstances when X behaves badly. By removing the functionality you also adds another cost, the cost of extra data loss and other problems, i.e. with file systems, as user, when ctrl-alt-backspace does not work, finally will press the reset button or power cycle their computer. As I see it the benefit from having ctrl-alt-backspace is greater than the cost of a few loosing data.
I.e. how often this problem happens, people accidentally pressing ctrl-alt-backspace, is a valid question in this discussion. Does it happen more often than X crashes? Or less? This is an important parameter in the decision to keep, or not keep, the functionality.
While you certainly can present an argument based on that you have _never_ done this by accident, I'd like to point you at http://en.wikipedia.org/wiki/Argument_from_ignorance for reasons why this is, IMHO, not a sustainable platform to argue from.
Sigh! I just mentioned that *I* had never accidentally pressed this key combination during all those years that I have been using Linux, and therefore find it strange that this has become such a big issue that some wants to remove this functionality. It was *NOT* ment to be a platform to argue from, such an argument would be plainly stupid. If I during all my years have never accidentally pressed this combination, how often does it happen to others? I.e. for me X crashes have happened way more often then me accidentally pressing ctrl-alt-backspace. How is it for others?
fwiw, about once a week or so for me (I have a lot of shortcuts on ctrl+alt).
btw, it's quite interesting to look at the key press/release events as you're typing you'll notice that often you're typing multiple letter before releasing the first.
Cheers, Peter
Lars E. Pettersson wrote:
The important thing is that this removal can create problems for new user of Linux. Why should we learn them to restart the whole machine when it is not needed, and even may create extra data loss, and even problems with the file system.
It occurs to me... has anyone considered that we are teaching users that GNU/Linux is just as bad as Windows with this change?
It used to be: no, no, this is *GNU/Linux*! An X crash isn't the total system failure it is in Windows, just hit c-a-bs and you're instantly back at your login.
Now we're making X problems into: yeah, you have to reboot, wait for the system to come back... oh, and wait for it to run scandi^Wfsck (and better pray the FS didn't get hosed)...
So much for GNU/Linux being better than Windows.
On Mon, Apr 20, 2009 at 11:20:10AM -0500, Matthew Woehlke wrote:
Lars E. Pettersson wrote:
The important thing is that this removal can create problems for new user of Linux. Why should we learn them to restart the whole machine when it is not needed, and even may create extra data loss, and even problems with the file system.
It occurs to me... has anyone considered that we are teaching users that GNU/Linux is just as bad as Windows with this change?
It used to be: no, no, this is *GNU/Linux*! An X crash isn't the total system failure it is in Windows, just hit c-a-bs and you're instantly back at your login.
Now we're making X problems into: yeah, you have to reboot, wait for the system to come back... oh, and wait for it to run scandi^Wfsck (and better pray the FS didn't get hosed)...
So much for GNU/Linux being better than Windows.
The world isn't black or white. Since the invention of light we have greyscales too (I believe this even was a zero-day update, according to the book of Genesis).
So their X crashed. Tell them that this can happen and if they tick the checkbox next time they can hit c-a-b to avoid rebooting.
They know now that the shortcut exists, they know they have enabled it explictly, they know how to disable it if they don't want it. They can even select the keyboard sequence they want (once we actually have a few to select from).
You have tought them something, and maybe they feel better for learning a bit of how their computer works, and that they can with little effort configure the computer to suit their needs better.
Then again, maybe they don't want to learn this. Maybe they don't want to hit c-a-b because they don't understand or care that crashing the windowing system != crashing the operating system (and hitting the power button is something they're comfortable with).
Or maybe somewhere in the middle.
Cheers, Peter
Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead?
Abusing the Fedora as Guinea pigs, as they've done many times before.
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote:
Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead?
Abusing the Fedora as Guinea pigs, as they've done many times before.
By doing the change upstream?
On Friday 17 April 2009 23:17:44 Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote: By doing the change upstream?
And this may also be the time to drop this "upstream" point to. There is no KMS upstream, but we have it here. nv is default upstream. Upstream defaults aren't that important. If there were no per-distro defaults, there would be only one distro - the one with upstream defaults.
Suren
On Fri, Apr 17, 2009 at 3:41 PM, Suren Karapetyan wrote:
On Friday 17 April 2009 23:17:44 Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote: By doing the change upstream?
And this may also be the time to drop this "upstream" point to. There is no KMS upstream, but we have it here. nv is default upstream. Upstream defaults aren't that important. If there were no per-distro defaults, there would be only one distro - the one with upstream defaults.
Suren
That is such a strong argument. But it is so true.
e.g. I don't think we have many packages in Fedora that uses /usr/local as their prefix, which is a very common upstream default.
We first lost CTRL+ALT-F7, now CTRL+ALT+BACKSPACE. What is next? The "kill" command? I think that is left to the imagination of our upstreams.
Orcan
On Fri, 2009-04-17 at 16:18 -0400, Orcan Ogetbil wrote:
e.g. I don't think we have many packages in Fedora that uses /usr/local as their prefix, which is a very common upstream default.
That's not a valid example. This discrepancy is exactly what the defaults are intended to preserve. The whole point is that stuff sourced from your distribution and stuff you build yourself from source code wind up in two different places so you can easily tell one from the other (and, in the worst case, wipe out /usr/local and know you're back at a stock system).
This is not a useful case to argue that *all* upstream defaults shouldn't be considered in setting Fedora defaults.
On Fri, 2009-04-17 at 16:18 -0400, Orcan Ogetbil wrote:
We first lost CTRL+ALT-F7, now CTRL+ALT+BACKSPACE. What is next? The "kill" command? I think that is left to the imagination of our upstreams.
They may take our gettys, they may take our zapper, but they'll never take... OUR FREEDOM!
On Fri, Apr 17, 2009 at 07:17:44PM +0100, Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote:
Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead?
Abusing the Fedora as Guinea pigs, as they've done many times before.
By doing the change upstream?
The name sounds familiar...are you by any chance the same Matthew Garrett that ran for Debian Project leader and then went on to Ubuntu?
On Fri, Apr 17, 2009 at 10:13:45PM +0100, Dan wrote:
On Fri, Apr 17, 2009 at 07:17:44PM +0100, Matthew Garrett wrote:
By doing the change upstream?
The name sounds familiar...are you by any chance the same Matthew Garrett that ran for Debian Project leader and then went on to Ubuntu?
Yes.
Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote:
Christopher Stone wrote:
2009/4/16 Jesse Keating jkeating@redhat.com:
On Fri, 2009-04-17 at 01:37 +0500, Suren Karapetyan wrote:
You will not find anyone @RH who will agree that previews behavior was better
On the contrary, you'll likely find many people who work for Red Hat that would prefer the old behavior, they're just not wasting their time by arguing about it on a distribution mailing list.
So what are they doing instead?
Abusing the Fedora as Guinea pigs, as they've done many times before.
By doing the change upstream?
When "upstreams" are identical to the fedora maintainers, the "upstream argument" is moot. They are in a position to commit any stupidity they want upstream and label it "upstream decsion".
If then, management and QA/control is dominated by "one governing state party" (Like in Fedora), all control and QA is effectively non-existent.
In other words, it's simply a matter of powers.
On Mon, Apr 20, 2009 at 07:56:34AM +0200, Ralf Corsepius wrote:
Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote:
Abusing the Fedora as Guinea pigs, as they've done many times before.
By doing the change upstream?
When "upstreams" are identical to the fedora maintainers, the "upstream argument" is moot. They are in a position to commit any stupidity they want upstream and label it "upstream decsion".
I think that basically all the Fedora X maintainers are upstream - however, not all of upstream are Fedora X maintainers.
On Mon, Apr 20, 2009 at 10:44 AM, Matthew Garrett mjg@redhat.com wrote:
On Mon, Apr 20, 2009 at 07:56:34AM +0200, Ralf Corsepius wrote:
Matthew Garrett wrote:
On Fri, Apr 17, 2009 at 07:02:20PM +0200, Ralf Corsepius wrote:
Abusing the Fedora as Guinea pigs, as they've done many times before.
By doing the change upstream?
When "upstreams" are identical to the fedora maintainers, the "upstream argument" is moot. They are in a position to commit any stupidity they want upstream and label it "upstream decsion".
I think that basically all the Fedora X maintainers are upstream - however, not all of upstream are Fedora X maintainers.
Just one question.
