After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 720656 - Some confusion with the lock screen/shield.
Some confusion with the lock screen/shield.
Status: RESOLVED OBSOLETE
Product: gnome-shell
Classification: Core
Component: lock-screen
3.26.x
Other All
: Normal normal
: ---
Assigned To: gnome-shell-maint
gnome-shell-maint
Depends on:
Blocks:
 
 
Reported: 2013-12-18 05:36 UTC by Sri Ramkrishna
Modified: 2021-07-05 14:34 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Sri Ramkrishna 2013-12-18 05:36:47 UTC
I'm putting this on behalf of Jonathan Corbet, based on a conversation on google+ that is archived here:

https://plus.google.com/104175436979387006170/posts/HKGb9jTnZ8e

One problem with the design and I've had the same feeling as well as that it takes extra steps to get to the point of logging in.  While you are able to just type your password it isn't exactly intuitive.  The reason is the following:

Every cue indicates a physical intervention to remove the shield.  You have animated arrows that ask you to remove the shield using a visual sweep of the arrow, or a keyboard shortcut like 'esc' to remove it.  Even though you know that you could just type your password, there is just too many hints that causes you to want to use the esc key or a mouse action to reomve the barrier.

It's the shield that can be problematic.  While we use it for notifications, it might be worth investigating whether people actually notice or even care that there are notifications there.

It might be worth re-vistiting this and doing some user tests to see if people truly find the notifications and the shield useful.  Indications from the community so far has been non-commital.  My personal experience is that while it is useful to know that you have mail and how much, I only spend about less than a second before I am already hitting 'esc', or using the mouse to remove the shield and logging in.

In the meanwhile, I'm hoping that we can fix the extension to remove the shield so it is useful to remove it.  It would be awesome to look at this and see if we can utilize existing screen space and still allow a user/password prompt for those who dont really want to know any information.  After all, a lot of us lead busy lives and we don't actually care about extraneous information on the screen a lot of the times because during the times of the break we are probably thinking of something else and when we are finally in front of the screen we are focused on doing some work especially from a corproate setting.
Comment 1 Giovanni Campagna 2013-12-18 09:58:08 UTC
I'd rather have this option in core (as an hidden/tweak-tool setting), rather than maintaining an extension that should not work at all.
Comment 2 Milan Bouchet-Valat 2013-12-18 13:34:15 UTC
I think one of Jonathan's points you summary is missing is that even when the session is not locked (i.e. no need to type your password), the shield activates. This is terribly frustrating when you just stopped typing or moving the pointer for a few moments (e.g. listening for a podcast a reading news while eating). Before 3.6, moving the mouse pointer was enough to wake up the screen and access your session; now you need to roll the curtain up, which requires keeping the mouse button pressed and performing a relatively complex movement.

I think it would make sense to only activate the shield when the session is locked, which usually happens after a longer delay than merely turning off the screen. If it's not locked, just turn on the screen, show the normal session applications, and pop up new notifications (if any).
Comment 3 Allan Day 2013-12-18 15:05:45 UTC
(In reply to comment #0)
> I'm putting this on behalf of Jonathan Corbet ...

One thing that isn't entirely clear to me here - if someone learns that typing their password skips the lock screen, why would it remain an issue for them? It's pretty easy to avoid the thing completely, if you want to.


(In reply to comment #2)
> I think one of Jonathan's points you summary is missing is that even when the
> session is not locked (i.e. no need to type your password), the shield
> activates. ...

That seems like a separate issue.
Comment 4 Giovanni Campagna 2013-12-18 18:14:28 UTC
(In reply to comment #3)
> (In reply to comment #0)
> > I'm putting this on behalf of Jonathan Corbet ...
> 
> One thing that isn't entirely clear to me here - if someone learns that typing
> their password skips the lock screen, why would it remain an issue for them?
> It's pretty easy to avoid the thing completely, if you want to.

I believe part of the problem comes from people who don't have a password, or don't want to type it (ie, after screen blanking, before screen locking). We have the shield at that time, and you can lift it with space or esc, but not say ctrl, alt or shift.
Btw, at that time the shield is also invisible (monitor is turned off to save power), so becomes naturally less useful.
Comment 5 Sri Ramkrishna 2013-12-18 18:28:47 UTC
It actually depends.  For instance, conceptually you're presented with a wall. 
Everything in my mind tells me it is a wal, e.g. it's called a 'shield', you
have a visual queue with arrows animating upwards.  Sometimes it's really hard
not keep responding to those cues, and hit the 'esc' key or use the mouse to
move the shield up.  Even though I know I can start hitting the keys all visual
cues/hints are telling me otherwise.  It's actually hard because you're trying
to ignore what the screen is telling me.

The other issue, is that most of the time when I'm at my screen the computer
screen is dark as the machine has gone to sleep since for instnace.  The amount
of time I spend caring in that window after the machine wakes up is pretty
small and I just want to type my password in and login and start doing whatever
tasks that needs to be done.  My situation might be unique in this instance, I
don't knokw.
Comment 6 Allan Day 2014-01-14 18:20:28 UTC
(In reply to comment #5)
> It actually depends.  For instance, conceptually you're presented with a wall. 
> Everything in my mind tells me it is a wal, e.g. it's called a 'shield', you
> have a visual queue with arrows animating upwards.  Sometimes it's really hard
> not keep responding to those cues, and hit the 'esc' key or use the mouse to
> move the shield up. Even though I know I can start hitting the keys all visual
> cues/hints are telling me otherwise.  It's actually hard because you're trying
> to ignore what the screen is telling me.

