RFC: System Settings categorisation overhaul

Darío Andrés andresbajotierra at gmail.com
Sun Oct 4 01:32:46 BST 2009

On Sat, Oct 3, 2009 at 9:09 PM, Aaron J. Seigo <aseigo at kde.org> wrote:
> On October 3, 2009, Darío Andrés wrote:
>> > i'm not saying it couldn't be better, but let's be sure it gets done
>> > better instead of trying and running the risk of probably ending up with
>> > another not- great arrangement.
>> That's why we are discussing it here, to come up with something
>> better, which should be easy and discoverable.
> kde-core-devel is not a good place to find out how average computer users
> perceive hierarchies of data that reflect the technical details of the system.
> it's probably one of the worst, actually, because it is populated by deeply
> advanced users who know all the details from the inside.

Ok... we choosed a bad place to discuss this..
This whole thing started on forums.k.o , may be it should continue
there.. (lots of users, not so much devs)

>> If it is done properly, it should help us to get something better; and
> i don't actually share the same confidence :)

Bah, to be sincere ... I don't have confidence at all about this ... nevermind.

>> the "relearning" process should not hurt at all.
> it does hurt. every time we change this structure, we cause problems for
> users. we hear about it. and it's the kind of problems that are really hard to
> justify. it doesn't really provide any new advantage to these people, it just
> discards their hard earned knowledge and exchanges it for another arcane set
> of hierarchical modules.
> (and yes, search is a far better approach here :)

I like the search approach too

>> > "Plasma Containments" is two jargon words put together. we should not be
>> > using jargon _anywhere_ in our UIs that the average user is meant to go
>> > through. in this case the name should be "Desktop Activities". it's a
>> > moot point in thi case, though, as this panel doesn't belong there at all
>> > (unless it's something other than what the name suggests?)
>> Agreed. (I never liked "Plasma Containments", but it was the first
>> name that come to my mind to describe it, sorry about that)
> what settings does this panel contain?

That KCM would be: http://www.notmart.org/images/netbooksettings.png ?
( http://www.notmart.org/index.php/Software/Trying_the_netbook_project_made_ )

>> > what's the difference between "Connections Manager" and "Connection
>> > Settings"? what "Resources" are being "shared"?
>> "Connections Manager" is just a dummy group for NetworkManager/WiCD ,
>> as I don't have such KCM I don't know how to group them nor name them
>> properly.
>> "Connections Settings" is the former "Network Settings" , proxy and
>> other connection preferences, timeouts and so on. Ben pointed that it
>> would make sense to merge this into "Connections Manager"..
> probably; use testing would help us with that.
>> "Resource sharing" should be "File sharing"
> probably a bit better; user testing would help us with that.
>> > is "Multimedia" really hardware?
>> "Multimedia" includes the Phonon KCM which is about configuring audio
>> and video output and settings, that kind includes hardware
>> configurations ("to which speaker should X category sounds be
>> played..."). but it is not strictly about hardware.. Where would you
>> put it.. ? could it fit on "Workspace"?
> i don't think so as it isn't exclusive to the workspace. my point here was
> that it's ambiguous, and so we really need to test this on actual people.
>> > would it make more sense to have Workspace / Applications / System
>> > instead of Personal / W & A Behaviour / Hardware?
>> Yes, as I said before it would make sense. The problem could appear
>> when defining which settings is from Applications, which one is from
>> Workspace and which one could be related to both.
>> About the last category, are you suggesting to merge "Hardware" +
>> "System Administration" ?
> perhaps; it would require user testing. :)
>> > i really think this needs to be done with a greater amount of usability
>> > input and principles applied and with subsequent rounds of testing.
>> Agreed. Should we redirect this to kde-usability ?
> yes and no. yes in that kde-usability is the right area, and no in that we
> really shouldn't be discussing this on a KDE mailing list, we should be
> testing people.
> these kinds of tests are actually pretty easy to do compared to other kinds of
> usability research. we could each go out and test 3-5 people and have a great
> sampling. Celeste could help set up some of the parameters. i think card
> sorting would be perfect here.

OK, agreed on doing some user testing to resolve this.... (or should
we just implement the "improved searching" function and forget all the
mess about reordering SS ?)


> --
> Aaron J. Seigo
> humru othro a kohnu se
> GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43
> KDE core developer sponsored by Qt Development Frameworks

More information about the kde-core-devel mailing list