How can fedora ever be objective about an issue like this if: (quote) "all the Fedora X maintainers are upstream"
.... ? i'm interested in your reply on this one. (and a reply from all RH/fedora ppl in this thread btw)
On Mon, Apr 20, 2009 at 01:16:34PM +0200, Mark wrote:
On Mon, Apr 20, 2009 at 10:44 AM, Matthew Garrett mjg@redhat.com wrote:
I think that basically all the Fedora X maintainers are upstream - however, not all of upstream are Fedora X maintainers.
Just one question.
How can fedora ever be objective about an issue like this if: (quote) "all the Fedora X maintainers are upstream"
.... ? i'm interested in your reply on this one. (and a reply from all RH/fedora ppl in this thread btw)
What do you mean by objective? It's the role of package maintainers to do their best at ensuring that the software works well in Fedora and is as bug free as possible. It's not the role of package mantainers to make changes that they disagree with, no matter how many users ask for them.
On 04/16/2009 11:44 AM, Jesse Keating wrote:
On Thu, 2009-04-16 at 19:33 +0200, Ralf Corsepius wrote:
If they feel sufficiently pissed, they will quit and leave Fedora alone.
People threaten that all the time, and yet they never seem to follow through.
Ehmm... there was a time when Gnome's brain-dead file picker and swapping of Cancel/Ok button orders did it for me -- I switched to KDE permanently. There's a limit, where you just have enough. -> Some people follow through...
Jesse Keating wrote:
On Thu, 2009-04-16 at 19:33 +0200, Ralf Corsepius wrote:
If they feel sufficiently pissed, they will quit and leave Fedora alone.
People threaten that all the time, and yet they never seem to follow through.
Yes, * as a developer, I like Fedora * as sysadmin, I find Fedora has derailed and become increasingly unusable. * as desktop user, it doesn't need much and I can't avoid "calling Fedora's desktop" crap. * as Fedora project participant, Fedora's leadership and some @RH's behavior gradually drives me off.
... the problem is lack of better alternatives.
.. and me not having giving up hope on Fedora.
Ralf
On Wed, 15 Apr 2009, Adam Williamson wrote:
way. All I've ever said is that it isn't a big deal.
"17000 threads disagree with that opinion"
Put it this way - the time lost by the possible drawbacks of *either* approach is extremely unlikely *ever* to add up to the amount of time and energy smart people have wasted in the seventeen thousand threads about this.
Are you going to wait for another 17000 threads to be wasted before you think it is worth to reconsider?
Indeed, I could have fixed a reported ldns bug in the time this happened. I blame it on Xorg and Fedora that the bug is still there :P
"possible drawbacks" of the suse solution are not there. See my previous email on the 10 years of lack of requesting a way to disable ctrl-alt-bksp.
People involved with Fedora are passionate. That's why they are involved. You can't have the cake and eat it too. The fact that I maintain 20 packages is the same reason I wrote 20 emails about ctrl-alt-bksp. I want it to be done right, and where possible to suppot multiple right ways.
There is little effort and no harm involved in the suse patch, and there is no reason not to accomodate these people.
My last patches to the nsd package was to acocmodate chroot. I despise chroot. There is no sane reason for it compared to selinux. But there was little effort involved to accomodate this person, so why not. And now fedora is better for people who like and who dislike chroot. That is how contributors and users should interact. Note the lack of 17000 emails on the nsd chroot modifications.
Paul
On Thu, 2009-04-16 at 14:30 -0400, Paul Wouters wrote:
On Wed, 15 Apr 2009, Adam Williamson wrote:
In a corporate environment the network will be managed by a sysadmin who will easily be able to change the default.
I am sorry, I must have missed the new kickstart/anaconda option......
Someone posted a very simple kickstart way to do this earlier in the thread.
(waaaay earlier.)
On 04/15/2009 08:59 PM, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why? There is no need to reboot the machine just because an application, X in this case, causes problem. You only have to kill X, and restart it, this is just what ctrl-alt-backspace have done in an excellent way for years. I frankly can not see why this feature should go, just because some few accidentally have pressed just that key sequence.
A better way to fix this might be to have a dialog. You presses ctrl-alt-backspace, a dialog comes up saying, I will restart X in yy seconds, with a cancel button. If X is toast, the alarm will the restart X. If this is feasible, would this alternative not be a better one than removing a feature that many likes and want to stay.
By the way, I can not recall seeing anyone complaining about that ctrl-alt-backspace have caused them problems. Could someone please give some links to discussions where this actually have happened.
/Lars
On Thu, 2009-04-16 at 10:35 +0200, Lars E. Pettersson wrote:
On 04/15/2009 08:59 PM, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why?
I don't *care*. Stop trying to argue with me. My only position on this is it really doesn't matter as much as you all seem to think it does. I don't care if it stays the way it is or gets patched, I just wish everyone would quit wasting time bikeshedding about it.
On Thu, Apr 16, 2009 at 8:34 AM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2009-04-16 at 10:35 +0200, Lars E. Pettersson wrote:
On 04/15/2009 08:59 PM, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why?
I don't *care*. Stop trying to argue with me. My only position on this is it really doesn't matter as much as you all seem to think it does. I don't care if it stays the way it is or gets patched, I just wish everyone would quit wasting time bikeshedding about it.
Funny how on every single thread about this issue people are either vehemently against the change or they don't care.
I have yet to find someone who actually thinks changing the defaults is a good idea.
And BTW, your comment about the amount of time wasted in this thread vs the time that will be wasted by this change is hilarious, it proves you have no clue.
I have yet to find someone who actually thinks changing the defaults is a good idea.
I think that's a good idea : 1. I used it once: accidentally 2. if I don't know that the shortcut exists, then there should be no way for me to do something bad accidentally 3. if I know the shortcut exists, then it should be easy for me to make it work
Previously, only the third item was provided. This change introduces the second one too (yes, the third one is still provided: it _is_ easy to make it work, before the change it worked out of box).
BTW, the main reason you actually see more people complaining about the change is because the people that this change is made for have absolutely no idea that this shortcut exists and that it is being disabled by default. How could they be happy about something they don't know ?
And BTW, your comment about the amount of time wasted in this thread vs the time that will be wasted by this change is hilarious, it proves you have no clue.
So you're saying that we should waste even more time in this thread so that the amount of time wasted here is finally superior to the time wasted by this change ?
Yeah, that's gonna be fun :)
----------
Mathieu Bridon (bochecha)
On 04/16/2009 05:34 PM, Adam Williamson wrote:
On Thu, 2009-04-16 at 10:35 +0200, Lars E. Pettersson wrote:
On 04/15/2009 08:59 PM, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why?
I don't *care*. Stop trying to argue with me. My only position on this is it really doesn't matter as much as you all seem to think it does. I don't care if it stays the way it is or gets patched, I just wish everyone would quit wasting time bikeshedding about it.
OK, so we should just sit here doing nothing when we see a change that *we* consider bad? Nah, better to speak up, I think.
You have the notion that this does not matter, but I, and apparently quite a few others, think it does.
Yes, they can just reboot, pressing ctrl-alt-del instead. But wait, that is another shortcut that people accidentally can press. Should we "zap" that one also? If not, why should we zap ctrl-alt-backspace, but not ctrl-alt-del? And if we zap both, they have to press the reset button, or switch the computer off and on again, a process that might cause loss of data, and even file system problems.
So my thinking is that this really matters.
Sure, a user might press ctrl-alt-backspace accidentally, but me feeling is that this is *not* such a recurrent event that we should remove a very useful function/keystroke from the system.
/Lars
On Thu, 2009-04-16 at 20:29 +0200, Lars E. Pettersson wrote:
On 04/16/2009 05:34 PM, Adam Williamson wrote:
On Thu, 2009-04-16 at 10:35 +0200, Lars E. Pettersson wrote:
On 04/15/2009 08:59 PM, Adam Williamson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why?
I don't *care*. Stop trying to argue with me. My only position on this is it really doesn't matter as much as you all seem to think it does. I don't care if it stays the way it is or gets patched, I just wish everyone would quit wasting time bikeshedding about it.
OK, so we should just sit here doing nothing when we see a change that *we* consider bad? Nah, better to speak up, I think.
No, I think after the twenty-three thousandth thread on the fiftieth different mailing list rehearses the same three arguments *again*, people should cut their damn losses and do something productive instead.
You have the notion that this does not matter, but I, and apparently quite a few others, think it does.
I have the notion that it doesn't matter anywhere near enough to be worth this amount of grief. As I said: grow a sense of proportion.
