https://bugs.kde.org/show_bug.cgi?id=390634
Pedro V <voidpointertonull+bugskde...@gmail.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |voidpointertonull+bugskdeor
| |g...@gmail.com
--- Comment #12 from Pedro V <voidpointertonull+bugskde...@gmail.com> ---
Isn't this mostly a duplicate of the significantly older Bug 336369, or is this
supposed to be Bug 336369 + Bug 392798 ?
Generally it seems like fragmentation of discussion isn't ideal as in the end
whether the operation is initiated with a HID device or with an ACPI event, the
problem is the security consideration.
Regarding the security aspect, for those with not much imagination I'd
recommend to envision a setup where only a keyboard and a mouse is available to
the user, the host itself isn't in reach. There are other possible setups, and
even if they aren't the most common, one of the charms of KDE is usually
striking a quite okay balance between security and convenience with default
settings, and covering a lot of use cases with settings.
It's definitely interesting though that in this case there seems to be really
just a regression in convenience instead of improvement in security:
- Not sure if it's a result of me enabling hibernation with overly permissive
configuration a while ago, but hibernate and sleep options are present for me
by default which are not as bad as shutdown and restart, but can be still
disruptive
- Initiating user switching really still present restart and shutdown options
as of Plasma 5.27.8, and they do work, so there was no security improvement,
just some obfuscation
I generally do agree with the idea that most setups are likely to be single
user, so this is a low priority issue, but I can envision a single checkbox
option controlling whether these controls should be available without
authentication covering common needs:
- Single user setups could have convenient controls
- Multi-user / more secure setups would still need to login first, but would
have the desired security
It's likely a good idea though to revert the change that ended up being just
security through obscurity, both confusing a lot of users thinking a feature is
missing, and also giving the false sense of security.
--
You are receiving this mail because:
You are watching all bug changes.
