GNOME Desktop Development List

Davyd Madeley | 9 Apr 2006 15:29

[gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:

> Since the original announcement mail about gnome-power-manager, we have
> moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> starting to integrate with other parts of the GNOME application stack.
> Lots of new functionality has been added, and lots of polish has been
> applied. See the screenshots area of my website[5] for some cool
> screenshots of the latest stuff in the 2-15 branch.

I would like to see g-p-m fragmented into three parts.

 * A daemon with no GTK+ dependance that would be suitable for
   cross-desktop use
 * A capplet (this exists today)
 * A notification area icon (libnotify dependance goes here)

This would allow us to more easily address integration issues with
GNOME and other desktops and it means that we can aim at avoiding
notification area pollution (because session initialised notification
icons are a violation of all that is good and right).

--d

--

-- 
Davyd Madeley

http://www.davyd.id.au/
08B0 341A 0B9B 08BB 2118  C060 2EDD BB4F 5191 6CDA

headers

Rodrigo Moya | 10 Apr 2006 11:43

[gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> 
> > Since the original announcement mail about gnome-power-manager, we have
> > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > starting to integrate with other parts of the GNOME application stack.
> > Lots of new functionality has been added, and lots of polish has been
> > applied. See the screenshots area of my website[5] for some cool
> > screenshots of the latest stuff in the 2-15 branch.
> 
> I would like to see g-p-m fragmented into three parts.
> 
>  * A daemon with no GTK+ dependance that would be suitable for
>    cross-desktop use
>  * A capplet (this exists today)
>  * A notification area icon (libnotify dependance goes here)
> 
> This would allow us to more easily address integration issues with
> GNOME and other desktops and it means that we can aim at avoiding
> notification area pollution (because session initialised notification
> icons are a violation of all that is good and right).
> 
since there is already a daemon (HAL and underlying power save software,
like pmu, powersave, etc), why do we need another daemon? I think the
current g-p-m architecture, with the 'daemon' being also the
notification icon makes a lot of sense. If adding the tray icon on
startup is wrong, then I guess we can easily delay the addition of the
icon, like we do with the typing break, for instance.
--

-- 
Rodrigo Moya <rodrigo@...>

(Continue reading)

headers

Ben Maurer | 11 Apr 2006 01:03

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 10 Apr 2006, Rodrigo Moya wrote:
> since there is already a daemon (HAL and underlying power save software,
> like pmu, powersave, etc), why do we need another daemon? I think the
> current g-p-m architecture, with the 'daemon' being also the
> notification icon makes a lot of sense. If adding the tray icon on
> startup is wrong, then I guess we can easily delay the addition of the
> icon, like we do with the typing break, for instance.

I'd point out that (as mentioned on my blog) using `daemons' that are 
tasked with putting an icon in the panel cause extra memory usage. On my 
laptop, this is responsible for 3MB of private, dirty rss. My laptop is 
*plugged in* and it is using this.

The task of `display an icon on the panel when battery is being used' 
should not require this amount of memory.

-- Ben

headers

Steve Frécinaux | 9 Apr 2006 17:13

Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Davyd Madeley wrote:
> I would like to see g-p-m fragmented into three parts.
> 
>  * A daemon with no GTK+ dependance that would be suitable for
>    cross-desktop use
>  * A capplet (this exists today)
>  * A notification area icon (libnotify dependance goes here)
> 
> This would allow us to more easily address integration issues with
> GNOME and other desktops and it means that we can aim at avoiding
> notification area pollution (because session initialised notification
> icons are a violation of all that is good and right).

On my own I'd like to see a real gnome-terminal applet instead of that 
notification icon, eventually replacing the current batstat applet...

headers

Richard Hughes | 9 Apr 2006 15:47

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> 
> > Since the original announcement mail about gnome-power-manager, we have
> > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > starting to integrate with other parts of the GNOME application stack.
> > Lots of new functionality has been added, and lots of polish has been
> > applied. See the screenshots area of my website[5] for some cool
> > screenshots of the latest stuff in the 2-15 branch.
> 
> I would like to see g-p-m fragmented into three parts.
> 
>  * A daemon with no GTK+ dependance that would be suitable for
>    cross-desktop use
>  * A capplet (this exists today)
>  * A notification area icon (libnotify dependance goes here)

Umm, no.

The IPC between these components would be horrific and over-complicated
for no actual gain. KDE are quite happy with their own power management
applications, and no KDE developer has ever mentioned to me that they
would want such a cross-desktop daemon.

> This would allow us to more easily address integration issues with
> GNOME and other desktops and it means that we can aim at avoiding
> notification area pollution (because session initialised notification
> icons are a violation of all that is good and right).

g-p-m can default to only displaying when the battery power is critical,

(Continue reading)

headers

Shaun McCance | 10 Apr 2006 16:53

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > 
> > > Since the original announcement mail about gnome-power-manager, we have
> > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > starting to integrate with other parts of the GNOME application stack.
> > > Lots of new functionality has been added, and lots of polish has been
> > > applied. See the screenshots area of my website[5] for some cool
> > > screenshots of the latest stuff in the 2-15 branch.
> > 
> > I would like to see g-p-m fragmented into three parts.
> > 
> >  * A daemon with no GTK+ dependance that would be suitable for
> >    cross-desktop use
> >  * A capplet (this exists today)
> >  * A notification area icon (libnotify dependance goes here)
> 
> Umm, no.
> 
> The IPC between these components would be horrific and over-complicated
> for no actual gain. KDE are quite happy with their own power management
> applications, and no KDE developer has ever mentioned to me that they
> would want such a cross-desktop daemon.

Does their power management thing use DBus, and
if so, do we share a common interface?  I care
much less about shared code than about shared
interfaces.

(Continue reading)

headers

Holger Macht | 10 Apr 2006 18:02

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon 10. Apr - 09:53:04, Shaun McCance wrote:
> On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> > On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > > 
> > > > Since the original announcement mail about gnome-power-manager, we have
> > > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > > starting to integrate with other parts of the GNOME application stack.
> > > > Lots of new functionality has been added, and lots of polish has been
> > > > applied. See the screenshots area of my website[5] for some cool
> > > > screenshots of the latest stuff in the 2-15 branch.
> > > 
> > > I would like to see g-p-m fragmented into three parts.
> > > 
> > >  * A daemon with no GTK+ dependance that would be suitable for
> > >    cross-desktop use
> > >  * A capplet (this exists today)
> > >  * A notification area icon (libnotify dependance goes here)
> > 
> > Umm, no.
> > 
> > The IPC between these components would be horrific and over-complicated
> > for no actual gain. KDE are quite happy with their own power management
> > applications, and no KDE developer has ever mentioned to me that they
> > would want such a cross-desktop daemon.
> 
> Does their power management thing use DBus, and
> if so, do we share a common interface?  I care
> much less about shared code than about shared
> interfaces.

(Continue reading)

headers

Matthew Garrett | 10 Apr 2006 18:45

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, Apr 10, 2006 at 06:02:53PM +0200, Holger Macht wrote:

> Ubuntu guys would also take it IMHO if there would be a gnome
> client for it. 

We'll take it if there's a gnome client /and/ it provides the same 
degree of useful functionality as g-p-m currently does. But I think 
we'll have the opportunity to discuss that later this week :)

> And power management is something real complex. It's not just popping up
> some notifications. It's definitely worth to reside in an own daemon like
> NetworkManager which has root privileges and can do CPU frequency scaling,
> throttling, harddisk adjustments, runtime device power management, battery
> management, proper suspend implementation, CPU hotplugging and a lot more.

Woah, hold on there. There's several issues involved here. Hal 
already provides mechanisms to trigger suspend, and it doesn't seem 
illogical to provide functionality like runtime device power management 
in there as well. If I select a wired network in NetworkManager, it 
makes sense for the device to be powered up. If it's going to be talking 
to hal to find out what capabilities the device has /anyway/, I think it 
makes sense for hal to be the basic interface for managing the power 
state. Otherwise NetworkManager has to start talking to the powersave 
daemon as well, which seems awkward.

I'd much rather have a user-level powersave daemon that collates 
available information (such as asking Networkmanager whether any devices 
are up or not) and makes policy decisions that are enacted via hal than 
have policy management and enactment more tightly coupled.

(Continue reading)

headers

David Zeuthen | 11 Apr 2006 17:47

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 17:45 +0100, Matthew Garrett wrote:
> Now, that's a slightly separate problem - that is, the fact that most 
> DBus applications are focused on the "User logged in" case. David 
> Zeuthen's been working on that, and I think there's a more elegant 
> solution than "Run as root, and fall back to a default policy if there's 
> no client".

Yea, I should be able to finish this (PolicyKit etc) soon and then g-p-m
can run even when no user is logged in. When we're there we can do other
stuff such as

 - Integrate with g-d-m so you can get a power icon

 - Make a "Make these settings system-wide" button in the gnome
   power preferences that uses PolicyKit to auth

and other exciting stuff.

I think it's just crazy talk to propose yet another daemon running as
root just to solve the not-so-interesting edge case of enforcing policy
when no one is logged in. Btw, I've suggested to some NetworkManager
developers to behave more like gnome-power-manager (so we don't need a
NetworkManager daemon running as root and an ugly split into system and
session daemon) and they liked it. I've been wanting to propose this on
the NM list but I've been very busy with other stuff.

So, I don't really think it's interesting for g-p-m to add a whole new
dependency on Powersave; of course, distributors can do whatever they
want (and they do) but sooner or later their solution will be replaced
with something that GNOME includes and then they need to make the switch

(Continue reading)

headers

Weinehall David (Nokia-M/Tampere | 12 Apr 2006 09:55

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On tis, 2006-04-11 at 11:47 -0400, ext David Zeuthen wrote:
[snip]

> I think it's just crazy talk to propose yet another daemon running as
> root just to solve the not-so-interesting edge case of enforcing policy
> when no one is logged in. Btw, I've suggested to some NetworkManager
> developers to behave more like gnome-power-manager (so we don't need a
> NetworkManager daemon running as root and an ugly split into system and
> session daemon) and they liked it. I've been wanting to propose this on
> the NM list but I've been very busy with other stuff.

[snip]

Not all GNOME users are regular desktop systems.  Some systems (like the
Nokia 770, for instance), run power management solutions that are
completely customised at the moment because they need to have working
power management even when there's no user session running.

IMHO, all hardware specific power management features
(suspend/hibernate/cpufreq/voltage scaling, etc) should be abstracted to
HAL; that's what it's for -- Hardware Abstraction.  A power management
policy daemon (running even when there's no X), that communicates with
the kernel through HAL, and with user applications through DBus, can
then make appropriate policy based decisions, based on configuration
values (read from a conf-file, gconf, a plugin-system, whatever).  The
only thing needed in GNOME/KDE would be a graphical frontend that shows
notifications/status (the UI-part of gnome-power-manager), and a
configuration interface (for changing the policies used by the daemon).

Regards: David Weinehall

(Continue reading)

headers

Richard Hughes | 11 Apr 2006 23:48

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tue, 2006-04-11 at 11:47 -0400, David Zeuthen wrote:
> I agree it's worthwhile renaming g-p-m's session bus interface from
> org.gnome.PowerManager to org.freedesktop.PowerManager after discussion
> and review on xdg-list. Richard, you should raise this on
> xdg-list <at> freedesktop.org  describing what the interface does (just write
> a small spec) and whether everyone is OK with this interface.

Sounds like a good idea, I'll do that. Thanks.

> p.s. : I'm on PTO right now so it will take a while for me to respond to
> replies

No probs, enjoy!

Richard.

headers

Richard Hughes | 11 Apr 2006 23:48

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tue, 2006-04-11 at 11:47 -0400, David Zeuthen wrote:
> I agree it's worthwhile renaming g-p-m's session bus interface from
> org.gnome.PowerManager to org.freedesktop.PowerManager after discussion
> and review on xdg-list. Richard, you should raise this on
> xdg-list@...  describing what the interface does (just write
> a small spec) and whether everyone is OK with this interface.

Sounds like a good idea, I'll do that. Thanks.

> p.s. : I'm on PTO right now so it will take a while for me to respond to
> replies

No probs, enjoy!

Richard.

headers

Richard Hughes | 10 Apr 2006 18:38

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:02 +0200, Holger Macht wrote:
> No, KDE people are definitely not pleased with their current
> implementation. They still use klaptop by default which is nearly
> completely unmaintained.
> 
> Maybe Richard never heard KDE guys saiing that they want a cross desktop
> daemon, but some of them do. For me personally, it's simply a some kind of
> good software design to share common code and interfaces. And I repeated
> myswlf already too often.