On 04/16/2009 08:41 PM, Adam Williamson wrote:
On Thu, 2009-04-16 at 20:29 +0200, Lars E. Pettersson wrote:
OK, so we should just sit here doing nothing when we see a change that *we* consider bad? Nah, better to speak up, I think.
No, I think after the twenty-three thousandth thread on the fiftieth different mailing list rehearses the same three arguments *again*, people should cut their damn losses and do something productive instead.
Perhaps it is time listen to those "twenty-three thousandth thread on the fiftieth different mailing list". People whine for a reason, not just for the fun of it.
You have the notion that this does not matter, but I, and apparently quite a few others, think it does.
I have the notion that it doesn't matter anywhere near enough to be worth this amount of grief. As I said: grow a sense of proportion.
Speaking for myself, this keystroke is one of the things in the Linux world I really like. A very easy way to fix a problem that sadly still happens way to often, a stuck X-session. Simple and elegant. So for *me* it matters, and apparently so also for a bunch of others.
Proportions? Well, look at all those threads you mentioned, and you see the proportions...
/Lars
On 04/17/2009 08:45 PM, Kevin Kofler wrote:
Lars E. Pettersson wrote:
Yes, they can just reboot, pressing ctrl-alt-del instead. But wait, that is another shortcut that people accidentally can press. Should we "zap" that one also?
X11 already eats Ctrl+Alt+Del.
Well, it isn't removed, as ctrl-alt-backspace is proposed to be (it was that I meant wit "zap"), it shows a dialog.
Not sure what happens if X have crashed and one presses ctrl-alt-del though, perhaps someone else who have experienced this could tell us? Will it automatically shut down the system after 60 seconds, even though X is not responding?
So in a situation where X have crashed, and neither ctrl-alt-del (if that is the case) nor ctrl-alt-backspace works, one have to press the reset button or power cycle the computer, a, in my humble opinion, very bad and brutal way of solving a problem that was elegantly solved with ctrl-alt-backspace.
/Lars
On Fri, Apr 17, 2009 at 09:05:37PM +0200, Lars E. Pettersson wrote:
On 04/17/2009 08:45 PM, Kevin Kofler wrote:
X11 already eats Ctrl+Alt+Del.
Well, it isn't removed, as ctrl-alt-backspace is proposed to be (it was that I meant wit "zap"), it shows a dialog.
ctrl+alt+del has no special status within X. It's simply a keybinding that can be grabbed by the desktop and used to display something. ctrl+alt+bs now has an identical status.
Not sure what happens if X have crashed and one presses ctrl-alt-del though, perhaps someone else who have experienced this could tell us? Will it automatically shut down the system after 60 seconds, even though X is not responding?
That depends on the state the system is in. If X has crashed and exited, ctrl+alt+del will reset the system as if you were at the console. If X has stopped drawing on screen but is still processing the event loop, a dialog will pop up and shut the machine down after 60 seconds. If X isn't responding to anything, ctrl+alt+del will do nothing.
So in a situation where X have crashed, and neither ctrl-alt-del (if that is the case) nor ctrl-alt-backspace works, one have to press the reset button or power cycle the computer, a, in my humble opinion, very bad and brutal way of solving a problem that was elegantly solved with ctrl-alt-backspace.
I don't /think/ (but will defer to people more familiar with this part of the server) that it's likely that you'll end up in situations where ctrl+alt+del won't work but ctrl+alt+bs will.
On 04/17/2009 09:25 PM, Matthew Garrett wrote:
ctrl+alt+del has no special status within X. It's simply a keybinding that can be grabbed by the desktop and used to display something. ctrl+alt+bs now has an identical status.
OK, but it (ctrl-alt-backspace) is not supposed to be grabbed by the desktop to display something (by default), is that correct? At least as things are at the moment? Have I gotten this right?
/Lars
On Fri, Apr 17, 2009 at 09:41:54PM +0200, Lars E. Pettersson wrote:
On 04/17/2009 09:25 PM, Matthew Garrett wrote:
ctrl+alt+del has no special status within X. It's simply a keybinding that can be grabbed by the desktop and used to display something. ctrl+alt+bs now has an identical status.
OK, but it (ctrl-alt-backspace) is not supposed to be grabbed by the desktop to display something (by default), is that correct? At least as things are at the moment? Have I gotten this right?
Currently ctrl+alt+backspace will not be grabbed by anything. If you're running gnome you can bind it through the keyboard shortcuts - this is the same place as ctrl+dlt+delete is bound.
On Fri, Apr 17, 2009 at 09:41:54PM +0200, Lars E. Pettersson wrote:
On 04/17/2009 09:25 PM, Matthew Garrett wrote:
ctrl+alt+del has no special status within X. It's simply a keybinding that can be grabbed by the desktop and used to display something. ctrl+alt+bs now has an identical status.
OK, but it (ctrl-alt-backspace) is not supposed to be grabbed by the desktop to display something (by default), is that correct? At least as things are at the moment? Have I gotten this right?
Clients can grab the shortcut. If zapping is enabled though, the server zaps before the client gets the event and regardless of an client holding the grab.
So there's no reason why the DE couldn't pop up a dialog box in the order "This shortcut zap the server. Do you want to permanently enable zapping?" I guess this would be handy for those that prefer zapping instead of hitting log out (and I am quite guilty of said behaviour myself)
Maybe we could make this more user-friendly by adding a talking paperclip too.
Cheers, Peter
On Thu, 16 Apr 2009, Lars E. Pettersson wrote:
Haha :) No, it's just that, as far as I can see, the impact on newbies is that we tell them to reboot instead of doing ctrl-alt-backspace.
Yes, but why? There is no need to reboot the machine just because an application, X in this case, causes problem. You only have to kill X, and restart it, this is just what ctrl-alt-backspace have done in an excellent way for years. I frankly can not see why this feature should go, just because some few accidentally have pressed just that key sequence.
And in fact, anyone claiming that this is a good solution has a very bad case of Windows mentality. Rebooting a machine is not nice. It might affect other people. What if someone is running X to use virt-manager, and his X session is stuck? Rebooting would be an awful solution. And we will have many people who atthat point will be hitting ctrl-alt-bksp to see nothing happen. Sure, they can try and ssh in, and sure they should have had sysctl enabled, but this is really stretching the gains of the fix versus the costs of the fix.
Every release (except f11) people have complained about the sbin directories only being in root's path. Ever release people have complained about the -y option for autofsck. every release some people have complained about the mv/cp/rm aliases for root. EVery release people have complained about the initrd's on dom0 not containing xenblock so the same initrd cannot be used to boot guests.
But until ctrl-alt-bksp was disabled, I've never seen a flameware on accidental key presses killing X on the fedora lists. And yes, it has happened to me too, and I once confirmed ctrl-bksp would kill my X session without the alt key, with no stuck alt keys, but it has happend to me twice in the last 10 years. X locking up however, occurs to me somewhere at least once every month or two. And I don't want to walk to another machine to figure out the dhcp ip i got, hope i have ssh enabled, just so I don't have to trash my machine.
There are solutions to please all parties involved, the suse patch. Not doing that is simply ignoring user demand, and will contribute to a lesser user experience.
Paul
On 2009/04/13 13:32 (GMT-0530) Rahul Sundaram composed:
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally.
I prefer to think no three key combination can be struck by accident, but rather only by incompetence of keyboard design and/or user. There's a limit to how far anyone or anything can go to prevent "accidents".
As long as standard keyboards continue to keep some considerable distance between any two of those three keys, Ctrl-Alt-BS should by default continue to function as it did last century.
2009/4/13 Felix Miata mrmazda@ij.net:
On 2009/04/13 13:32 (GMT-0530) Rahul Sundaram composed:
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally.
I prefer to think no three key combination can be struck by accident, but rather only by incompetence of keyboard design and/or user. There's a limit to how far anyone or anything can go to prevent "accidents".
As long as standard keyboards continue to keep some considerable distance between any two of those three keys, Ctrl-Alt-BS should by default continue to function as it did last century.
Yes please.
If the only reason we are disabling c-a-b is due to the corner case of:
"People who somehow manage to - quite incredibly - push this combo by accident"
then someone is getting patch-happy and this needs a quick revert. The number of users who hit this by accident will always be less than those that require it to restart X. It seems clear that the majority of users require this functionality and that disabling it is not the best way of improving the X server.