So you are waking the screen up and then typing your password? How do you tend to wake the screen up? What is your usual routine for unlocking?

...
> The amount
> of time I spend caring in that window after the machine wakes up is pretty
> small ...

Sorry, I can't parse that sentence!
Comment 7 Bastien Nocera 2014-11-09 20:00:11 UTC
Questions were asked in comment 6.
Comment 8 Nicholas Miell 2015-06-07 22:42:40 UTC
My usual unblanking routine for the 25 years prior to the introduction of the GNOME screen shield was hitting the spacebar or moving the mouse (no password to unblank in my case).

I had to retrain myself to exclusively press the Esc key to unblank the screen because nothing else worked, and even figuring that out took some Googling because the screen shield doesn't have a user interface, it just sits their inert and occasionally blinks a non-functional arrow (clicking it does nothing, not that it looks like a clickable UI element) at you after a long delay. (Clicking and dragging the screen shield as that animated arrow is meant to imply is absurdly slow and why anybody thought this was a good design is mystifying.)
Comment 9 Matthias Clasen 2015-06-08 19:53:23 UTC
(In reply to Nicholas Miell from comment #8)
> My usual unblanking routine for the 25 years prior to the introduction of
> the GNOME screen shield was hitting the spacebar or moving the mouse (no
> password to unblank in my case).

Both of these work fine here - wiggling the mouse wakes up the screen, and hitting space wakes up the screen and lifts the shield.
Comment 10 Nicholas Miell 2015-06-09 02:12:11 UTC
You're confusing "displaying the screen shield" with "unblanking", as far as I'm concerned, the screen shield is a very boring screensaver.

I'd expect mouse movement to dismiss the screen shield entirely.

Its nice that keyboard input works now, though. That's definitely an improvement over the original release.
Comment 11 ville.ranki 2017-10-11 09:55:25 UTC
Previously to dismiss screen saver and get the password prompt it was enough to move the mouse. Now a mouse swipe from bottom to top is required. This would make sense on a touch screen tablet, but on pc it's useless and confusing.

My first reaction to screen shield was "something is broken, the password prompt doesn't work when i move the mouse". According to related comments on the Internet, I'm not alone. So this is a usability bug.

Suggestion: get rid of it.
Comment 12 James Roper 2017-10-30 03:57:14 UTC
I've just switched to Gnome 3 from Unity (as part of upgrading to Ubuntu 17.10), and I found this very unintuitive too. The screen has an animation saying to swipe up, my first reaction to this was "wtf? I'm not on a phone, my monitor isn't touch sensitive, why is it asking me to swipe up?" So I spent the first few hours clicking and dragging to get to the password prompt, until I started researching and found you could just start typing your password. There are no visual cues on the screen to indicate that you can do this, there is only one visual cue for how to get rid of that screen, that's the swipe up cue.

If you can just start typing your password to clear it, then why not put a password input field on the screen to indicate that? Why can't the notification screen have a password input field on it if it accepts your password? That's the way that 99.99999% of applications tell you that you can enter your password, they put a password input field there, why would Gnome break from this UI indicator? Typing my password when there's no highlighted input field for it to go in feels so wrong and dangerous - over the years I have trained myself to be very careful that I'm always typing my password into an active password field, because getting that wrong is what leads to things like tweeting your password, or typing your password in the clear on a screen in the middle of a presentation to a conference room full of people. Now it feels like I have to unlearn this mental training to use Gnome.

As for notifications - I don't know what Gnome would ever have to notify me that I would be interested in. I have a phone in my pocket, notifications for email and calendar events go to that, if I just want to check my notifications, I'll pull my phone out of my pocket and check it, takes 2 seconds. I would never go to all the effort of opening my laptop just to check notifications, I only ever open my laptop if I have a specific task that I want to achieve, in that case, showing me notifications before login is just another thing preventing me from getting to doing the thing that I came to do.
Comment 13 Bill Hayden 2018-05-16 19:00:30 UTC
If I could vote this bug up 1000 times, I would.  I don't agree with the need for a shield, and I don't personally know anyone that does. I don't agree that it should be the default, and I don't personally know anyone that does.  But whoever thought there should be a shield, turned on by default, *with no built-in way to turn it off* should simply not be writing designs and coding for GNOME, especially since this atrocious behavior has now been standard for 7 (!) GNOME releases as I write this.

Please put this option in core.
Comment 14 Florian Müllner 2018-05-16 19:25:40 UTC
The new unlock/login designs[0] put away the separate screen shield altogether, so this bug is likely obsolete - the one thing that's possibly missing there is the question of whether the password-less single-user case should be special-cased (by simply "unlocking" via any user activity). Should we move that discussion to the new design issue?

[0] https://gitlab.gnome.org/GNOME/gnome-shell/issues/276
Comment 15 GNOME Infrastructure Team 2021-07-05 14:34:12 UTC
GNOME is going to shut down bugzilla.gnome.org in favor of  gitlab.gnome.org.
As part of that, we are mass-closing older open tickets in bugzilla.gnome.org
which have not seen updates for a longer time (resources are unfortunately
quite limited so not every ticket can get handled).

If you can still reproduce the situation described in this ticket in a recent
and supported software version, then please follow
  https://wiki.gnome.org/GettingInTouch/BugReportingGuidelines
and create a new ticket at
  https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/

Thank you for your understanding and your help.