As too have I. I really don't see why we need to add *another* layer of
abstraction just so desktops can standardise on 10% for a battery
critical low notification.

The differences between GNOME and KDE in configuration, language,
politics, HIG and loads of other stuff makes cross platform API choices
very difficult.

> Hal is an '_H_ardware _A_bstraction _L_ayer' and
> _no_ power management daemon. Hal should provide device information like
> battery information and nothing more.

But that's what it does.

It provides information such as battery.charge_level.percentage and
methods such as Suspend() and Hibernate(), anything else is out of the
scope of HAL.

<snip some other stuff>

(Continue reading)

headers

Andrew Sobala | 10 Apr 2006 19:41

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Richard Hughes wrote:
> On Mon, 2006-04-10 at 18:02 +0200, Holger Macht wrote:
>   
>> Richard, please think about your current opinion and maybe try to help to
>> get a good solution for the GNOME desktop which usees one common backend.
>>     
>
> gnome-power-manager is a 400k binary. It uses gconf to store a few
> daemon settings and preferences.
>
> HAL does all the heavy lifting doing all the quirks and talking to stuff
> in /proc and /sys.
>
> g-p-m is like the cherry on the cake, small and simple.
>
> It really doesn't do much more than:
>
> "If battery charge < 10% then notify the user"
> "If battery charge < 5% then suspend if HAL thinks we can"
>
> I really don't see what the big issue is.
>   
I really don't think there is (or should be) an issue with GNOME and KDE 
using different daemons to implement power management policy. That's 
what the difference between desktops *is*.

However, Bob being able to send out one DBus message 
("InhibitInactiveSleep") in BobsCrackLadenAndCPUIntensiveApp that is 
picked up by the power management daemon in GNOME and KDE, whichever the 
app is running under, may be a good thing. And if people really want to

(Continue reading)

headers

Richard Hughes | 10 Apr 2006 20:07

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:41 +0100, Andrew Sobala wrote:
> Richard Hughes wrote:
> > On Mon, 2006-04-10 at 18:02 +0200, Holger Macht wrote:
> >   
> >> Richard, please think about your current opinion and maybe try to help to
> >> get a good solution for the GNOME desktop which usees one common backend.
> >>     
> >
> > gnome-power-manager is a 400k binary. It uses gconf to store a few
> > daemon settings and preferences.
> >
> > HAL does all the heavy lifting doing all the quirks and talking to stuff
> > in /proc and /sys.
> >
> > g-p-m is like the cherry on the cake, small and simple.
> >
> > It really doesn't do much more than:
> >
> > "If battery charge < 10% then notify the user"
> > "If battery charge < 5% then suspend if HAL thinks we can"
> >
> > I really don't see what the big issue is.
> >   
> I really don't think there is (or should be) an issue with GNOME and KDE 
> using different daemons to implement power management policy. That's 
> what the difference between desktops *is*.
> 
> However, Bob being able to send out one DBus message 
> ("InhibitInactiveSleep") in BobsCrackLadenAndCPUIntensiveApp that is 
> picked up by the power management daemon in GNOME and KDE, whichever the

(Continue reading)

headers

Ikke | 10 Apr 2006 20:01

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:41 +0100, Andrew Sobala wrote:
> However, Bob being able to send out one DBus message 
> ("InhibitInactiveSleep") in BobsCrackLadenAndCPUIntensiveApp that is 
> picked up by the power management daemon in GNOME and KDE, whichever the 
> app is running under, may be a good thing. And if people really want to 
> standardise this between desktops, this should be the point of 
> standardisation.
Agree, the DBus interface should most certainly be common (/me remembers
CDIS and hides in a dark corner).
Eg as a Gentoo user I don't want my system to suspend when I'm compiling
something and leave my workstation. GPM got a call for this (indeed,
InhibitInactiveSleep, although it's implementation as far as I saw might
be flawed a little), but you can't expect every developer doing
something like
	if(user_runs_gnome)
		do_dbus_call("org.gnome.PowerManager", fooargs,...);
	if(user_runs_kde)
		do_dbus_call("org.kde.KPowerDaemon", fooargs,...);
	if(user_runs_some_other_power_manager)
		do_something_else();
etc

As mentioned before, a shared base interface is a must, extra
implementation-specific calls can be added on top of that when
appropriate.

Just my .02,

Ikke
http://www.eikke.com

(Continue reading)

headers

Ikke | 10 Apr 2006 20:01

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:41 +0100, Andrew Sobala wrote:
> However, Bob being able to send out one DBus message 
> ("InhibitInactiveSleep") in BobsCrackLadenAndCPUIntensiveApp that is 
> picked up by the power management daemon in GNOME and KDE, whichever the 
> app is running under, may be a good thing. And if people really want to 
> standardise this between desktops, this should be the point of 
> standardisation.
Agree, the DBus interface should most certainly be common (/me remembers
CDIS and hides in a dark corner).
Eg as a Gentoo user I don't want my system to suspend when I'm compiling
something and leave my workstation. GPM got a call for this (indeed,
InhibitInactiveSleep, although it's implementation as far as I saw might
be flawed a little), but you can't expect every developer doing
something like
	if(user_runs_gnome)
		do_dbus_call("org.gnome.PowerManager", fooargs,...);
	if(user_runs_kde)
		do_dbus_call("org.kde.KPowerDaemon", fooargs,...);
	if(user_runs_some_other_power_manager)
		do_something_else();
etc

As mentioned before, a shared base interface is a must, extra
implementation-specific calls can be added on top of that when
appropriate.

Just my .02,

Ikke
http://www.eikke.com

(Continue reading)

headers

Richard Hughes | 10 Apr 2006 20:07

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:41 +0100, Andrew Sobala wrote:
> Richard Hughes wrote:
> > On Mon, 2006-04-10 at 18:02 +0200, Holger Macht wrote:
> >   
> >> Richard, please think about your current opinion and maybe try to help to
> >> get a good solution for the GNOME desktop which usees one common backend.
> >>     
> >
> > gnome-power-manager is a 400k binary. It uses gconf to store a few
> > daemon settings and preferences.
> >
> > HAL does all the heavy lifting doing all the quirks and talking to stuff
> > in /proc and /sys.
> >
> > g-p-m is like the cherry on the cake, small and simple.
> >
> > It really doesn't do much more than:
> >
> > "If battery charge < 10% then notify the user"
> > "If battery charge < 5% then suspend if HAL thinks we can"
> >
> > I really don't see what the big issue is.
> >   
> I really don't think there is (or should be) an issue with GNOME and KDE 
> using different daemons to implement power management policy. That's 
> what the difference between desktops *is*.
> 
> However, Bob being able to send out one DBus message 
> ("InhibitInactiveSleep") in BobsCrackLadenAndCPUIntensiveApp that is 
> picked up by the power management daemon in GNOME and KDE, whichever the

(Continue reading)

headers

Richard Hughes | 10 Apr 2006 18:38

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 18:02 +0200, Holger Macht wrote:
> No, KDE people are definitely not pleased with their current
> implementation. They still use klaptop by default which is nearly
> completely unmaintained.
> 
> Maybe Richard never heard KDE guys saiing that they want a cross desktop
> daemon, but some of them do. For me personally, it's simply a some kind of
> good software design to share common code and interfaces. And I repeated
> myswlf already too often.

As too have I. I really don't see why we need to add *another* layer of
abstraction just so desktops can standardise on 10% for a battery
critical low notification.

The differences between GNOME and KDE in configuration, language,
politics, HIG and loads of other stuff makes cross platform API choices
very difficult.

> Hal is an '_H_ardware _A_bstraction _L_ayer' and
> _no_ power management daemon. Hal should provide device information like
> battery information and nothing more.

But that's what it does.

It provides information such as battery.charge_level.percentage and
methods such as Suspend() and Hibernate(), anything else is out of the
scope of HAL.

<snip some other stuff>

(Continue reading)

headers

Rodrigo Moya | 10 Apr 2006 17:10

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 09:53 -0500, Shaun McCance wrote:
> On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> > On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > > 
> > > > Since the original announcement mail about gnome-power-manager, we have
> > > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > > starting to integrate with other parts of the GNOME application stack.
> > > > Lots of new functionality has been added, and lots of polish has been
> > > > applied. See the screenshots area of my website[5] for some cool
> > > > screenshots of the latest stuff in the 2-15 branch.
> > > 
> > > I would like to see g-p-m fragmented into three parts.
> > > 
> > >  * A daemon with no GTK+ dependance that would be suitable for
> > >    cross-desktop use
> > >  * A capplet (this exists today)
> > >  * A notification area icon (libnotify dependance goes here)
> > 
> > Umm, no.
> > 
> > The IPC between these components would be horrific and over-complicated
> > for no actual gain. KDE are quite happy with their own power management
> > applications, and no KDE developer has ever mentioned to me that they
> > would want such a cross-desktop daemon.
> 
> Does their power management thing use DBus,
>
they use HAL if we're talking about kpowersave

(Continue reading)

headers

Jaap Haitsma | 9 Apr 2006 19:56

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > 
> > > Since the original announcement mail about gnome-power-manager, we have
> > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > starting to integrate with other parts of the GNOME application stack.
> > > Lots of new functionality has been added, and lots of polish has been
> > > applied. See the screenshots area of my website[5] for some cool
> > > screenshots of the latest stuff in the 2-15 branch.
> > 
> > I would like to see g-p-m fragmented into three parts.
> > 
> >  * A daemon with no GTK+ dependance that would be suitable for
> >    cross-desktop use
> >  * A capplet (this exists today)
> >  * A notification area icon (libnotify dependance goes here)
> 
> Umm, no.
> 
> The IPC between these components would be horrific and over-complicated
> for no actual gain. KDE are quite happy with their own power management
> applications, and no KDE developer has ever mentioned to me that they
> would want such a cross-desktop daemon.
> 
Richard,

As far as I understand the code of GPM splitting up GPM in a "daemon"
and a "notication area icon"/applet would not be so hard.

(Continue reading)

headers

Soeren Sonnenburg | 9 Apr 2006 20:52

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 09 Apr 2006 17:56:50 +0000, Jaap Haitsma wrote:

> On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
>> On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
>> > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
[...]
>> > I would like to see g-p-m fragmented into three parts.
>> > 
>> >  * A daemon with no GTK+ dependance that would be suitable for
>> >    cross-desktop use
>> >  * A capplet (this exists today)
>> >  * A notification area icon (libnotify dependance goes here)
>> 
>> Umm, no.
[...]
> So in my opinion it's not that difficult, or am I missing something?

Well and there is something like this already in pbbuttons 

http://pbbuttons.berlios.de/

it has a daemon pbbuttonsd (watching apm/acpi/pmu), nice osd via
gtkpbbuttons (look at these great screenshots
http://pbbuttons.berlios.de/projects/gtkpbbuttons/screen.html), and
configuration manager via powerprefs ... and oh well everything can even
be controlled via cmdline.

So it is doable and has been done...

Soeren.

(Continue reading)

headers

Richard Hughes | 9 Apr 2006 21:42

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 20:52 +0200, Soeren Sonnenburg wrote:
> On Sun, 09 Apr 2006 17:56:50 +0000, Jaap Haitsma wrote:
> 
> > On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> >> On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> >> > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> [...]
> >> > I would like to see g-p-m fragmented into three parts.
> >> > 
> >> >  * A daemon with no GTK+ dependance that would be suitable for
> >> >    cross-desktop use
> >> >  * A capplet (this exists today)
> >> >  * A notification area icon (libnotify dependance goes here)
> >> 
> >> Umm, no.
> [...]
> > So in my opinion it's not that difficult, or am I missing something?
> 
> Well and there is something like this already in pbbuttons 
> 
> http://pbbuttons.berlios.de/
> it has a daemon pbbuttonsd (watching apm/acpi/pmu), nice osd via
> gtkpbbuttons (look at these great screenshots

Erm... it supported /dev/pmu (only Apple hardware) only last time I
looked, although I could be wrong if it's added lots of new stuff very
quickly.

> http://pbbuttons.berlios.de/projects/gtkpbbuttons/screen.html), and
> configuration manager via powerprefs ... and oh well everything can even

(Continue reading)

headers

Soeren Sonnenburg | 9 Apr 2006 22:00

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 20:42 +0100, Richard Hughes wrote:
> On Sun, 2006-04-09 at 20:52 +0200, Soeren Sonnenburg wrote:
> > On Sun, 09 Apr 2006 17:56:50 +0000, Jaap Haitsma wrote:
> > > On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> > >> On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > >> > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > [...]
> > >> > I would like to see g-p-m fragmented into three parts.
> > >> > 
> > >> >  * A daemon with no GTK+ dependance that would be suitable for
> > >> >    cross-desktop use
[...]
> > >> Umm, no.
> > [...]
> > > So in my opinion it's not that difficult, or am I missing something?
> > 
> > Well and there is something like this already in pbbuttons 
[...]
> You can continue to use pbbuttonsd if you like it. We can discuss
> pbbuttonsd if it is proposed for inclusion.