As someone previously put, this increases the number of post-install operations required - qv. nautilus spatial behaviour. Please don't bring back the kickstart counter for this as asking users to configure kickstarts is not sane.
I'm disappointed that despite numerous user requests this has not been changed back. I'm well aware of the policy to follow upstream as closely as possible but I believe user experience should override this.
Felix Miata wrote:
On 2009/04/13 13:32 (GMT-0530) Rahul Sundaram composed:
Regardless of which default setting you prefer, the clear evidence is that people do end up doing it accidentally.
I prefer to think no three key combination can be struck by accident, but rather only by incompetence of keyboard design and/or user.
I think, the standard situation to "accidentally" hit "ctrl-alt-bs" is when people start hammering their keyboard with random hits in situations of "stuck systems"/"systems running wild" and when actually wanting to press "clrl-alt-del".
Makes me wonder, if they start "hammering their keyboard", because x has already raped "clt-alt-del" ;)
There's a limit to how far anyone or anything can go to prevent "accidents".
Yes, you'll alway find somebody who will try to dry a cat in microwave or lets his kids play with explosives.
Ralf
On Mon, 2009-04-13 at 02:50 -0400, Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 2:41 AM, Rahul Sundaram wrote:
Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Many people in thread here and other places like the freedesktop mailing list and in http://lwn.net/Articles/327141/ have claimed to accidentally tripped this key combo. It is not that hard.
Rahul
I really don't know what to say. It feels like the probability of burning my hair on the stove while sticking my foot in the freezer.
Anyway, I opened a poll at fedoraforum about this: http://www.fedoraforum.org/forum/showthread.php?t=219559
I do wonder what will come out. Of course this will not reflect the true statistics of all Fedora userbase. It never does. But it might give an idea.
I noticed one of the first comments is someone who missed one of the points of Peters work. At GDM you can still C-A-B all you want, the user preference only takes control when you login. So in those cases where you have gdm in the wrong resolution etc you can still quickly do it.
Its unlikely anyone minds losing their username they typed at the gdm prompt.
Dave.
Orcan
Dave Airlie wrote:
I noticed one of the first comments is someone who missed one of the points of Peters work. At GDM you can still C-A-B all you want, the user preference only takes control when you login.
Uh, if the default is to have it disabled (and it currently is), GDM would have to explicitly enable it for that to work. As would KDM and all the other login managers. I don't see this actually implemented in Rawhide's GDM. The option would have to default to enabled for things to just work.
Kevin Kofler
I wrote:
Uh, if the default is to have it disabled (and it currently is), GDM would have to explicitly enable it for that to work. As would KDM and all the other login managers. I don't see this actually implemented in Rawhide's GDM. The option would have to default to enabled for things to just work.
Hmmm, reading Peter Hutterer's reply further down the thread, it looks like the default for the xkb setting _is_ "on", it's just gnome-keyboard-properties which turns it off by default. This makes sense.
Kevin Kofler
On Wed, Apr 15, 2009 at 10:01:13AM +0200, Kevin Kofler wrote:
I wrote:
Uh, if the default is to have it disabled (and it currently is), GDM would have to explicitly enable it for that to work. As would KDM and all the other login managers. I don't see this actually implemented in Rawhide's GDM. The option would have to default to enabled for things to just work.
Hmmm, reading Peter Hutterer's reply further down the thread, it looks like the default for the xkb setting _is_ "on", it's just gnome-keyboard-properties which turns it off by default. This makes sense.
A bit of clarification here - just in case: The default keymap does not include the symbols anymore. fedora-setup-keyboard (which also sets the layout based on /etc/sysconfig/keyboard) enables the option at X server startup. hence it is on in gdm (or anytime after startup).
When gdm hands over to gnome, gnome applies the keyboard settings as stored in gconf, and unless the option is set in gnome, it won't be turned on again. This uses the same mechanism that allows us to start with whatever is in sysconfig/keyboard at gdm and then apply a custom layout once the user logs in. The visible effect is that it's on in gdm and off in the session.
For those not running gnome-settings-daemon it remains on until the respective desktop environment changes to a layout without setting the XKB options.
Cheers, Peter
Peter Hutterer wrote:
For those not running gnome-settings-daemon it remains on until the respective desktop environment changes to a layout without setting the XKB options.
OK.
KDE does not set the keyboard layout at all by default, so I presume this means KDE will keep the Ctrl+Alt+BkSp on by default. For those users who enabled keyboard layouts in KDE, what setting they'll get depends on whether they check the xkb option (and users upgrading from older releases will have it unchecked).
Kevin Kofler
"OO" == Orcan Ogetbil oget.fedora@gmail.com writes:
OO> I really don't know what to say. It feels like the probability of OO> burning my hair on the stove while sticking my foot in the OO> freezer.
Erm, I can press Ctrl-Alt-Backspace using just one thumb. Please don't assume that everyone's keyboard looks like yours. (That said, I've still never triggered it accidentally.)
- J<
Rahul Sundaram wrote:
Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Many people in thread here and other places like the freedesktop mailing list and in http://lwn.net/Articles/327141/ have claimed to accidentally tripped this key combo.
They have learnd their lesson and won't do it again.
People will have to understand that "Linux is not Windoze" and that it's naive to expect all OSes to behave the same rsp to adopt the design flaws of other OSes.
It'll be the same people who will complain, when they find their X system misbehaving and they are able to cope with the situation in reasonable manners because Red Hat's xorg maintainers have broken "ctl-alt-bs".
Those people who now are complaining, are those who know how helpful "clt-alt-bs" is in certain situations.
It is not that hard.
It's not that hard to press "ctl-alt-del", "ctl-alt-prsrcn", "ctl-S" and many other keyboard short cuts, either.
Rahul Sundaram wrote:
Orcan Ogetbil wrote:
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it."
Does this imply that people can press CTRL+ALT+Backspace accidentally??
I'm speechless.
Many people in thread here and other places like the freedesktop mailing list and in http://lwn.net/Articles/327141/ have claimed to accidentally tripped this key combo. It is not that hard.
Rahul
I think it's great that this feature can now be turned on or off via the GUI rather than having to manually edit xorg.conf. I also agree with the general principle that default settings should normally be aimed at beginning users, since it's much easier for experienced users to change the settings to their liking. So I think the pertinent question is, which users are typically more experienced, the ones who want to use Ctrl-Alt-Backspace, or the ones who regularly use complicated keyboard combinations that can lead to Ctrl-Alt-Backspace being hit accidentally? (This isn't a rhetorical question, I really don't know, though I'm in the former category.)
On 04/13/2009 04:31 PM, Andre Robatino wrote:
I think it's great that this feature can now be turned on or off via the GUI rather than having to manually edit xorg.conf. I also agree with the general principle that default settings should normally be aimed at beginning users, since it's much easier for experienced users to change the settings to their liking.
In that case I vote that ctrl-alt-backspace should be on by default.
The reason? Well, we do not want to learn new users that in case of X-windows trouble, reboot the whole machine (which in most cases is not needed). Instead, we instruct them to use ctrl-alt-backspace, as we have done for years now, to kill just the X-window.
If anyone have problem with ctrl-alt-backspace, he/she could then change the default to his/her liking. But the default ought to be on.
/Lars (I have been using emacs since 1987, and never accidentally pressed ctrl-alt-backspace...)
On Mon, Apr 13, 2009 at 12:58 AM, Christopher Stone chris.stone@gmail.com wrote: [snip]
Good thing we will never see a vote on this, because we have to keep the x.org egos inflated. They could never possibly make a brain dead lemming like decision.
You should take a moment to consider the tone of your message. To me it's coming across as unnecessarily abusive, I doubt I'm alone.
Have some perspective— we're talking about a keypress that shuts down the X-server. There is room here for people who are intelligent, knowledgeable, and well intentioned to disagree.
On Mon, Apr 13, 2009 at 2:29 AM, Orcan Ogetbil oget.fedora@gmail.com wrote:
Another quote: "Those who want to use the computer but not have to know about it's internals should not be able to accidentally trigger it." Does this imply that people can press CTRL+ALT+Backspace accidentally?? I'm speechless.
I have, quite a few times. I've always slapped my forehead after doing it and went on with life. I just considered it my own error and it never occurred to me to disable it. I was somewhat surprised to hear that some other experienced users had *not* triggered it accidentally at least once.
On Sun, Apr 12, 2009 at 09:58:50PM -0700, Christopher Stone wrote:
On Sun, Apr 12, 2009 at 8:18 PM, Tom "spot" Callaway tcallawa@redhat.com wrote:
Excuse me Tom, this article is so bad I have to rip it apart.
Assume that the server never supported zapping in the past. Now we add a feature that immediately and without asking terminates your session, shuts down all applications, logs you out, brings down your wireless network in the process, shuts down your VPN and generally makes the computer giggle at you.
That is one way to look at it. You could also see it as adding a cool innovative new feature which allows inexperienced users to easily recover from a bad X state.
Experienced users will benefit from zapping. So make it accessible to them.
No. Inexperienced users will benefit from zapping. So it should be enabled by default.
Could you imagine if this was actually put up for a vote?
"Stay with default" option would be hands-down winner.
Good thing we will never see a vote on this, because we have to keep the x.org egos inflated. They could never possibly make a brain dead lemming like decision.
A vote of a non-representative group is still non-representative even if you increase the sample size. so I'm not sure how a vote on fedora-devel should be worth more or less than a vote on xorg-devel, or gnome-devel, or... There are also a number of users that don't know what a mailing list is and could not partake in such a vote. (I have evidence of at least one, presumably there are others). if you have _actual_ representative data, I'll happily listen. seriously.
Please remember too that we're talking about a feature that is accessible to every GUI user, not just administrators, developers, geeks, or some other more experienced group.
I'm also not quite sure how that has to do with inflating egos. I've spent some time last week to actually make you (well, not particularly you, but..) happy because: - fedora-setup-keyboard merges this key automatically now. - the C-A-B shortcut is enabled during gdm, which is usually when you notice that something output-related is broken. - c-a-b works if you start your custom X session, it gets disabled (or not) when gnome-keyboard-properties applies the user-configured settings. - gnome-keyboard-properties provides a simple checkbox to enable it if needed. - the xkeyboard-config rules allow for different combinations for Terminate_Server, which is an improvement to the hardcoded value before. - the keyboard driver now uses XKB instead of its own hardcoded zapping.
With the exception that the default is still disabled, everything around is now working properly and pretty much like layouts generally do. I'm sorry that you don't like the new default, but at least I've tried hard to make it as easy for you to enable it again should you need it.
Cheers, Peter, who asks for a at least a polite tone in the reply.
On Mon, Apr 13, 2009 at 5:50 PM, Peter Hutterer peter.hutterer@who-t.net wrote:
On Sun, Apr 12, 2009 at 09:58:50PM -0700, Christopher Stone wrote:
Good thing we will never see a vote on this, because we have to keep the x.org egos inflated. They could never possibly make a brain dead lemming like decision.
A vote of a non-representative group is still non-representative even if you increase the sample size. so I'm not sure how a vote on fedora-devel should be worth more or less than a vote on xorg-devel, or gnome-devel, or... There are also a number of users that don't know what a mailing list is and could not partake in such a vote. (I have evidence of at least one, presumably there are others). if you have _actual_ representative data, I'll happily listen. seriously.
Actually, I was thinking about a formal vote, like they do for FESCo elections, or voting for a Fedora code name. It would not be representative of every single Fedora end user, but it would represent the people who contribute to Fedora. I hope you will happily listen to the results of such a vote if it ever occurs, which it wont, so the point is moot.
Please remember too that we're talking about a feature that is accessible to every GUI user, not just administrators, developers, geeks, or some other more experienced group.
Indeed, please remember that this feature is most beneficial to non-administrators, and non-developers, and non-geeks or some other non-experienced group. It is the experienced ones who are able to switch virtual consoles and recover from X properly. It is the non-experienced ones who need a short-cut key. Especially when it comes to handling support calls/chats.
I'm also not quite sure how that has to do with inflating egos. I've spent some time last week to actually make you (well, not particularly you, but..) happy because:
It's quite simple. You have to have a hyper-inflated ego in order to actually think that disabling a crash recovery feature in your software will somehow be beneficial, or is no longer used or needed. If the x.org guys didn't have huge egos they would have rejected disabling the zap feature outright on the grounds that their software is no where near stable enough to illicit such a change in the defaults.
- fedora-setup-keyboard merges this key automatically now.
- the C-A-B shortcut is enabled during gdm, which is usually when you notice
that something output-related is broken.
- c-a-b works if you start your custom X session, it gets disabled (or not)
when gnome-keyboard-properties applies the user-configured settings.
- gnome-keyboard-properties provides a simple checkbox to enable it if needed.
- the xkeyboard-config rules allow for different combinations for
Terminate_Server, which is an improvement to the hardcoded value before.
- the keyboard driver now uses XKB instead of its own hardcoded zapping.
This is awesome, I hope that this work goes towards making it easy to *remove* the don't zap. As it should be enabled by default. Should you happen to be in the extreme minority of people who do not want ctrl-alt-bksp enabled by default, it is now easy for you to disable it. Thanks to everyone for their contributions, now if only we can get the defaults correct....
On Mon, Apr 13, 2009 at 5:50 PM, Peter Hutterer peter.hutterer@who-t.net wrote:
A vote of a non-representative group is still non-representative even if you increase the sample size. so I'm not sure how a vote on fedora-devel should be worth more or less than a vote on xorg-devel, or gnome-devel, or...
You should also note that on every single mailing list, including x.org's, the consensus was overwhelming against changing the defaults. You will not be able to find any thread on any mailing list where the majority of people think ctrl-alt-bksp should be removed. Try to prove me wrong.
The decision was made on IRC by a couple of guys with no logs.
On Mon, Apr 13, 2009 at 9:10 PM, Christopher Stone chris.stone@gmail.com wrote:
On Mon, Apr 13, 2009 at 5:50 PM, Peter Hutterer The decision was made on IRC by a couple of guys with no logs.
It would be nice if there were _at least_ some logs so the rest of us could follow the logic.
On 04/13/2009 06:50 PM, Peter Hutterer wrote:
A vote of a non-representative group is still non-representative even if you increase the sample size. so I'm not sure how a vote on fedora-devel should be worth more or less than a vote on xorg-devel, or gnome-devel, or... There are also a number of users that don't know what a mailing list is and could not partake in such a vote. (I have evidence of at least one, presumably there are others). if you have _actual_ representative data, I'll happily listen. seriously.
Interesting logic -- first establish that it's not possible to get representative data, then offer to listen to "_actual_ representative" data... Catch 22?
I've spent some time last week to actually make you (well, not particularly you, but..) happy because:
- fedora-setup-keyboard merges this key automatically now.
- the C-A-B shortcut is enabled during gdm, which is usually when you notice that something output-related is broken.
- c-a-b works if you start your custom X session, it gets disabled (or not) when gnome-keyboard-properties applies the user-configured settings.
- gnome-keyboard-properties provides a simple checkbox to enable it if needed.
- the xkeyboard-config rules allow for different combinations for Terminate_Server, which is an improvement to the hardcoded value before.
- the keyboard driver now uses XKB instead of its own hardcoded zapping.
With the exception that the default is still disabled, everything around is now working properly and pretty much like layouts generally do. I'm sorry that you don't like the new default, but at least I've tried hard to make it as easy for you to enable it again should you need it.
This sounds sane. Is it also the case with other desktops (e.g. KDE), window managers (should I choose to run, eh, AfterStep, instead of full blown desktop environment)?
Thanks for your work, Peter.
On Tuesday 14 April 2009 05:50:34 Peter Hutterer wrote:
- the keyboard driver now uses XKB instead of its own hardcoded zapping.
Does this solution have the same chance to kill X as the hardcoded one had? I mean: is it possible for situations where this solution doesn't work but the old one does? (like broken grabs...)
Cheers, Suren
On Tue, Apr 14, 2009 at 10:54:10AM +0500, Suren Karapetyan wrote:
On Tuesday 14 April 2009 05:50:34 Peter Hutterer wrote:
- the keyboard driver now uses XKB instead of its own hardcoded zapping.
Does this solution have the same chance to kill X as the hardcoded one had? I mean: is it possible for situations where this solution doesn't work but the old one does? (like broken grabs...)
There's no difference to before.
Let me explain in a bit more detail. zapping consists of two parts: - the Terminate_Server XKB symbol triggers the action, if you don't have this symbol, you can't trigger it. - The DontZap setting controls whether the Terminate_Server action will actually do anything when you trigger it.