This is supposed to say it is nicely doable *if* necessary. I don't know
whether doing it at this stage for g-p-m is necessary or not. But on the
long run it is definitely beneficial to have a single daemon for all
freedesktop's.

Soeren.

PS: FWIW, I would probably switch to g-p-m and add OSD support if I feel
I like it.

(Continue reading)

headers

Richard Hughes | 9 Apr 2006 20:05

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 19:56 +0200, Jaap Haitsma wrote:
> On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> > On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > > 
> > > > Since the original announcement mail about gnome-power-manager, we have
> > > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > > starting to integrate with other parts of the GNOME application stack.
> > > > Lots of new functionality has been added, and lots of polish has been
> > > > applied. See the screenshots area of my website[5] for some cool
> > > > screenshots of the latest stuff in the 2-15 branch.
> > > 
> > > I would like to see g-p-m fragmented into three parts.
> > > 
> > >  * A daemon with no GTK+ dependance that would be suitable for
> > >    cross-desktop use
> > >  * A capplet (this exists today)
> > >  * A notification area icon (libnotify dependance goes here)
> > 
> > Umm, no.
> > 
> > The IPC between these components would be horrific and over-complicated
> > for no actual gain. KDE are quite happy with their own power management
> > applications, and no KDE developer has ever mentioned to me that they
> > would want such a cross-desktop daemon.
> > 
> Richard,
> 
> 
> As far as I understand the code of GPM splitting up GPM in a "daemon"

(Continue reading)

headers

Jaap Haitsma | 9 Apr 2006 21:12

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

> > Richard,
> > 
> > 
> > As far as I understand the code of GPM splitting up GPM in a "daemon"
> > and a "notication area icon"/applet would not be so hard.
> > 
> > They are pretty independent from each other. 
> > 
> > The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > take appropriate actions etc. If people run the daemon then they get all
> > the power management features. 
> > 
> > The applet/"notification area icon" just needs to watch the batteries
> > (code of the daemon can be reused  )and show the status by changing
> > it's icon and displaying notifications. 
> > 
> > The only message I see that the "daemon" might want to send to the
> > applet is a message that the system is going to suspend/hibernate and
> > that is already something we want to do to notify other apps that the
> > system is going to suspend/sleep and that they need to take appropriate
> > actions if necessary.
> > 
> > So in my opinion it's not that difficult, or am I missing something?
> 
> But what's the point?
> 

1. It's good design to split up parts which are doing different things
( You can also put all your code in one source file, but that's not good
design )

(Continue reading)

headers

David Zeuthen | 10 Apr 2006 00:31

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 21:12 +0200, Jaap Haitsma wrote:
> 1. It's good design to split up parts which are doing different things
> ( You can also put all your code in one source file, but that's not good
> design )

Even though D-BUS indeed makes it easy to do IPC I fail to see why you'd
want to split g-p-m in two processes. Remeber that g-p-m is a lot more
than an icon; it also pops up dialogs from time to time...

> 2. An applet would be much more consistent with how GNOME works at the
> moment. If I want to add something to the panel I just add there by
> doing "Add to panel" and if I want to remove it I choose "Remove from
> panel". GNOME unlike windows luckily doesn't put many stuff
> automagically in the panel 

Suppose g-p-m was an applet instead of using the notification area. Do
you think it's a good idea to let users remove this applet? 

Basically I think the notification area is the right answer for UI
gizmos that directly relate to the state of the computer while applet is
the answer for UI gizmos that doesn't.

    David

headers

Andrew Sobala | 9 Apr 2006 21:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Jaap Haitsma wrote:
>>> Richard,
>>>
>>>
>>> As far as I understand the code of GPM splitting up GPM in a "daemon"
>>> and a "notication area icon"/applet would not be so hard.
>>>
>>> They are pretty independent from each other. 
>>>
>>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
>>> take appropriate actions etc. If people run the daemon then they get all
>>> the power management features. 
>>>
>>> The applet/"notification area icon" just needs to watch the batteries
>>> (code of the daemon can be reused  )and show the status by changing
>>> it's icon and displaying notifications. 
>>>
>>> The only message I see that the "daemon" might want to send to the
>>> applet is a message that the system is going to suspend/hibernate and
>>> that is already something we want to do to notify other apps that the
>>> system is going to suspend/sleep and that they need to take appropriate
>>> actions if necessary.
>>>
>>> So in my opinion it's not that difficult, or am I missing something?
>>>       
>> But what's the point?
>>
>>     
>
> 1. It's good design to split up parts which are doing different things

(Continue reading)

headers

Luis Villa | 9 Apr 2006 22:47

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Andrew Sobala <aes@...> wrote:
> Jaap Haitsma wrote:
> >>> Richard,
> >>>
> >>>
> >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> >>> and a "notication area icon"/applet would not be so hard.
> >>>
> >>> They are pretty independent from each other.
> >>>
> >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> >>> take appropriate actions etc. If people run the daemon then they get all
> >>> the power management features.
> >>>
> >>> The applet/"notification area icon" just needs to watch the batteries
> >>> (code of the daemon can be reused  )and show the status by changing
> >>> it's icon and displaying notifications.
> >>>
> >>> The only message I see that the "daemon" might want to send to the
> >>> applet is a message that the system is going to suspend/hibernate and
> >>> that is already something we want to do to notify other apps that the
> >>> system is going to suspend/sleep and that they need to take appropriate
> >>> actions if necessary.
> >>>
> >>> So in my opinion it's not that difficult, or am I missing something?
> >>>
> >> But what's the point?
> >>
> >>
> >

(Continue reading)

headers

Corey Burger | 9 Apr 2006 22:57

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
> On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > Jaap Haitsma wrote:
> > >>> Richard,
> > >>>
> > >>>
> > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > >>> and a "notication area icon"/applet would not be so hard.
> > >>>
> > >>> They are pretty independent from each other.
> > >>>
> > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > >>> take appropriate actions etc. If people run the daemon then they get all
> > >>> the power management features.
> > >>>
> > >>> The applet/"notification area icon" just needs to watch the batteries
> > >>> (code of the daemon can be reused  )and show the status by changing
> > >>> it's icon and displaying notifications.
> > >>>
> > >>> The only message I see that the "daemon" might want to send to the
> > >>> applet is a message that the system is going to suspend/hibernate and
> > >>> that is already something we want to do to notify other apps that the
> > >>> system is going to suspend/sleep and that they need to take appropriate
> > >>> actions if necessary.
> > >>>
> > >>> So in my opinion it's not that difficult, or am I missing something?
> > >>>
> > >> But what's the point?
> > >>
> > >>

(Continue reading)

headers

Luis Villa | 9 Apr 2006 23:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Corey Burger <corey.burger <at> gmail.com> wrote:
> On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
> > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > Jaap Haitsma wrote:
> > > >>> Richard,
> > > >>>
> > > >>>
> > > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > > >>> and a "notication area icon"/applet would not be so hard.
> > > >>>
> > > >>> They are pretty independent from each other.
> > > >>>
> > > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > > >>> take appropriate actions etc. If people run the daemon then they get all
> > > >>> the power management features.
> > > >>>
> > > >>> The applet/"notification area icon" just needs to watch the batteries
> > > >>> (code of the daemon can be reused  )and show the status by changing
> > > >>> it's icon and displaying notifications.
> > > >>>
> > > >>> The only message I see that the "daemon" might want to send to the
> > > >>> applet is a message that the system is going to suspend/hibernate and
> > > >>> that is already something we want to do to notify other apps that the
> > > >>> system is going to suspend/sleep and that they need to take appropriate
> > > >>> actions if necessary.
> > > >>>
> > > >>> So in my opinion it's not that difficult, or am I missing something?
> > > >>>
> > > >> But what's the point?
> > > >>

(Continue reading)

headers

intech | 12 Apr 2006 06:24

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
> Wrong. It only matters when you're getting so low you are in danger of
> losing work, or when the status changes, or in a couple other corner
> cases which can be designed for. It is *not* the most important thing-
> the most important thing is whatever work I'm actually *doing*.

Its not the most important thing, but it is something of some
importance. If it was the most important thing, it should intrude on
your current work and get more in your face. A simple notification
icon doesn't mean it requires your urgent attention, just that it is
something to worry about at some point in the future. If you are
plugged into the wall and your battery is full, nothing is happening
at all. Nothing that you should be aware of at all.

> I strongly recommend reading 'Designing From Both Sides of the
> Screen', where one of the simple design heuristics is to make software
> that acts like a butler (or in this case, a chauffeur.) As you drive
> around town, does your chauffeur say 'by the way sir, the gas tank is
> now 59% full.' (minutes pass) 'oh, now 58% full sir.' No. If your
> chauffeur did that, you'd fire him for being an irritating idiot. A
> good chauffeur tells you 'Sir, the tank is very nearly empty- shall I
> find a station?', and a great chauffeur asks you once 'how early would
> you like me to warn you about the gas, sir?' and then remembers that
> in the future. When you pull the plug out of the wall, I mean, when
> you come upon the sign that says 'huge desert- no gas for a long way',
> a good chauffeur says 'Sir, we only have enough gas for 299 miles at
> current consumption- would you like me to turn around?'
I think taking it as a more similiar example, it would be the gas gage
in a car. The way most that I've seen work is that it has the
indicator that is always there, and then has an additional light/beep

(Continue reading)

headers

Elijah Newren | 10 Apr 2006 01:22

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
> On 4/9/06, Corey Burger <corey.burger <at> gmail.com> wrote:

> > I completely disagree. There are a few good reasons why an icon should
> > be displayed all the time
> >
> > 1. What state the battery is in is always relevant. Power is the
> > single most important thing on a laptop. Without it, you are going
> > nowhere.
>
> Wrong. It only matters when you're getting so low you are in danger of
> losing work, or when the status changes, or in a couple other corner
> cases which can be designed for.

Wrong.   It matters unless the battery is fully charged and the
laptop is plugged in.  IMNSHO anyway.

> It is *not* the most important thing- the most important thing is whatever work
> I'm actually *doing*.

Agreed.

> I strongly recommend reading 'Designing From Both Sides of the
> Screen', where one of the simple design heuristics is to make software
> that acts like a butler (or in this case, a chauffeur.) As you drive
> around town, does your chauffeur say 'by the way sir, the gas tank is
> now 59% full.' (minutes pass) 'oh, now 58% full sir.' No. If your
> chauffeur did that, you'd fire him for being an irritating idiot. A
> good chauffeur tells you 'Sir, the tank is very nearly empty- shall I
> find a station?', and a great chauffeur asks you once 'how early would

(Continue reading)

headers

Elijah Newren | 10 Apr 2006 01:22

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Luis Villa <luis.villa@...> wrote:
> On 4/9/06, Corey Burger <corey.burger@...> wrote:

> > I completely disagree. There are a few good reasons why an icon should
> > be displayed all the time
> >
> > 1. What state the battery is in is always relevant. Power is the
> > single most important thing on a laptop. Without it, you are going
> > nowhere.
>
> Wrong. It only matters when you're getting so low you are in danger of
> losing work, or when the status changes, or in a couple other corner
> cases which can be designed for.

Wrong.   It matters unless the battery is fully charged and the
laptop is plugged in.  IMNSHO anyway.

> It is *not* the most important thing- the most important thing is whatever work
> I'm actually *doing*.

Agreed.