The keyboard driver had - from the times before XKB was around - code for so-called "Special Key" handling. Including a hard-coded path for C-A-B. This means that even if you move the Terminate_Server symbol to another key or remove it altogether, the keyboard driver would zap the server on C-A-B. This is what's been removed, it now posts the keycode, and passes through the xkb action handling.
The only time when the old code _could_ have worked but the new one didn't is if the server is stuck somewhere in the main loop but still handles signals fine - so the zapping would be triggered during the SIGIO handling.
I said "could", because the keyboard driver didn't use signals for input events, it just got polled (checked the git logs, goes back to the original X.Org import from Xfree86). So kbd was relying on a working main loop anyway.
The only difference is that we can move or remove the Terminate_Server symbol key now.
Makes sense?
Cheers, Peter
On Wednesday 15 April 2009 04:44:40 Peter Hutterer wrote:
On Tue, Apr 14, 2009 at 10:54:10AM +0500, Suren Karapetyan wrote:
On Tuesday 14 April 2009 05:50:34 Peter Hutterer wrote:
- the keyboard driver now uses XKB instead of its own hardcoded
zapping.
Does this solution have the same chance to kill X as the hardcoded one had? I mean: is it possible for situations where this solution doesn't work but the old one does? (like broken grabs...)
There's no difference to before.
Let me explain in a bit more detail. zapping consists of two parts:
- the Terminate_Server XKB symbol triggers the action, if you don't have
this symbol, you can't trigger it.
- The DontZap setting controls whether the Terminate_Server action will actually do anything when you trigger it.
The keyboard driver had - from the times before XKB was around - code for so-called "Special Key" handling. Including a hard-coded path for C-A-B. This means that even if you move the Terminate_Server symbol to another key or remove it altogether, the keyboard driver would zap the server on C-A-B. This is what's been removed, it now posts the keycode, and passes through the xkb action handling.
The only time when the old code _could_ have worked but the new one didn't is if the server is stuck somewhere in the main loop but still handles signals fine - so the zapping would be triggered during the SIGIO handling.
I said "could", because the keyboard driver didn't use signals for input events, it just got polled (checked the git logs, goes back to the original X.Org import from Xfree86). So kbd was relying on a working main loop anyway.
The only difference is that we can move or remove the Terminate_Server symbol key now.
Makes sense?
Cheers, Peter
Thanks for the explanation! It makes sense now.
Suren
On Wednesday 08 April 2009 08:39:23 Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
Indeed! Or MiHl proposed #494528 - RFE: Zap after warning same way as OpenSuse do it - one warning then action.
I can understand this behavior once drivers become rock solid, no lockups, no freezes, no hangs but their state these days is horrible... I understand why - not easy to access docs, few people who actually understand it and even fewer people who are working on it.
Jaroslav
Kevin Kofler
On Wed, 2009-04-08 at 08:39 +0200, Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
+1 [How does this voting system work anyways?]
Yes. In a multiuser system it makes a world of difference if you have to reboot to get X working again, or if just killing X is enough.
To me it sounds silly to disable a quintessential feature of X just because of a few broken programs that want to use Ctrl-Alt-BkSp for something else.
If a user really wants to disable Ctrl-Alt-BkSp in order to use it as input to a program, then s/he should modify xorg.conf.
PS. Personally I've had a lot more trouble with the sleep button on my keyboard instead of Ctrl-Alt-Bksp, since the former is a lot easier to trigger unwantedly :D
On Wednesday 08 April 2009 08:39:23 Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
Kevin Kofler
I've made public inquiry at www.abclinuxu.cz ( http://www.abclinuxu.cz/ankety/zruseni-ctrl-alt-backspace-v-x ) - CZECH only, but unfortunately I asked my question wrong way.
Instead of 'Disabling ctrl+alt+backspace by default in the X server is:' I used 'Removing ctrl+alt+backspace from the X server is:' a) good idea b) I don't care c) bad joke
results: voters: 1405 a) 3 % (45 votes) b) 7 % (92 votes) c) 90 % (1268 votes)
I agree this voting is not completely "valid" because of wrong question.
Michal
On 4/8/2009 2:39 AM, Kevin Kofler wrote:
David wrote:
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this.
And that's exactly what we're complaining about.
It looks very clear to me that the majority either wants Ctrl+Alt+BkSp enabled by default or doesn't care either way. Only very few people have ever accidentally triggered it.
And that I can understand. I'm one of the 'don't care' either way people. BTW. But I think that you need to go directly to the Xorg devs. Fedora has always, as far as I know, pretty much followed what upstream provides.
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves. :-P
David wrote:
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves.
Not quite: Setting up a distro and preparing packages to include them into a distro is a bit more than blindly following "upstreams".
It also means package maintainers to be "listening" to their "user-base", to communicate their user-base's concerns upstream and to deviate from upstream when necessary.
On 4/8/2009 11:25 AM, Ralf Corsepius wrote:
David wrote:
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves.
Not quite: Setting up a distro and preparing packages to include them into a distro is a bit more than blindly following "upstreams".
It also means package maintainers to be "listening" to their "user-base", to communicate their user-base's concerns upstream and to deviate from upstream when necessary.
This discussion has me thinking about the last time I used C-A-B. And quite honestly I can't recall the last time. Certainly not recently.
I do understand that some users have problems and that some users need this. However the way I see it is that the change has been made and will not be reverting.
Which means one of two things.
Learn to live with the change. Or you change it back yourself. Seems simple enough. One little change. I change many things from the defaults.
David wrote:
On 4/8/2009 11:25 AM, Ralf Corsepius wrote:
David wrote:
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves.
Not quite: Setting up a distro and preparing packages to include them into a distro is a bit more than blindly following "upstreams".
It also means package maintainers to be "listening" to their "user-base", to communicate their user-base's concerns upstream and to deviate from upstream when necessary.
This discussion has me thinking about the last time I used C-A-B. And quite honestly I can't recall the last time. Certainly not recently.
As far as I am concerned: - Ca. 4 weeks ago, when trying to get my netbook working with an external monitor - Today, when something, I don't know, crashed X and left me with an entirely black screen.
I do understand that some users have problems and that some users need this. However the way I see it is that the change has been made and will not be reverting.
Which means one of two things.
Learn to live with the change. Or you change it back yourself.
Or ... - exchange upstream - exchange the fedora packager - switch the distro. - fork the distro (This change alone is easy to patch).
I think it would be appropriate to have FESCo interfere and let them vote on this matter.
Seems simple enough. One little change. I change many things from the defaults.
With each fedora release, I increasing change more. I am seriously asking myself why I am using a distro which is increasingly divering from my needs, and which I experience to be increasingly less usable wrt. certain aspects.
On Wed, Apr 8, 2009 at 11:29 AM, Ralf Corsepius rc040203@freenet.de wrote:
David wrote:
enough. One little change. I change many things from the defaults.
With each fedora release, I increasing change more. I am seriously asking myself why I am using a distro which is increasingly divering from my needs, and which I experience to be increasingly less usable wrt. certain aspects.
My list of post install things to do is also growing. Partly due to the fact that I use the LiveCD for install now, which at least offsets some of the increasing post install setup time.
Arthur Pemberton wrote:
On Wed, Apr 8, 2009 at 11:29 AM, Ralf Corsepius rc040203@freenet.de wrote:
David wrote:
enough. One little change. I change many things from the defaults.
With each fedora release, I increasing change more. I am seriously asking myself why I am using a distro which is increasingly divering from my needs, and which I experience to be increasingly less usable wrt. certain aspects.
My list of post install things to do is also growing.
If all issues were so easy to fix - Unfortunately, there are serious issues with some packages which can't be fixed with %posts/%triggers ;)
In many such cases, my impression is, the causes are related to certain individuals, who are not able to separate their roles as "upstream developer", "packager" and (not too seldom) as "@RH".
On 4/8/2009 12:29 PM, Ralf Corsepius wrote:
David wrote:
On 4/8/2009 11:25 AM, Ralf Corsepius wrote:
David wrote:
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves.
Not quite: Setting up a distro and preparing packages to include them into a distro is a bit more than blindly following "upstreams".
It also means package maintainers to be "listening" to their "user-base", to communicate their user-base's concerns upstream and to deviate from upstream when necessary.
This discussion has me thinking about the last time I used C-A-B. And quite honestly I can't recall the last time. Certainly not recently.
As far as I am concerned:
- Ca. 4 weeks ago, when trying to get my netbook working with an
external monitor
- Today, when something, I don't know, crashed X and left me with an
entirely black screen.
I do understand that some users have problems and that some users need this. However the way I see it is that the change has been made and will not be reverting.
Which means one of two things.
Learn to live with the change. Or you change it back yourself.
Or ...
- exchange upstream
- exchange the fedora packager
- switch the distro.
- fork the distro (This change alone is easy to patch).
I think it would be appropriate to have FESCo interfere and let them vote on this matter.
Seems simple enough. One little change. I change many things from the defaults.
With each fedora release, I increasing change more. I am seriously asking myself why I am using a distro which is increasingly divering from my needs, and which I experience to be increasingly less usable wrt. certain aspects.
"switch the distro" won't work. All are going this way.
So your, and the others, options, are the same as when this started. Change it back yourself or live with it the way that it is now. A sub-option, if you want to call it an option, is to keep complaining. But you will still have to deal with one of the others.
I am sorry that you are unhappy but I am finished with this thread.
Bye
On Wed, 8 Apr 2009, David wrote:
"switch the distro" won't work. All are going this way.
So your, and the others, options, are the same as when this started. Change it back yourself or live with it the way that it is now. A sub-option, if you want to call it an option, is to keep complaining. But you will still have to deal with one of the others.
I am sorry that you are unhappy but I am finished with this thread.
On the contrary, look at the "sbin in regular user path". It came up repeatedly and lo and behold, we have it in Fedora 11 and it makes me happy. You describe it as "complaining" but you can describe it also as "pointing out an issue has not reached concensus"
We'll see where the ctrl-alt-bksp issue goes. And for tha matter, the same with the requests I've made in the past for UseDNS in the default ssh client config.
Paul
David wrote:
On 4/8/2009 12:29 PM, Ralf Corsepius wrote:
David wrote:
On 4/8/2009 11:25 AM, Ralf Corsepius wrote:
David wrote:
Said more simply? This discussion is on the wrong list. You're talking to the wrong people. Actually it looks more like you are talking to yourselves.
Not quite: Setting up a distro and preparing packages to include them into a distro is a bit more than blindly following "upstreams".
It also means package maintainers to be "listening" to their "user-base", to communicate their user-base's concerns upstream and to deviate from upstream when necessary.
This discussion has me thinking about the last time I used C-A-B. And quite honestly I can't recall the last time. Certainly not recently.
As far as I am concerned:
- Ca. 4 weeks ago, when trying to get my netbook working with an
external monitor
- Today, when something, I don't know, crashed X and left me with an
entirely black screen.
I do understand that some users have problems and that some users need this. However the way I see it is that the change has been made and will not be reverting.
Which means one of two things.
Learn to live with the change. Or you change it back yourself.
Or ...
- exchange upstream
- exchange the fedora packager
- switch the distro.
- fork the distro (This change alone is easy to patch).
I think it would be appropriate to have FESCo interfere and let them vote on this matter.
Seems simple enough. One little change. I change many things from the defaults.
With each fedora release, I increasing change more. I am seriously asking myself why I am using a distro which is increasingly divering from my needs, and which I experience to be increasingly less usable wrt. certain aspects.
"switch the distro" won't work. All are going this way.
You might want to have a second glance - SuSE Factory has this:
--- xorg-server-1.6.0/hw/xfree86/common/xf86Config.c.orig 2009-02-28 20:29:42.000000000 +0100 +++ xorg-server-1.6.0/hw/xfree86/common/xf86Config.c 2009-02-28 20:30:44.000000000 +0100 @@ -734,7 +734,7 @@ static OptionInfoRec FlagOptions[] = { { FLAG_DONTVTSWITCH, "DontVTSwitch", OPTV_BOOLEAN, {0}, FALSE }, { FLAG_DONTZAP, "DontZap", OPTV_BOOLEAN, - {0}, TRUE }, + {0}, FALSE }, { FLAG_ZAPWARNING, "ZapWarning", OPTV_BOOLEAN, {0}, FALSE }, { FLAG_DONTZOOM, "DontZoom", OPTV_BOOLEAN,
So your, and the others, options, are the same as when this started.
Correct - my opinion still is: upstream's decision is silly.
Also, provided the hazzle this issue has caused, I am deeply convinced the change would have been reverted in Fedora, if Fedora package maintainer wasn't a RH employee.
On Wed, Apr 8, 2009 at 3:35 PM, Ralf Corsepius rc040203@freenet.de wrote:
Correct - my opinion still is: upstream's decision is silly.
Anyone propose a patch that moves the zap to something suitably impossible to press accidentally… like ctrl-alt-backspace then type killxnow? :)
Also, provided the hazzle this issue has caused, I am deeply convinced the change would have been reverted in Fedora, if Fedora package maintainer wasn't a RH employee.
Oh no! RedHat has been subverted by the conspiracy of emacs users?!
Gregory Maxwell wrote:
On Wed, Apr 8, 2009 at 3:35 PM, Ralf Corsepius rc040203@freenet.de wrote:
Correct - my opinion still is: upstream's decision is silly.
Anyone propose a patch that moves the zap to something suitably impossible to press accidentally… like ctrl-alt-backspace then type killxnow? :)
As I wrote several time before in this threat: To me, c-a-bs is an emergency button. Any additional action but "immediately killing X" voids this aspect.
Also, provided the hazzle this issue has caused, I am deeply convinced the change would have been reverted in Fedora, if Fedora package maintainer wasn't a RH employee.
Oh no! RedHat has been subverted by the conspiracy of emacs users?!
No, experience in Fedora has told many @RedHat's simply don't care about opinion but their own.
It's what is commonly known as "Red Hat's arrogance and ignorance".
On Fri, 2009-04-10 at 10:54 +0200, Ralf Corsepius wrote:
Gregory Maxwell wrote:
On Wed, Apr 8, 2009 at 3:35 PM, Ralf Corsepius rc040203@freenet.de wrote:
Correct - my opinion still is: upstream's decision is silly.
Anyone propose a patch that moves the zap to something suitably impossible to press accidentally… like ctrl-alt-backspace then type killxnow? :)
As I wrote several time before in this threat: To me, c-a-bs is an emergency button. Any additional action but "immediately killing X" voids this aspect.
Also, provided the hazzle this issue has caused, I am deeply convinced the change would have been reverted in Fedora, if Fedora package maintainer wasn't a RH employee.
Oh no! RedHat has been subverted by the conspiracy of emacs users?!
No, experience in Fedora has told many @RedHat's simply don't care about opinion but their own.
It's what is commonly known as "Red Hat's arrogance and ignorance".
I think you got the causality wrong, Red Hat hired us because we were arrogant dicks that are very good at what we do in our own opinions.
We didn't become that way after we joined Red Hat.
Dave.
On Fri, Apr 10, 2009 at 4:54 AM, Ralf Corsepius rc040203@freenet.de wrote:
Gregory Maxwell wrote:
On Wed, Apr 8, 2009 at 3:35 PM, Ralf Corsepius rc040203@freenet.de wrote:
Correct - my opinion still is: upstream's decision is silly.
Anyone propose a patch that moves the zap to something suitably impossible to press accidentally… like ctrl-alt-backspace then type killxnow? :)
As I wrote several time before in this threat: To me, c-a-bs is an emergency button. Any additional action but "immediately killing X" voids this aspect.
Perhaps I was unclear? I wasn't suggesting there be a popup asking if you're really sure. I was suggesting that the key sequence be changed to something that is actually impossible to press accidentally. If such a change were made it would eliminate the only argument for disabling the zap button, as far as I can tell.
(And yes, it's quite possible to press the current button accidentally)
I think the biggest problem with that proposal is keymap issues, although that exists for c-a-bs too.
David wrote:
On 4/7/2009 4:31 PM, Paul Wouters wrote:
My first fedora 11 beta run-in with not having dontzap set to false.
I have a dual twin setup. This already requires an xorg.conf file, because in the no-config running, xorg decides you want to only use one of the two screens it finds. I don't know why, I consider it a bug.
I configure it to have dual screen. When I login, my window manager dies on me. My gnome panel only works in my left screen, but anything I select pops up on the right screen, out of focus. I can move my mouse into the other screen but I cannot get any focus. So I cannot get to a terminal or any other application.
So I ctrl-alt-f2. Now my left monitor goes into standby mode and my right monitor just shows a blinking cursor. The same for all VC's except alt-f7 which brings me back to my broken X session.
I had to login via ssh to fix my machine, because I could not restart X. Indeed, i did not have sysrq enabled yet, that's disabled per default it seems.