> I strongly recommend reading 'Designing From Both Sides of the
> Screen', where one of the simple design heuristics is to make software
> that acts like a butler (or in this case, a chauffeur.) As you drive
> around town, does your chauffeur say 'by the way sir, the gas tank is
> now 59% full.' (minutes pass) 'oh, now 58% full sir.' No. If your
> chauffeur did that, you'd fire him for being an irritating idiot. A
> good chauffeur tells you 'Sir, the tank is very nearly empty- shall I
> find a station?', and a great chauffeur asks you once 'how early would

(Continue reading)

headers

Andrew Sobala | 9 Apr 2006 23:08

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Corey Burger wrote:
> On 4/9/06, Luis Villa <luis.villa@...> wrote:
>   
>> On 4/9/06, Andrew Sobala <aes@...> wrote:
>>     
>>> It's worth pointing out that gnome-power-manager is very much a notifier
>>> rather than an interactive applet. If your power cable falls out, it
>>> pops up a message saying you've lost power. If you're working away from
>>> a power source, there's a battery indicator with how much power you've
>>> got left... that disappears when you're fully charged.
>>>
>>> (At least, that's how it's configured on my system.)
>>>       
>> This isn't the default, FWIW. I do agree that making this the default
>> behavior  would be the best approach- better, IMHO, than a regular
>> panel applet. I only want to know about power when something bad is
>> going wrong, which is exactly what the notification area is for. An
>> applet is all the time, and so is the current default behavior in the
>> notification area- both of which are broken.
>>
>> Luis
>>     
>
> I completely disagree. There are a few good reasons why an icon should
> be displayed all the time
>
> 1. What state the battery is in is always relevant. Power is the
> single most important thing on a laptop. Without it, you are going
> nowhere. Whether or not it is a notification icon or an applet is a
> detail I won't comment on.

(Continue reading)

headers

Rodrigo Moya | 10 Apr 2006 11:50

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 22:08 +0100, Andrew Sobala wrote:
> Corey Burger wrote:
> > On 4/9/06, Luis Villa <luis.villa@...> wrote:
> >   
> >> On 4/9/06, Andrew Sobala <aes@...> wrote:
> >>     
> >>> It's worth pointing out that gnome-power-manager is very much a notifier
> >>> rather than an interactive applet. If your power cable falls out, it
> >>> pops up a message saying you've lost power. If you're working away from
> >>> a power source, there's a battery indicator with how much power you've
> >>> got left... that disappears when you're fully charged.
> >>>
> >>> (At least, that's how it's configured on my system.)
> >>>       
> >> This isn't the default, FWIW. I do agree that making this the default
> >> behavior  would be the best approach- better, IMHO, than a regular
> >> panel applet. I only want to know about power when something bad is
> >> going wrong, which is exactly what the notification area is for. An
> >> applet is all the time, and so is the current default behavior in the
> >> notification area- both of which are broken.
> >>
> >> Luis
> >>     
> >
> > I completely disagree. There are a few good reasons why an icon should
> > be displayed all the time
> >
> > 1. What state the battery is in is always relevant. Power is the
> > single most important thing on a laptop. Without it, you are going
> > nowhere. Whether or not it is a notification icon or an applet is a

(Continue reading)

headers

Andrew Sobala | 10 Apr 2006 19:32

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Rodrigo Moya wrote:
> On Sun, 2006-04-09 at 22:08 +0100, Andrew Sobala wrote:
>   
>> Corey Burger wrote:
>>     
>>> On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
>>>   
>>>       
>>>> On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
>>>>     
>>>>         
>>>>> It's worth pointing out that gnome-power-manager is very much a notifier
>>>>> rather than an interactive applet. If your power cable falls out, it
>>>>> pops up a message saying you've lost power. If you're working away from
>>>>> a power source, there's a battery indicator with how much power you've
>>>>> got left... that disappears when you're fully charged.
>>>>>
>>>>> (At least, that's how it's configured on my system.)
>>>>>       
>>>>>           
>>>> This isn't the default, FWIW. I do agree that making this the default
>>>> behavior  would be the best approach- better, IMHO, than a regular
>>>> panel applet. I only want to know about power when something bad is
>>>> going wrong, which is exactly what the notification area is for. An
>>>> applet is all the time, and so is the current default behavior in the
>>>> notification area- both of which are broken.
>>>>
>>>> Luis
>>>>     
>>>>

(Continue reading)

headers

Davyd Madeley | 10 Apr 2006 14:10

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 11:50 +0200, Rodrigo Moya wrote:

> I don't understand why you don't care. Usually, AFAIK, batteries have a
> longer life if you charge them completely and then discharge them
> completely, so at least from my experience, you really care when it's
> fully charged so that you can unplug it from AC and start discharging.

This is true of NiCd and NiMH batteries. It is not true of Li-ION and
LiS batteries, which do not develop "memory" in the same way. I've seen
it recommended that such batteries instead be used in smaller bursts.
Completely discharging and then recharging Li-ION cells repeatedly can
actually cause them to flip polarity and "die".

--d

--

-- 
Davyd Madeley

http://www.davyd.id.au/
08B0 341A 0B9B 08BB 2118  C060 2EDD BB4F 5191 6CDA

headers

Andrew Sobala | 10 Apr 2006 19:32

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Rodrigo Moya wrote:
> On Sun, 2006-04-09 at 22:08 +0100, Andrew Sobala wrote:
>   
>> Corey Burger wrote:
>>     
>>> On 4/9/06, Luis Villa <luis.villa@...> wrote:
>>>   
>>>       
>>>> On 4/9/06, Andrew Sobala <aes@...> wrote:
>>>>     
>>>>         
>>>>> It's worth pointing out that gnome-power-manager is very much a notifier
>>>>> rather than an interactive applet. If your power cable falls out, it
>>>>> pops up a message saying you've lost power. If you're working away from
>>>>> a power source, there's a battery indicator with how much power you've
>>>>> got left... that disappears when you're fully charged.
>>>>>
>>>>> (At least, that's how it's configured on my system.)
>>>>>       
>>>>>           
>>>> This isn't the default, FWIW. I do agree that making this the default
>>>> behavior  would be the best approach- better, IMHO, than a regular
>>>> panel applet. I only want to know about power when something bad is
>>>> going wrong, which is exactly what the notification area is for. An
>>>> applet is all the time, and so is the current default behavior in the
>>>> notification area- both of which are broken.
>>>>
>>>> Luis
>>>>     
>>>>

(Continue reading)

headers

Vincent Untz | 10 Apr 2006 08:37

Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

[I killed most of the mail addresses in the cc line]

Le dimanche 09 avril 2006 à 22:08 +0100, Andrew Sobala a écrit :
> As a sidenote, I believe the windows slider was invented to leave some 
> room for the task bar when you have 40 icons in your notification area, 
> one for every application installed on the system. The GNOME 
> notification area isn't intended to be (and for the most part, isn't) 
> used in this way, so we don't need a way to hide icons that shouldn't be
> there in the first place.

Just a sidenote about notification icons: right now, I have 5 of them:

 + the alarm of evolution (I kind of agree it should be here, although
   I didn't click on it and it's now sitting there since at least 24
   hours...)

 + the gossip icon. It just sits there, while I have no message. It
   should be an applet.

 + the battery icon of gnome-power-manager. Because I like to know if
   my battery is fully charged or not, etc. Should be an applet. I
   should mention I'm insane since I also have the battstat applet 

 + the xchat-gnome notification plugin. The last version has a fix that
   make it only show an icon when there's a message for you. Much saner.

 + the reboot icon of ubuntu. I don't want to click on this, it will
   reboot! (well, it won't: it will just ask me if I want to reboot,
   but I already mentioned I'm insane)

(Continue reading)

headers

Rodney Dawes | 10 Apr 2006 18:21

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
>  + the gossip icon. It just sits there, while I have no message. It
>    should be an applet.

Making it an applet doesn't solve the problem. It's still the same icon
sitting on your panel, taking up space, doing absolutely nothing for
you. In fact, making it an applet would be a regression, as it would no
longer work under other desktops as well.

> There are some ways to fix this:
> 
>  + HIG HIG HIG

The HIG is a set of guidelines, not a set of rules that must be abided
for an application to use the tray. Putting something here doesn't
really solve the problem outside the scope of the core GNOME apps. It
just gives the people who will whine about the problem, someplace to
point at, while they are whining.

>  + try to make the notification area smart and force some icons to hide.
>    This will be inherently broken.

We should make it smart and allow expansion, and hide inactive and low
priority icons, like on Windows.

-- dobey

headers

Steve Frécinaux | 10 Apr 2006 19:36

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Rodney Dawes wrote:
> On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
>>  + the gossip icon. It just sits there, while I have no message. It
>>    should be an applet.
> 
> Making it an applet doesn't solve the problem. It's still the same icon
> sitting on your panel, taking up space, doing absolutely nothing for
> you. In fact, making it an applet would be a regression, as it would no
> longer work under other desktops as well.

But it would allow me to put the icon wherever I want to put it, and not 
in the notification area.

headers

Rodney Dawes | 10 Apr 2006 22:32

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 19:36 +0200, Steve Frécinaux wrote:
> Rodney Dawes wrote:
> > On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
> >>  + the gossip icon. It just sits there, while I have no message. It
> >>    should be an applet.
> > 
> > Making it an applet doesn't solve the problem. It's still the same icon
> > sitting on your panel, taking up space, doing absolutely nothing for
> > you. In fact, making it an applet would be a regression, as it would no
> > longer work under other desktops as well.
> 
> But it would allow me to put the icon wherever I want to put it, and not 
> in the notification area.

This is a configuration problem, not a user interface problem. Just
because you might want to do that, does not mean that all other users
will want to as well. In fact, if you want to say that, we can talk
about the majority of the Desktop market, and show how useful it is to
have them in the tray, because all of the IM clients are there in the
same area, so it makes it easy to use them all. And useful status
information is there too. You don't have to rove all around your desktop
looking for information, as it's all always in the same location.

But that's just me (and another 85% of the market).

-- dobey

headers

Steve Frécinaux | 10 Apr 2006 23:13

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Rodney Dawes wrote:
> This is a configuration problem, not a user interface problem. Just
> because you might want to do that, does not mean that all other users
> will want to as well. In fact, if you want to say that, we can talk
> about the majority of the Desktop market, and show how useful it is to
> have them in the tray, because all of the IM clients are there in the
> same area, so it makes it easy to use them all. And useful status
> information is there too. You don't have to rove all around your desktop
> looking for information, as it's all always in the same location.
> 
> But that's just me (and another 85% of the market).

I really, really don't care about the market. I'm not a carpet salesman.

About our current subject, I like my desktop being well organized. And 
as all my status monitors (I'm using a laptop, so I'm talking about 
network interfaces status and frequency monitor) sit on a part of my 
panel, I don't see any good reason why the battery monitor should be on 
a separate place, in the notification area, with no way for me to put it 
among the other ones.

And what happen is just the opposite of what you describe : that status 
icon is not in the right place, so I have to look around my desktop to 
find it. Worse, it's messed up with app icons (which I'm quite ok they 
stay in the notification area) and other real notification icons...

headers

Alex | 11 Apr 2006 13:21

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Sorry to enter the discussion that late but there were a few things
mentioned that really, really disturbed be.

Steve Frécinaux wrote:
> Rodney Dawes wrote:
>> This is a configuration problem, not a user interface problem. Just
>> because you might want to do that, does not mean that all other users
>> will want to as well. In fact, if you want to say that, we can talk
>> about the majority of the Desktop market, and show how useful it is to
>> have them in the tray, because all of the IM clients are there in the
>> same area, so it makes it easy to use them all. And useful status
>> information is there too. You don't have to rove all around your desktop
>> looking for information, as it's all always in the same location.
>>
>> But that's just me (and another 85% of the market).
> 
> I really, really don't care about the market. I'm not a carpet salesman.

Finally someone who said it. Please, stop talking about "the user", the
"majority of all users out there" or similar things.

In fact you all have an opinion - mine would be that I'd like to see the
old battstat "replaced" by gpm - but that's only my opinion. I would
never ever dare to _guess_ whate veryone else wants. This is exactly
what a few of you do (I don't want to mention names now).

The only real problem that Gnome-d-e ever had were developers and
"semi-developers" talking about what I want without really knowing that.
Finally there is someone who tries to satisfy /nearly/ everyone - may I
quote gnome-power-settings: "Never display icon" [...] "Always display

(Continue reading)

headers

Steve Frécinaux | 10 Apr 2006 23:13

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Rodney Dawes wrote:
> This is a configuration problem, not a user interface problem. Just
> because you might want to do that, does not mean that all other users
> will want to as well. In fact, if you want to say that, we can talk
> about the majority of the Desktop market, and show how useful it is to
> have them in the tray, because all of the IM clients are there in the
> same area, so it makes it easy to use them all. And useful status
> information is there too. You don't have to rove all around your desktop
> looking for information, as it's all always in the same location.
> 
> But that's just me (and another 85% of the market).

I really, really don't care about the market. I'm not a carpet salesman.

About our current subject, I like my desktop being well organized. And 
as all my status monitors (I'm using a laptop, so I'm talking about 
network interfaces status and frequency monitor) sit on a part of my 
panel, I don't see any good reason why the battery monitor should be on 
a separate place, in the notification area, with no way for me to put it 
among the other ones.

And what happen is just the opposite of what you describe : that status 
icon is not in the right place, so I have to look around my desktop to 
find it. Worse, it's messed up with app icons (which I'm quite ok they 
stay in the notification area) and other real notification icons...

headers

Matthew Garrett | 10 Apr 2006 18:35

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, Apr 10, 2006 at 12:21:58PM -0400, Rodney Dawes wrote:

> We should make it smart and allow expansion, and hide inactive and low
> priority icons, like on Windows.

If we can hide icons without losing important information, why are we 
showing them in the first place?
--

-- 
Matthew Garrett | mjg59 <at> srcf.ucam.org

headers

Rodney Dawes | 10 Apr 2006 22:29

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 17:35 +0100, Matthew Garrett wrote:
> On Mon, Apr 10, 2006 at 12:21:58PM -0400, Rodney Dawes wrote:
> 
> > We should make it smart and allow expansion, and hide inactive and low
> > priority icons, like on Windows.
> 
> If we can hide icons without losing important information, why are we 
> showing them in the first place?

This is not the correct question to the problem. *WE* aren't showing
them. However, *WE* also don't control all the win32 apps that one can
run, that puts an icon in the tray, or other apps which are not part of
GNOME itself, and need to show an icon on all desktops for some reason.

Also, it's quite annoying to have icons appearing and disappearing
constantly, because something happened, and then 3 seconds later the
icon is gone. And doing so, shuffles the icons around, making the ones
the user does care about, moving targets in some cases.

-- dobey

headers

Iain * | 11 Apr 2006 01:04

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On 4/10/06, Rodney Dawes <dobey <at> novell.com> wrote:
> And doing so, shuffles the icons around, making the ones
> the user does care about, moving targets in some cases.

See my proposal earlier.

headers

Iain * | 11 Apr 2006 01:04

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On 4/10/06, Rodney Dawes <dobey@...> wrote:
> And doing so, shuffles the icons around, making the ones
> the user does care about, moving targets in some cases.

See my proposal earlier.

headers

Steve Frécinaux | 10 Apr 2006 19:36

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Rodney Dawes wrote:
> On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
>>  + the gossip icon. It just sits there, while I have no message. It
>>    should be an applet.
> 
> Making it an applet doesn't solve the problem. It's still the same icon
> sitting on your panel, taking up space, doing absolutely nothing for
> you. In fact, making it an applet would be a regression, as it would no
> longer work under other desktops as well.

But it would allow me to put the icon wherever I want to put it, and not 
in the notification area.

headers

Matthew Garrett | 10 Apr 2006 18:35

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, Apr 10, 2006 at 12:21:58PM -0400, Rodney Dawes wrote:

> We should make it smart and allow expansion, and hide inactive and low
> priority icons, like on Windows.

If we can hide icons without losing important information, why are we 
showing them in the first place?
--

-- 
Matthew Garrett | mjg59@...

headers

Stanislav Brabec | 10 Apr 2006 12:58

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Vincent Untz writes:

>  + HIG HIG HIG

[Bug 99175] Need recommendations for notification area.
HIG | General | Ver: unspecified
http://bugzilla.gnome.org/show_bug.cgi?id=99175

--

-- 
Best Regards / S pozdravem,

Stanislav Brabec
software developer
---------------------------------------------------------------------
SuSE CR, s. r. o.                             e-mail: sbrabec <at> suse.cz
Drahobejlova 27                               tel: +420 296 542 382
190 00 Praha 9                                fax: +420 296 542 374
Czech Republic                                http://www.suse.cz/

headers

Rodrigo Moya | 10 Apr 2006 11:53

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
> Well, I really should have 6, but I'm not listening music right now
> (rhythmbox). And it could be 7, but I'm not connected with ekiga right
> now. Or 8 with NetworkManager. And even 9 since I use the keyboard
> typing break feature, but the icon disappeared???
> 
oh, that's a bug, please file it :)
--

-- 
Rodrigo Moya <rodrigo@...>

headers

Iain * | 10 Apr 2006 11:28

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On 4/10/06, Vincent Untz <vuntz <at> gnome.org> wrote:

> There are some ways to fix this:
>
>  + HIG HIG HIG
>
>  + make it possible to dynamically add an applet from a program. I'd
>    like to add the infrastructure for this during 2.16. Don't know if
>    I'll have time, but maybe someone is interested in working on
>    this? 
>
>  + try to make the notification area smart and force some icons to hide.
>    This will be inherently broken.

There are two different ways of looking at the icons
+ Icon is application and any windows created belong to the icon
    The program doesn't quit when you close windows. Gaim and Tomboy are
    examples
+ Icon is used for notification and is owned by the program. If the
program window
   is closed, the icon never appears again.

I think we need to accept that both of these are going to be used
whether we like it or not, and handle them both. Currently we only
"allow" the second way, and bitch about the first, but it is clear
that applications want the first one, and there are many situations
that the first is useful.

headers

Vincent Untz | 10 Apr 2006 13:18

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)


On Mon, April 10, 2006 11:28, Iain * wrote:
> On 4/10/06, Vincent Untz <vuntz <at> gnome.org> wrote:
>
>> There are some ways to fix this:
>>
>>  + HIG HIG HIG
>>
>>  + make it possible to dynamically add an applet from a program. I'd
>>    like to add the infrastructure for this during 2.16. Don't know if
>>    I'll have time, but maybe someone is interested in working on
>>    this? 
>>
>>  + try to make the notification area smart and force some icons to hide.
>>    This will be inherently broken.
>
> There are two different ways of looking at the icons
> + Icon is application and any windows created belong to the icon
>     The program doesn't quit when you close windows. Gaim and Tomboy are
>     examples
> + Icon is used for notification and is owned by the program. If the
> program window
>    is closed, the icon never appears again.
>
> I think we need to accept that both of these are going to be used
> whether we like it or not, and handle them both. Currently we only
> "allow" the second way, and bitch about the first, but it is clear
> that applications want the first one, and there are many situations
> that the first is useful.

(Continue reading)

headers

Sriram Ramkrishna | 10 Apr 2006 19:45

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

> You missed my second point. If the icon is the application, then we
> should provide an easy way for the application to use an applet
> instead of the notification area.

Sort of like "minimize to applet" or something?  That's not going to
work well until applets and application use the same main loop.  We
really need to move towards merging those together so we can do
a lot more interesting things with the desktop.

sri

headers

Iain * | 10 Apr 2006 15:31

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

> You missed my second point. If the icon is the application, then we
> should provide an easy way for the application to use an applet
> instead of the notification area.

Except they want it so that they work in KDE as well. That was the
original reason for the first offender (the redhat updater icon) to
use the notification area instead of an applet.

But I don't see that use of the notification area for this sort of
thing as "bad" persay. We just need to be able to handle it well.

iain

headers

Joe Shaw | 10 Apr 2006 21:03

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Hi,

On Mon, 2006-04-10 at 14:31 +0100, Iain * wrote:
> > You missed my second point. If the icon is the application, then we
> > should provide an easy way for the application to use an applet
> > instead of the notification area.
> 
> Except they want it so that they work in KDE as well. That was the
> original reason for the first offender (the redhat updater icon) to
> use the notification area instead of an applet.

I feel like the importance of this is understated.

Developers writing software for the "Linux desktop" will ignore
desktop-specific interface guidelines if they have to to reach both
desktops or achieve missing functionality.  All the panel-dockable
dealies I write (which have included Netapplet and now Beagle) are
notification area icons and not panel applets for this reason.

It's fine to debate the finer points of our UI guidelines with respect
to the notification area, but no one outside of GNOME is going to follow
them if it's the only means to put them on both desktops.

Joe

headers

Iain * | 10 Apr 2006 11:35

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

> There are two different ways of looking at the icons
> + Icon is application and any windows created belong to the icon
>     The program doesn't quit when you close windows. Gaim and Tomboy are
>     examples
> + Icon is used for notification and is owned by the program. If the
> program window
>    is closed, the icon never appears again.
>
> I think we need to accept that both of these are going to be used
> whether we like it or not, and handle them both. Currently we only
> "allow" the second way, and bitch about the first, but it is clear
> that applications want the first one, and there are many situations
> that the first is useful.

It might be useful to allow apps to define which category they are in,
and then their location when being added is changed.
With the notification area at the right hand side of panel

          [a][2][2][2][b][1][1][1][1]

[1] is an icon using case 1
[2] is an icon using case 2
[a] is where icons using case 2 are added because they appear and
disappear all the time, we don't want them to adjust the location of
the [1] icons
[b] is where icons using case 1 are added, because they are there for
a long time and people want to find them, so it'd be nice if they were
in a more permanent position and can't be moved around by the [2]s

<talking-crap>

(Continue reading)

headers

Emmanuele Bassi | 10 Apr 2006 11:34

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 10:28 +0100, Iain * wrote:

> >  + try to make the notification area smart and force some icons to hide.
> >    This will be inherently broken.
> 
> There are two different ways of looking at the icons
> + Icon is application and any windows created belong to the icon
>     The program doesn't quit when you close windows. Gaim and Tomboy are
>     examples

Tomboy also has a proper applet (which I'm using right now).

Gaim's icon is similar to the NetworkManager: it shows the state of the
connection, and changes when there's an "event"; one could also argue
that the 'Buddy List' is hardly Gaim's main window (I practically never
use it).

Ciao,
 Emmanuele

--

-- 
Emmanuele Bassi - <ebassi <at> gmail.com>
Log: http://log.emmanuelebassi.net

headers

Rodney Dawes | 10 Apr 2006 18:21

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

On Mon, 2006-04-10 at 08:37 +0200, Vincent Untz wrote:
>  + the gossip icon. It just sits there, while I have no message. It
>    should be an applet.

Making it an applet doesn't solve the problem. It's still the same icon
sitting on your panel, taking up space, doing absolutely nothing for
you. In fact, making it an applet would be a regression, as it would no
longer work under other desktops as well.

> There are some ways to fix this:
> 
>  + HIG HIG HIG

The HIG is a set of guidelines, not a set of rules that must be abided
for an application to use the tray. Putting something here doesn't
really solve the problem outside the scope of the core GNOME apps. It
just gives the people who will whine about the problem, someplace to
point at, while they are whining.

>  + try to make the notification area smart and force some icons to hide.
>    This will be inherently broken.

We should make it smart and allow expansion, and hide inactive and low
priority icons, like on Windows.

-- dobey

headers

Stanislav Brabec | 10 Apr 2006 12:58

Re: Notification icons hell (was Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager)

Vincent Untz writes:

>  + HIG HIG HIG

[Bug 99175] Need recommendations for notification area.
HIG | General | Ver: unspecified
http://bugzilla.gnome.org/show_bug.cgi?id=99175

--

-- 
Best Regards / S pozdravem,

Stanislav Brabec
software developer
---------------------------------------------------------------------
SuSE CR, s. r. o.                             e-mail: sbrabec@...
Drahobejlova 27                               tel: +420 296 542 382
190 00 Praha 9                                fax: +420 296 542 374
Czech Republic                                http://www.suse.cz/

headers

Scott J. Harmon | 9 Apr 2006 23:53

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Andrew Sobala wrote:
> Corey Burger wrote:
>> On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
>>  
>>> On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
>>>    
>>>> It's worth pointing out that gnome-power-manager is very much a
>>>> notifier
>>>> rather than an interactive applet. If your power cable falls out, it
>>>> pops up a message saying you've lost power. If you're working away from
>>>> a power source, there's a battery indicator with how much power you've
>>>> got left... that disappears when you're fully charged.
>>>>
>>>> (At least, that's how it's configured on my system.)
>>>>       
>>> This isn't the default, FWIW. I do agree that making this the default
>>> behavior  would be the best approach- better, IMHO, than a regular
>>> panel applet. I only want to know about power when something bad is
>>> going wrong, which is exactly what the notification area is for. An
>>> applet is all the time, and so is the current default behavior in the
>>> notification area- both of which are broken.
>>>
>>> Luis
>>>     
>>
>> I completely disagree. There are a few good reasons why an icon should
>> be displayed all the time
>>
>> 1. What state the battery is in is always relevant. Power is the
>> single most important thing on a laptop. Without it, you are going

(Continue reading)

headers

Paul Drain | 10 Apr 2006 03:38

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager


> > Nope. I'm working on a laptop at the moment, and I don't care that my
> > battery is fully charged. This is because it's plugged into the wall. If
> > I wasn't plugged into the wall, I'd start caring - but I'd also get a
> > battery symbol.
> 
> Am I the only one who mouses over the applet to see how much more time
> until the battery is fully charged?