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Sorry Paul. But it does not look like, from what I read here, that there will be a vote on this. It appears that you, and all others, have two options.
1: Live with it as default 'off'.
2: Change it to 'on' to suit yourself.
Great thing about Linux. You can do pretty much just as you wish.
Good luck.
the vote was taken in secret by the xorg folks a while ago, real community spirit in action!
phil
psmith wrote:
the vote was taken in secret by the xorg folks a while ago, real community spirit in action!
phil
So we can either write a new graphics server or let them get away with it. Would /you/ like to write a knew graphics server?
The term is meritocracy, not democracy. Someone's mistakes are going to show up in the product no matter what. It tends to work out for the best if its the mistakes of people who usually know what they're doing.
--CJD
Casey Dahlin wrote:
psmith wrote:
the vote was taken in secret by the xorg folks a while ago, real community spirit in action!
phil
So we can either write a new graphics server or let them get away with it. Would /you/ like to write a knew graphics server?
The term is meritocracy, not democracy. Someone's mistakes are going to show up in the product no matter what. It tends to work out for the best if its the mistakes of people who usually know what they're doing.
--CJD
but why the need for a *secret *vote? why not do it in the open so all of the x users can go look at the logs and at least see their reasoning behind this change that effects every x user?
better still the fedora 'community' can revert the upstream change, it has been done plenty of times before and i don't see why this should be different?
as for writing a new gfx server, well no but i will compile x with the standard behaviour set regardless
phil
Olivier Galibert wrote:
On Wed, Apr 08, 2009 at 02:46:14PM +0100, psmith wrote:
the vote was taken in secret by the xorg folks a while ago, real community spirit in action!
And the result of the vote was "don't change the default" iirc.
OG.
wtf! then if the secret x devs vote result was not to change the default then why did the change get in?
phil
On Wed, Apr 08, 2009 at 10:17:43PM +0200, Olivier Galibert wrote:
On Wed, Apr 08, 2009 at 09:10:06PM +0100, psmith wrote:
wtf! then if the secret x devs vote result was not to change the default then why did the change get in?
That's known as "last patch wins". Or "nyah nyah nyah I have commit access and you don't".
Or "The people who do the work get to make the decisions".
Matthew Garrett wrote:
On Wed, Apr 08, 2009 at 10:17:43PM +0200, Olivier Galibert wrote:
On Wed, Apr 08, 2009 at 09:10:06PM +0100, psmith wrote:
wtf! then if the secret x devs vote result was not to change the default then why did the change get in?
That's known as "last patch wins". Or "nyah nyah nyah I have commit access and you don't".
Or "The people who do the work get to make the decisions".
but the devs are the people who do the work, and if the majority of those devs voted no, why did the change go through?. this is getting more and more unbelievable as more facts come out
phil
On Wed, Apr 08, 2009 at 09:28:28PM +0100, psmith wrote:
Matthew Garrett wrote:
Or "The people who do the work get to make the decisions".
but the devs are the people who do the work, and if the majority of those devs voted no, why did the change go through?. this is getting more and more unbelievable as more facts come out
There was no vote amongst developers. A decision was made and there was no significant dissent amongst the developer body. Like most upstream projects, decisions about X result from gaining rough consensus amongst the relevant people. User viewpoints will be considered as part of this, but there's no concept of "This many people complained on the mailing list, therefore we are forbidden from carrying out this change".
Matthew Garrett wrote:
There was no vote amongst developers. A decision was made and there was no significant dissent amongst the developer body. Like most upstream projects, decisions about X result from gaining rough consensus amongst the relevant people. User viewpoints will be considered as part of this, but there's no concept of "This many people complained on the mailing list, therefore we are forbidden from carrying out this change".
I think the big problem is the "mailing list consensus" decision making process which is used by many Free Software projects. I've seen it happen in other projects too (e.g. GCC). How it usually goes: 1. maintainer X posts a patch to the mailing list for discussion, 2. some people post objections, others approve of the patch, 3. the discussion eventually dies off because people realize nobody is going to change their mind and move on - no visible consensus was ever reached, 4. maintainer X somehow concludes consensus was reached (the mechanisms at work are not clear to me: Maybe maintainer X subconciously doesn't notice the objections? Maybe they conclude that consensus was reached because no new objections are being raised, despite the existing ones not having been addressed? Maybe there is something which gives insiders a feeling of consensus which an outsider just doesn't see?) and commits the change.
Kevin Kofler
Rahul Sundaram wrote:
psmith wrote:
the vote was taken in secret by the xorg folks a while ago, real community spirit in action!
You are just totally making up things like that and repeating it endlessly. Stop doing so.
Rahul
well it so happens that an xorg dev posted this info, that it was taken in a closed vote, in response to someone asking for a link to the logs where the decision was made. when i get a minute i'll dig through the logs and find it
phil
On Wednesday 08 April 2009 01:31:41 Paul Wouters wrote:
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Paul
Time for some stats.
From all the people who replied to this thread:
19 were for reverting the change, 1 of them from RH. 7 were for keeping the change, 5 of them from RH. And a lot of others who were neutral (at least I couldn't decide on which side they were).
Suren
On Sun, Apr 19, 2009 at 11:22 PM, Suren Karapetyan surenkarapetyan@gmail.com wrote:
On Wednesday 08 April 2009 01:31:41 Paul Wouters wrote:
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Paul
Time for some stats.
From all the people who replied to this thread:
19 were for reverting the change, 1 of them from RH. 7 were for keeping the change, 5 of them from RH. And a lot of others who were neutral (at least I couldn't decide on which side they were).
Suren
So if you keep people out that work for redhat you get: 18 in favor of the old default 2 in favor of the new default.
I think we have a clear winner. But o wait, a vote like this isn't subjective/valid/wished or whatever the reason is. (sarcastic, i hope you notice. don't bother replying with this as quote)
It is interesting to see that the the majority of the people are in favor of the old behaviour but the majority of redhat is in favor of the new one.
I wonder if this new default is ever goingto be changed back to it's old state or that we just have to "bite" trough the change till we get used to it.
Btw. i don't know if i'm in your 19 number as well but i'm havily in favor of the OLD default so could me in there ^_^
On Sun, Apr 19, 2009 at 8:19 PM, Mark markg85@gmail.com wrote:
On Sun, Apr 19, 2009 at 11:22 PM, Suren Karapetyan surenkarapetyan@gmail.com wrote:
On Wednesday 08 April 2009 01:31:41 Paul Wouters wrote:
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Paul
Time for some stats.
From all the people who replied to this thread:
19 were for reverting the change, 1 of them from RH. 7 were for keeping the change, 5 of them from RH. And a lot of others who were neutral (at least I couldn't decide on which
side
they were).
Suren
So if you keep people out that work for redhat you get: 18 in favor of the old default 2 in favor of the new default.
I think we have a clear winner. But o wait, a vote like this isn't subjective/valid/wished or whatever the reason is. (sarcastic, i hope you notice. don't bother replying with this as quote)
It is interesting to see that the the majority of the people are in favor of the old behaviour but the majority of redhat is in favor of the new one.
I wonder if this new default is ever goingto be changed back to it's old state or that we just have to "bite" trough the change till we get used to it.
Btw. i don't know if i'm in your 19 number as well but i'm havily in favor of the OLD default so could me in there ^_^
Poll or not, mark me as 20.
On Sun, Apr 19, 2009 at 9:13 PM, L. Gabriel Somlo somlo@cmu.edu wrote:
On Sun, 19 Apr 2009, Dr. Diesel wrote:
Poll or not, mark me as 20.
Allright, 21 in favor of the old default :)
I like to have ctrl-alt-backspace, and at the same time prefer to stick as close to a default install as possible.
Short of forking Fedora and/or Xorg I don't see the powers that be changing their mind.
On Sun, Apr 19, 2009 at 5:22 PM, Suren Karapetyan wrote:
On Wednesday 08 April 2009 01:31:41 Paul Wouters wrote:
+1 for reverting to the old behaviour of having ctrl-alt-backspace kill the current X session.
Paul
Time for some stats.
From all the people who replied to this thread:
19 were for reverting the change, 1 of them from RH. 7 were for keeping the change, 5 of them from RH. And a lot of others who were neutral (at least I couldn't decide on which side they were).
That makes me wonder if having pushed ctrl+alt+backspace accidentally at least once is an employment criteria at Redhat.
Or maybe it's coincidence.
Orcan