Certainly not.

I'd be right miffed if this particular functionality wasn't available by
default on a laptop anymore -- especially when you're at a conference,
or some other 'on-AC-but-time-limited-by-other-commitments' event and
you *need* to make sure you have enough battery life available to
actually perform a demo.

I also might care about the icon, if say -- i've got a problem with ACPI
not detecting my battery -- yeah, the laptop's OK when you're plugged
into the wall (which I am now), but *all your work goes boom!* when you
disconnect the power.

Using Luis' 'chauffeur scenario' -- the latter case is kind of like your
driver turning around to you and saying, 'sir, we've left the gas tank
on the road, about ten miles back.'

Most Annoying.

Paul

(Continue reading)

headers

Elijah Newren | 10 Apr 2006 01:14

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> Andrew Sobala wrote:
> > Nope. I'm working on a laptop at the moment, and I don't care that my
> > battery is fully charged. This is because it's plugged into the wall. If
> > I wasn't plugged into the wall, I'd start caring - but I'd also get a
> > battery symbol.
>
> Am I the only one who mouses over the applet to see how much more time
> until the battery is fully charged?

Definitely not; I rely on this frequently.  I'd be heavily annoyed if
the applet wasn't showing (and no other equally easy way of obtaining
this information was available) when my laptop is plugged in and not
fully charged.  If it's both plugged in and fully charged then I'd be
fine with it not being there, as long as that was the only case.

headers

Andrew Sobala | 10 Apr 2006 01:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Elijah Newren wrote:
> On 4/9/06, Scott J. Harmon <harmon@...> wrote:
>   
>> Am I the only one who mouses over the applet to see how much more time
>> until the battery is fully charged?
>>     
>
> Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> the applet wasn't showing (and no other equally easy way of obtaining
> this information was available) when my laptop is plugged in and not
> fully charged.  If it's both plugged in and fully charged then I'd be
> fine with it not being there, as long as that was the only case.
>   
Hmm. In this configuration, it is.

--
Andrew

headers

Corey Burger | 10 Apr 2006 01:28

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> Elijah Newren wrote:
> > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> >
> >> Am I the only one who mouses over the applet to see how much more time
> >> until the battery is fully charged?
> >>
> >
> > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > the applet wasn't showing (and no other equally easy way of obtaining
> > this information was available) when my laptop is plugged in and not
> > fully charged.  If it's both plugged in and fully charged then I'd be
> > fine with it not being there, as long as that was the only case.
> >
> Hmm. In this configuration, it is.

The problem with hiding in this case is that the user must know that
when they are plugged in and fully charged, the icon will vanish,
rather than just looking and seeing that they are fully charged and
plugged in. Ouch.

Corey

headers

Xavier Bestel | 10 Apr 2006 11:59

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 01:28, Corey Burger wrote:
> On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > Elijah Newren wrote:
> > > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> > >
> > >> Am I the only one who mouses over the applet to see how much more time
> > >> until the battery is fully charged?
> > >>
> > >
> > > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > > the applet wasn't showing (and no other equally easy way of obtaining
> > > this information was available) when my laptop is plugged in and not
> > > fully charged.  If it's both plugged in and fully charged then I'd be
> > > fine with it not being there, as long as that was the only case.
> > >
> > Hmm. In this configuration, it is.
> 
> The problem with hiding in this case is that the user must know that
> when they are plugged in and fully charged, the icon will vanish,
> rather than just looking and seeing that they are fully charged and
> plugged in. Ouch.

Well, if once the battery reaches 100%, a short-lived notification
bubble says so then the icon can disappear without harm. No need to
pollute the notification area with a "battery is full" icon, OTOH on the
road the fuel gauge is important.

IMHO the best design is found in PocketPC2003: the battery icon starts
to appear only when it's half-empty.

(Continue reading)

headers

Christian Fredrik Kalager Schaller | 10 Apr 2006 12:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 11:59 +0200, Xavier Bestel wrote:
> On Mon, 2006-04-10 at 01:28, Corey Burger wrote:
> > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > Elijah Newren wrote:
> > > > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> > > >
> > > >> Am I the only one who mouses over the applet to see how much more time
> > > >> until the battery is fully charged?
> > > >>
> > > >
> > > > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > > > the applet wasn't showing (and no other equally easy way of obtaining
> > > > this information was available) when my laptop is plugged in and not
> > > > fully charged.  If it's both plugged in and fully charged then I'd be
> > > > fine with it not being there, as long as that was the only case.
> > > >
> > > Hmm. In this configuration, it is.
> > 
> > The problem with hiding in this case is that the user must know that
> > when they are plugged in and fully charged, the icon will vanish,
> > rather than just looking and seeing that they are fully charged and
> > plugged in. Ouch.
> 
> Well, if once the battery reaches 100%, a short-lived notification
> bubble says so then the icon can disappear without harm. No need to
> pollute the notification area with a "battery is full" icon, OTOH on the
> road the fuel gauge is important.
> 
> IMHO the best design is found in PocketPC2003: the battery icon starts
> to appear only when it's half-empty.

(Continue reading)

headers

Rodrigo Moya | 10 Apr 2006 13:22

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 12:25 +0200, Christian Fredrik Kalager Schaller
wrote:
> On Mon, 2006-04-10 at 11:59 +0200, Xavier Bestel wrote:
> > On Mon, 2006-04-10 at 01:28, Corey Burger wrote:
> > > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > > Elijah Newren wrote:
> > > > > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> > > > >
> > > > >> Am I the only one who mouses over the applet to see how much more time
> > > > >> until the battery is fully charged?
> > > > >>
> > > > >
> > > > > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > > > > the applet wasn't showing (and no other equally easy way of obtaining
> > > > > this information was available) when my laptop is plugged in and not
> > > > > fully charged.  If it's both plugged in and fully charged then I'd be
> > > > > fine with it not being there, as long as that was the only case.
> > > > >
> > > > Hmm. In this configuration, it is.
> > > 
> > > The problem with hiding in this case is that the user must know that
> > > when they are plugged in and fully charged, the icon will vanish,
> > > rather than just looking and seeing that they are fully charged and
> > > plugged in. Ouch.
> > 
> > Well, if once the battery reaches 100%, a short-lived notification
> > bubble says so then the icon can disappear without harm. No need to
> > pollute the notification area with a "battery is full" icon, OTOH on the
> > road the fuel gauge is important.
> >

(Continue reading)

headers

Jeroen Zwartepoorte | 10 Apr 2006 13:29

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

I'm curious: what role does the powersave daemon fulfill? I'm running
FC5 and afaik, HAL/g-p-m gets it information directly from the OS
(ACPI)?

Regards,

Jeroen

On 4/10/06, Rodrigo Moya <rodrigo <at> gnome-db.org> wrote:
> On Mon, 2006-04-10 at 12:25 +0200, Christian Fredrik Kalager Schaller
> wrote:
> > On Mon, 2006-04-10 at 11:59 +0200, Xavier Bestel wrote:
> > > On Mon, 2006-04-10 at 01:28, Corey Burger wrote:
> > > > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > > > Elijah Newren wrote:
> > > > > > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> > > > > >
> > > > > >> Am I the only one who mouses over the applet to see how much more time
> > > > > >> until the battery is fully charged?
> > > > > >>
> > > > > >
> > > > > > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > > > > > the applet wasn't showing (and no other equally easy way of obtaining
> > > > > > this information was available) when my laptop is plugged in and not
> > > > > > fully charged.  If it's both plugged in and fully charged then I'd be
> > > > > > fine with it not being there, as long as that was the only case.
> > > > > >
> > > > > Hmm. In this configuration, it is.
> > > >
> > > > The problem with hiding in this case is that the user must know that

(Continue reading)

headers

Rodrigo Moya | 10 Apr 2006 14:33

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 13:29 +0200, Jeroen Zwartepoorte wrote:
> I'm curious: what role does the powersave daemon fulfill? I'm running
> FC5 and afaik, HAL/g-p-m gets it information directly from the OS
> (ACPI)?
> 
HAL scripts (in $prefix/libexec/hal iirc) use the underlying daemon's
commands to suspend/hibernate/set power save profile.
--

-- 
Rodrigo Moya <rodrigo <at> gnome-db.org>

headers

David Zeuthen | 11 Apr 2006 18:05

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 14:33 +0200, Rodrigo Moya wrote:
> On Mon, 2006-04-10 at 13:29 +0200, Jeroen Zwartepoorte wrote:
> > I'm curious: what role does the powersave daemon fulfill? I'm running
> > FC5 and afaik, HAL/g-p-m gets it information directly from the OS
> > (ACPI)?
> > 
> HAL scripts (in $prefix/libexec/hal iirc) use the underlying daemon's
> commands to suspend/hibernate/set power save profile.

In other words, the powersave daemon isn't really that useful for GNOME
on SUSE. I guess that once we get g-p-m to run when no one is logged in,
you guys can remove it completely on GNOME installations 

     David

headers

Danny Kukawka | 11 Apr 2006 19:13

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tuesday 11 April 2006 18:05, David Zeuthen wrote:
> On Mon, 2006-04-10 at 14:33 +0200, Rodrigo Moya wrote:
> > On Mon, 2006-04-10 at 13:29 +0200, Jeroen Zwartepoorte wrote:
> > > I'm curious: what role does the powersave daemon fulfill? I'm running
> > > FC5 and afaik, HAL/g-p-m gets it information directly from the OS
> > > (ACPI)?
> >
> > HAL scripts (in $prefix/libexec/hal iirc) use the underlying daemon's
> > commands to suspend/hibernate/set power save profile.
>
> In other words, the powersave daemon isn't really that useful for GNOME
> on SUSE. I guess that once we get g-p-m to run when no one is logged in,
> you guys can remove it completely on GNOME installations 

Sorry, that must say this: This is bulls***. 

There is no powermanagement on SUSE without powersave daemon. The powersave 
daemon provide much more functionality than g-p-m make available for the user 
at the moment. 

Powersave will never be removed from any SUSE/Novell products, because you 
can't use all the features powersave provide atm, as e.g. trigger suspend or 
change cpufrq policy, in this case. If you uninstall powersave from your SUSE 
Gnome installation also g-p-m can't e.g. suspend, also not via HAL. 

Yes, powersave handle powermanagement if there is no user and also if there is 
no X running. And this work perfect for SUSE with KDE, GNOME and every other 
windowmanager. But this is not the only reason to run powersave. The primary 
intention to develop powersave was not to have powermanagement if no user 
logged in - it was to allow powermanagement.

(Continue reading)

headers

Richard Hughes | 11 Apr 2006 23:34

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tue, 2006-04-11 at 19:13 +0200, Danny Kukawka wrote:
<snip other stuff>

> IMO the idea to start g-p-m in gdm is not really a good solution if you use 
> more than one desktop environment -  a cross desktop solution would be much 
> more better. 

Generic freedesktop-type DBUS API calls for stuff like the inhibit
interface (like has been suggested recently on d-d-l) is a very good
idea and something I wish to work towards. Coding anything else other
than a simple API specification to be cross-desktop is not something
that I think is viable, but this is only my opinion.

> Btw. I don't understand: What's the problem with a system daemon (with a well 
> defined DBUS-Interface, with a well defined default policy, without need a 
> session daemon) specialized to powermanagement. You can use provided 
> functions, but you don't need to use all available if you don't need them in 
> g-p-m. The daemon is _complete_ desktop _independent_, I think this was the 
> whish of Davyd Madeley (* A daemon with no GTK+ dependance that would be 
> suitable for cross-desktop use).

What if you want to interact with the user in the session? Like getting
the idle time for that session, or interacting with that instance of the
screensaver? Or notifying the user or displaying status? Or interacting
with other applications so that a complete integrated system is
possible? Then you need a session daemon also.

> This kind of discussion is not really helpful for anyone. I would like to see 
> a discussion about a cross desktop solution (if not about a special daemon 
> about a well defined, fullfeatured, cross desktop dbus interface which could

(Continue reading)

headers

David Zeuthen | 13 Apr 2006 21:52

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager


(dropping desktop-devel-list from the Cc)

On Tue, 2006-04-11 at 19:13 +0200, Danny Kukawka wrote:
> Btw. I don't understand: What's the problem with a system daemon (with a well 
> defined DBUS-Interface, with a well defined default policy, without need a 
> session daemon) specialized to powermanagement. You can use provided 
> functions, but you don't need to use all available if you don't need them in 
> g-p-m. The daemon is _complete_ desktop _independent_, I think this was the 
> whish of Davyd Madeley (* A daemon with no GTK+ dependance that would be 
> suitable for cross-desktop use).

Desktop independency for the sake of.. well desktop independency... is
just pretty point-less. 

Here are *some* of the problems

 1. You need something running in the session anyway
    - to provide a D-BUS interface for inhibiting actions
    - to read the users settings

 2. It's completely stupid to read system wide settings from a different
    location than per-user settings. Code duplication is bad.

 3. Having a system-wide daemon encourages very silly things
    - read plain text system wide settings from /etc instead of the
      default area of your desktop configuration system - what happens
      when your desktop configuration grows an LDAP backend? 
    - do completely silly things like grepping the process list and
      refuse to suspend e.g. when a process named "mplayer", "totem"

(Continue reading)

headers

Danny Kukawka | 14 Apr 2006 01:17

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Thursday 13 April 2006 21:52, David Zeuthen wrote:
> (dropping desktop-devel-list from the Cc)

Why?

> On Tue, 2006-04-11 at 19:13 +0200, Danny Kukawka wrote:
> > Btw. I don't understand: What's the problem with a system daemon (with a
> > well defined DBUS-Interface, with a well defined default policy, without
> > need a session daemon) specialized to powermanagement. You can use
> > provided functions, but you don't need to use all available if you don't
> > need them in g-p-m. The daemon is _complete_ desktop _independent_, I
> > think this was the whish of Davyd Madeley (* A daemon with no GTK+
> > dependance that would be suitable for cross-desktop use).
>
> Desktop independency for the sake of.. well desktop independency... is
> just pretty point-less.

Sorry, I don't understand what you try to tell me.

> Here are *some* of the problems
>
>  2. It's completely stupid to read system wide settings from a different
>     location than per-user settings. Code duplication is bad.

I don't see this problem. The deamon can read his settings from one place and 
can get changed via the dbus-interface from any connected client. It's 
complete uninteresting for the client from where the daemon get his settings 
and also complete uninteresting for the daemon from where the client read his 
settings. There is nothing stupid.

(Continue reading)

headers

David Zeuthen | 14 Apr 2006 02:06

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Fri, 2006-04-14 at 01:17 +0200, Danny Kukawka wrote:
> On Thursday 13 April 2006 21:52, David Zeuthen wrote:
> > (dropping desktop-devel-list from the Cc)
> 
> Why?

Because a discussion on how g-p-m works is off-topic on d-d-l? The
thread started with whether GNOME would include g-p-m in the 2.16
desktop release and people focused on two things.

> > On Tue, 2006-04-11 at 19:13 +0200, Danny Kukawka wrote:
> > > Btw. I don't understand: What's the problem with a system daemon (with a
> > > well defined DBUS-Interface, with a well defined default policy, without
> > > need a session daemon) specialized to powermanagement. You can use
> > > provided functions, but you don't need to use all available if you don't
> > > need them in g-p-m. The daemon is _complete_ desktop _independent_, I
> > > think this was the whish of Davyd Madeley (* A daemon with no GTK+
> > > dependance that would be suitable for cross-desktop use).
> >
> > Desktop independency for the sake of.. well desktop independency... is
> > just pretty point-less.
> 
> Sorry, I don't understand what you try to tell me.

As I said I think it's *great* to see GNOME and KDE interoperate and I'm
a big supporter of that. But one trend is that people get so eager about
it that they forget what problem they're trying to solve. KDE, GNOME,
Enlightenment and other desktops are different desktops because their
creators have different goals.

(Continue reading)

headers

Stefan Seyfried | 14 Apr 2006 21:35

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager


Hi,

Just some short annotations...

On Thu, Apr 13, 2006 at 08:06:55PM -0400, David Zeuthen wrote:

> In other words, once g-p-m is able to run when no-one is logged in (e.g.
> system-wide), we simply add a button
> 
>  [Make settings system-wide]
> 
> and the user authenticates; we write the settings to the default area of
> e.g. gconf. We can play other tricks to do this automatically if there
> is only one user on the system, only show this button if the user has
> certain privileges (group membership) etc. etc. 
> 
> If g-p-m relies on e.g. the powersave daemon there is no such guarantee
> because it also reads settings from /etc/foobar. Sure, the Powersave
> daemon may be changed to not do this but what's the point of the daemon
> then?

Well, it would not be too hard to tell the powersave daemon "write your
current configuration as systm wide default". It is not there yet, but
it could be added.

> actually real networking and all the corner cases that implies. Have you
> ever tried debugging a program that relies on two processes and IPC?
> It's much harder than if it's a single process.

(Continue reading)

headers

Richard Hughes | 15 Apr 2006 14:58

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Fri, 2006-04-14 at 21:35 +0200, Stefan Seyfried wrote:
> But you need the same daemon twice. Once for GNOME, once for KDE. That's
> code duplication, i'd say.

This is the KDE and GNOME way of doing things.

There are 3 ways to write a desktop application.

* using KDE technologies (C++, KDevelop, QT) so that it integrates well,
(but looks and feels wrong in GNOME)
* using GNOME technologies (C, Glade, GTK+) so that it integrates well,
(but looks and feels wrong in KDE)
* using custom technologies so that program does not integrate or feel
right with either KDE or GNOME.

Sharing anything other than a API spec is a kiss of death in my opinion.

> The problem is simply that you will always have some legacy applications
> that do not know about DBUS, inhibitors etc. RealPlayer anyone?

There's no reason RealPlayer wouldn't use a new inhibit interface if we
said "this will make RealPlayer work with GNOME"

> But of course you can leave the blacklist part out of gnome-powersave if
> this is the only problem 

I don't think a blacklist is the way to go in my opinion. And it's
certainly not the only problem.

> > The way this works with gnome-screensaver and gnome-power-manager is

(Continue reading)

headers

Richard Hughes | 11 Apr 2006 23:34

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tue, 2006-04-11 at 19:13 +0200, Danny Kukawka wrote:
<snip other stuff>

> IMO the idea to start g-p-m in gdm is not really a good solution if you use 
> more than one desktop environment -  a cross desktop solution would be much 
> more better. 

Generic freedesktop-type DBUS API calls for stuff like the inhibit
interface (like has been suggested recently on d-d-l) is a very good
idea and something I wish to work towards. Coding anything else other
than a simple API specification to be cross-desktop is not something
that I think is viable, but this is only my opinion.

> Btw. I don't understand: What's the problem with a system daemon (with a well 
> defined DBUS-Interface, with a well defined default policy, without need a 
> session daemon) specialized to powermanagement. You can use provided 
> functions, but you don't need to use all available if you don't need them in 
> g-p-m. The daemon is _complete_ desktop _independent_, I think this was the 
> whish of Davyd Madeley (* A daemon with no GTK+ dependance that would be 
> suitable for cross-desktop use).

What if you want to interact with the user in the session? Like getting
the idle time for that session, or interacting with that instance of the
screensaver? Or notifying the user or displaying status? Or interacting
with other applications so that a complete integrated system is
possible? Then you need a session daemon also.

> This kind of discussion is not really helpful for anyone. I would like to see 
> a discussion about a cross desktop solution (if not about a special daemon 
> about a well defined, fullfeatured, cross desktop dbus interface which could

(Continue reading)

headers

Danny Kukawka | 11 Apr 2006 19:13

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Tuesday 11 April 2006 18:05, David Zeuthen wrote:
> On Mon, 2006-04-10 at 14:33 +0200, Rodrigo Moya wrote:
> > On Mon, 2006-04-10 at 13:29 +0200, Jeroen Zwartepoorte wrote:
> > > I'm curious: what role does the powersave daemon fulfill? I'm running
> > > FC5 and afaik, HAL/g-p-m gets it information directly from the OS
> > > (ACPI)?
> >
> > HAL scripts (in $prefix/libexec/hal iirc) use the underlying daemon's
> > commands to suspend/hibernate/set power save profile.
>
> In other words, the powersave daemon isn't really that useful for GNOME
> on SUSE. I guess that once we get g-p-m to run when no one is logged in,
> you guys can remove it completely on GNOME installations 

Sorry, that must say this: This is bulls***. 

There is no powermanagement on SUSE without powersave daemon. The powersave 
daemon provide much more functionality than g-p-m make available for the user 
at the moment. 

Powersave will never be removed from any SUSE/Novell products, because you 
can't use all the features powersave provide atm, as e.g. trigger suspend or 
change cpufrq policy, in this case. If you uninstall powersave from your SUSE 
Gnome installation also g-p-m can't e.g. suspend, also not via HAL. 

Yes, powersave handle powermanagement if there is no user and also if there is 
no X running. And this work perfect for SUSE with KDE, GNOME and every other 
windowmanager. But this is not the only reason to run powersave. The primary 
intention to develop powersave was not to have powermanagement if no user 
logged in - it was to allow powermanagement.

(Continue reading)

headers

Richard Hughes | 10 Apr 2006 12:24

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Mon, 2006-04-10 at 12:25 +0200, Christian Fredrik Kalager Schaller
wrote:
> On Mon, 2006-04-10 at 11:59 +0200, Xavier Bestel wrote:
> > On Mon, 2006-04-10 at 01:28, Corey Burger wrote:
> > > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > > Elijah Newren wrote:
> > > > > On 4/9/06, Scott J. Harmon <harmon <at> ksu.edu> wrote:
> > > > >
> > > > >> Am I the only one who mouses over the applet to see how much more time
> > > > >> until the battery is fully charged?
> > > > >>
> > > > >
> > > > > Definitely not; I rely on this frequently.  I'd be heavily annoyed if
> > > > > the applet wasn't showing (and no other equally easy way of obtaining
> > > > > this information was available) when my laptop is plugged in and not
> > > > > fully charged.  If it's both plugged in and fully charged then I'd be
> > > > > fine with it not being there, as long as that was the only case.
> > > > >
> > > > Hmm. In this configuration, it is.
> > > 
> > > The problem with hiding in this case is that the user must know that
> > > when they are plugged in and fully charged, the icon will vanish,
> > > rather than just looking and seeing that they are fully charged and
> > > plugged in. Ouch.
> > 
> > Well, if once the battery reaches 100%, a short-lived notification
> > bubble says so then the icon can disappear without harm. No need to
> > pollute the notification area with a "battery is full" icon, OTOH on the
> > road the fuel gauge is important.
> >

(Continue reading)

headers

Paul Drain | 10 Apr 2006 03:38

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager


> > Nope. I'm working on a laptop at the moment, and I don't care that my
> > battery is fully charged. This is because it's plugged into the wall. If
> > I wasn't plugged into the wall, I'd start caring - but I'd also get a
> > battery symbol.
> 
> Am I the only one who mouses over the applet to see how much more time
> until the battery is fully charged?

Certainly not.

I'd be right miffed if this particular functionality wasn't available by
default on a laptop anymore -- especially when you're at a conference,
or some other 'on-AC-but-time-limited-by-other-commitments' event and
you *need* to make sure you have enough battery life available to
actually perform a demo.

I also might care about the icon, if say -- i've got a problem with ACPI
not detecting my battery -- yeah, the laptop's OK when you're plugged
into the wall (which I am now), but *all your work goes boom!* when you
disconnect the power.

Using Luis' 'chauffeur scenario' -- the latter case is kind of like your
driver turning around to you and saying, 'sir, we've left the gas tank
on the road, about ten miles back.'

Most Annoying.

Paul

(Continue reading)

headers

Elijah Newren | 10 Apr 2006 01:14

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Scott J. Harmon <harmon@...> wrote:
> Andrew Sobala wrote:
> > Nope. I'm working on a laptop at the moment, and I don't care that my
> > battery is fully charged. This is because it's plugged into the wall. If
> > I wasn't plugged into the wall, I'd start caring - but I'd also get a
> > battery symbol.
>
> Am I the only one who mouses over the applet to see how much more time
> until the battery is fully charged?

Definitely not; I rely on this frequently.  I'd be heavily annoyed if
the applet wasn't showing (and no other equally easy way of obtaining
this information was available) when my laptop is plugged in and not
fully charged.  If it's both plugged in and fully charged then I'd be
fine with it not being there, as long as that was the only case.

headers

Scott J. Harmon | 9 Apr 2006 23:53

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Andrew Sobala wrote:
> Corey Burger wrote:
>> On 4/9/06, Luis Villa <luis.villa@...> wrote:
>>  
>>> On 4/9/06, Andrew Sobala <aes@...> wrote:
>>>    
>>>> It's worth pointing out that gnome-power-manager is very much a
>>>> notifier
>>>> rather than an interactive applet. If your power cable falls out, it
>>>> pops up a message saying you've lost power. If you're working away from
>>>> a power source, there's a battery indicator with how much power you've
>>>> got left... that disappears when you're fully charged.
>>>>
>>>> (At least, that's how it's configured on my system.)
>>>>       
>>> This isn't the default, FWIW. I do agree that making this the default
>>> behavior  would be the best approach- better, IMHO, than a regular
>>> panel applet. I only want to know about power when something bad is
>>> going wrong, which is exactly what the notification area is for. An
>>> applet is all the time, and so is the current default behavior in the
>>> notification area- both of which are broken.
>>>
>>> Luis
>>>     
>>
>> I completely disagree. There are a few good reasons why an icon should
>> be displayed all the time
>>
>> 1. What state the battery is in is always relevant. Power is the
>> single most important thing on a laptop. Without it, you are going

(Continue reading)

headers

Richard Hughes | 9 Apr 2006 23:09

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 13:57 -0700, Corey Burger wrote:
> On 4/9/06, Luis Villa <luis.villa <at> gmail.com> wrote:
> > On 4/9/06, Andrew Sobala <aes <at> gnome.org> wrote:
> > > Jaap Haitsma wrote:
> > > >>> Richard,
> > > >>>
> > > >>>
> > > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > > >>> and a "notication area icon"/applet would not be so hard.
> > > >>>
> > > >>> They are pretty independent from each other.
> > > >>>
> > > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > > >>> take appropriate actions etc. If people run the daemon then they get all
> > > >>> the power management features.
> > > >>>
> > > >>> The applet/"notification area icon" just needs to watch the batteries
> > > >>> (code of the daemon can be reused  )and show the status by changing
> > > >>> it's icon and displaying notifications.
> > > >>>
> > > >>> The only message I see that the "daemon" might want to send to the
> > > >>> applet is a message that the system is going to suspend/hibernate and
> > > >>> that is already something we want to do to notify other apps that the
> > > >>> system is going to suspend/sleep and that they need to take appropriate
> > > >>> actions if necessary.
> > > >>>
> > > >>> So in my opinion it's not that difficult, or am I missing something?
> > > >>>
> > > >> But what's the point?
> > > >>

(Continue reading)

headers

Luis Villa | 9 Apr 2006 23:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Corey Burger <corey.burger@...> wrote:
> On 4/9/06, Luis Villa <luis.villa@...> wrote:
> > On 4/9/06, Andrew Sobala <aes@...> wrote:
> > > Jaap Haitsma wrote:
> > > >>> Richard,
> > > >>>
> > > >>>
> > > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > > >>> and a "notication area icon"/applet would not be so hard.
> > > >>>
> > > >>> They are pretty independent from each other.
> > > >>>
> > > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > > >>> take appropriate actions etc. If people run the daemon then they get all
> > > >>> the power management features.
> > > >>>
> > > >>> The applet/"notification area icon" just needs to watch the batteries
> > > >>> (code of the daemon can be reused  )and show the status by changing
> > > >>> it's icon and displaying notifications.
> > > >>>
> > > >>> The only message I see that the "daemon" might want to send to the
> > > >>> applet is a message that the system is going to suspend/hibernate and
> > > >>> that is already something we want to do to notify other apps that the
> > > >>> system is going to suspend/sleep and that they need to take appropriate
> > > >>> actions if necessary.
> > > >>>
> > > >>> So in my opinion it's not that difficult, or am I missing something?
> > > >>>
> > > >> But what's the point?
> > > >>

(Continue reading)

headers

Richard Hughes | 9 Apr 2006 23:09

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 13:57 -0700, Corey Burger wrote:
> On 4/9/06, Luis Villa <luis.villa@...> wrote:
> > On 4/9/06, Andrew Sobala <aes@...> wrote:
> > > Jaap Haitsma wrote:
> > > >>> Richard,
> > > >>>
> > > >>>
> > > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > > >>> and a "notication area icon"/applet would not be so hard.
> > > >>>
> > > >>> They are pretty independent from each other.
> > > >>>
> > > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > > >>> take appropriate actions etc. If people run the daemon then they get all
> > > >>> the power management features.
> > > >>>
> > > >>> The applet/"notification area icon" just needs to watch the batteries
> > > >>> (code of the daemon can be reused  )and show the status by changing
> > > >>> it's icon and displaying notifications.
> > > >>>
> > > >>> The only message I see that the "daemon" might want to send to the
> > > >>> applet is a message that the system is going to suspend/hibernate and
> > > >>> that is already something we want to do to notify other apps that the
> > > >>> system is going to suspend/sleep and that they need to take appropriate
> > > >>> actions if necessary.
> > > >>>
> > > >>> So in my opinion it's not that difficult, or am I missing something?
> > > >>>
> > > >> But what's the point?
> > > >>

(Continue reading)

headers

Corey Burger | 9 Apr 2006 22:57

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Luis Villa <luis.villa@...> wrote:
> On 4/9/06, Andrew Sobala <aes@...> wrote:
> > Jaap Haitsma wrote:
> > >>> Richard,
> > >>>
> > >>>
> > >>> As far as I understand the code of GPM splitting up GPM in a "daemon"
> > >>> and a "notication area icon"/applet would not be so hard.
> > >>>
> > >>> They are pretty independent from each other.
> > >>>
> > >>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
> > >>> take appropriate actions etc. If people run the daemon then they get all
> > >>> the power management features.
> > >>>
> > >>> The applet/"notification area icon" just needs to watch the batteries
> > >>> (code of the daemon can be reused  )and show the status by changing
> > >>> it's icon and displaying notifications.
> > >>>
> > >>> The only message I see that the "daemon" might want to send to the
> > >>> applet is a message that the system is going to suspend/hibernate and
> > >>> that is already something we want to do to notify other apps that the
> > >>> system is going to suspend/sleep and that they need to take appropriate
> > >>> actions if necessary.
> > >>>
> > >>> So in my opinion it's not that difficult, or am I missing something?
> > >>>
> > >> But what's the point?
> > >>
> > >>

(Continue reading)

headers

David Zeuthen | 10 Apr 2006 00:31

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 21:12 +0200, Jaap Haitsma wrote:
> 1. It's good design to split up parts which are doing different things
> ( You can also put all your code in one source file, but that's not good
> design )

Even though D-BUS indeed makes it easy to do IPC I fail to see why you'd
want to split g-p-m in two processes. Remeber that g-p-m is a lot more
than an icon; it also pops up dialogs from time to time...

> 2. An applet would be much more consistent with how GNOME works at the
> moment. If I want to add something to the panel I just add there by
> doing "Add to panel" and if I want to remove it I choose "Remove from
> panel". GNOME unlike windows luckily doesn't put many stuff
> automagically in the panel 

Suppose g-p-m was an applet instead of using the notification area. Do
you think it's a good idea to let users remove this applet? 

Basically I think the notification area is the right answer for UI
gizmos that directly relate to the state of the computer while applet is
the answer for UI gizmos that doesn't.

    David

headers

Andrew Sobala | 9 Apr 2006 21:25

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Jaap Haitsma wrote:
>>> Richard,
>>>
>>>
>>> As far as I understand the code of GPM splitting up GPM in a "daemon"
>>> and a "notication area icon"/applet would not be so hard.
>>>
>>> They are pretty independent from each other. 
>>>
>>> The "daemon" just has to watch batteries, laptop lid, hardware keys and
>>> take appropriate actions etc. If people run the daemon then they get all
>>> the power management features. 
>>>
>>> The applet/"notification area icon" just needs to watch the batteries
>>> (code of the daemon can be reused  )and show the status by changing
>>> it's icon and displaying notifications. 
>>>
>>> The only message I see that the "daemon" might want to send to the
>>> applet is a message that the system is going to suspend/hibernate and
>>> that is already something we want to do to notify other apps that the
>>> system is going to suspend/sleep and that they need to take appropriate
>>> actions if necessary.
>>>
>>> So in my opinion it's not that difficult, or am I missing something?
>>>       
>> But what's the point?
>>
>>     
>
> 1. It's good design to split up parts which are doing different things

(Continue reading)

headers

Jeff Spaleta | 9 Apr 2006 20:22

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On 4/9/06, Jaap Haitsma <jaap@...> wrote:
> The only message I see that the "daemon" might want to send to the
> applet is a message that the system is going to suspend/hibernate and
> that is already something we want to do to notify other apps that the
> system is going to suspend/sleep and that they need to take appropriate
> actions if necessary.
>
> So in my opinion it's not that difficult, or am I missing something?

I thought this was part of the power of dbus enabled technologies...
the ability to write
small daemon processes which use the bus to communicate to/from
whatever clients want to be listening.  Right now it might seem that
the desktop applet is the primary and only mechanism that informs
desktop users. But as a sysadmin I want the Network Manager stuff to
be as flexible as possible so that I can script my own dbus clients
which trigger actions like email or logging events by the same
communication mechanism that triggers UI changes in the desktop applet
used to inform the active desktop user of power related events.

I'm not going to fight to have the standard g-p-m ui and configuration
mechanisms incorporating the functionality that I need locally as a
sysadmin.  My needs are absolutely based on local policy crackrock and
I respect the bounds of the usage-case that is being designed for with
the applet and the exposed configuration UI.

To do what I need to do to write specialized local dbus clients for my
local needs as a sysadmin, I don't need anything more than a level
playing field with regard to available communication from the "daemon"
to the applet "client".   While I don't need the applet to be stripped

(Continue reading)

headers

Richard Hughes | 9 Apr 2006 20:05

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 19:56 +0200, Jaap Haitsma wrote:
> On Sun, 2006-04-09 at 14:47 +0100, Richard Hughes wrote:
> > On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> > > On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> > > 
> > > > Since the original announcement mail about gnome-power-manager, we have
> > > > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > > > starting to integrate with other parts of the GNOME application stack.
> > > > Lots of new functionality has been added, and lots of polish has been
> > > > applied. See the screenshots area of my website[5] for some cool
> > > > screenshots of the latest stuff in the 2-15 branch.
> > > 
> > > I would like to see g-p-m fragmented into three parts.
> > > 
> > >  * A daemon with no GTK+ dependance that would be suitable for
> > >    cross-desktop use
> > >  * A capplet (this exists today)
> > >  * A notification area icon (libnotify dependance goes here)
> > 
> > Umm, no.
> > 
> > The IPC between these components would be horrific and over-complicated
> > for no actual gain. KDE are quite happy with their own power management
> > applications, and no KDE developer has ever mentioned to me that they
> > would want such a cross-desktop daemon.
> > 
> Richard,
> 
> 
> As far as I understand the code of GPM splitting up GPM in a "daemon"

(Continue reading)

headers

Steve Frécinaux | 9 Apr 2006 17:13

[gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

Davyd Madeley wrote:
> I would like to see g-p-m fragmented into three parts.
> 
>  * A daemon with no GTK+ dependance that would be suitable for
>    cross-desktop use
>  * A capplet (this exists today)
>  * A notification area icon (libnotify dependance goes here)
> 
> This would allow us to more easily address integration issues with
> GNOME and other desktops and it means that we can aim at avoiding
> notification area pollution (because session initialised notification
> icons are a violation of all that is good and right).

On my own I'd like to see a real gnome-terminal applet instead of that 
notification icon, eventually replacing the current batstat applet...

headers

Richard Hughes | 9 Apr 2006 15:47

Re: [gpm] Re: Gnome 2.16 Module Proposal: GNOME Power Manager

On Sun, 2006-04-09 at 21:29 +0800, Davyd Madeley wrote:
> On Sun, Apr 09, 2006 at 01:47:12PM +0100, Richard Hughes wrote:
> 
> > Since the original announcement mail about gnome-power-manager, we have
> > moved the mailing list to gnome.org, are now hosted on gnome.org, and am
> > starting to integrate with other parts of the GNOME application stack.
> > Lots of new functionality has been added, and lots of polish has been
> > applied. See the screenshots area of my website[5] for some cool
> > screenshots of the latest stuff in the 2-15 branch.
> 
> I would like to see g-p-m fragmented into three parts.
> 
>  * A daemon with no GTK+ dependance that would be suitable for
>    cross-desktop use
>  * A capplet (this exists today)
>  * A notification area icon (libnotify dependance goes here)

Umm, no.

The IPC between these components would be horrific and over-complicated
for no actual gain. KDE are quite happy with their own power management
applications, and no KDE developer has ever mentioned to me that they
would want such a cross-desktop daemon.

> This would allow us to more easily address integration issues with
> GNOME and other desktops and it means that we can aim at avoiding
> notification area pollution (because session initialised notification
> icons are a violation of all that is good and right).

g-p-m can default to only displaying when the battery power is critical,

(Continue reading)