[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Ui enhancements
From:       "Aaron J. Seigo" <aseigo () olympusproject ! org>
Date:       2003-01-26 6:44:17
[Download RAW message or body]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Saturday 25 January 2003 03:06, Michael Kreitzer wrote:
> > We just have had a very heated discussion on the kde-usability list on
> > adding more configurability to KDE and most people there would say no to
> > add any single more thing to configure almost regardless how useful or
> > slick  it would be, as it would add complexity to an already  complex
> > system. In this case the readability problem can be solved just by using
> > another background.

i don't think this is an accurate portrayal of the general sentiment on 
kde-usability. there is a deffinite sensitivity to not making aboslutely 
every single thing possible configurable as that can have serious collateral 
affects (such as learnability, effectivity, being able to offer quality tech 
support, efficiency, etc). on the other hand, there most deffinitely ARE 
things that can and should be made configurable. and some things may be 
configurable but do not belong in a KDE control panel, since that leads to 
configuration panels that have so many microconfigurations that they become 
unusable.

the idea of a "power tweaker's" config tool has been discussed and i think 
gained serious traction as a good middle ground for those items that would be 
really nice to have the option of configuring but are too detailed or too 
detrimental in the common case to offer in a prominent fashion.

if one looks at the track record of actual code produced by the kde-usability 
team you may notice that a number of new options and features have been 
added. but it is often done in such a way that it appears simpler. on the 
other hand, we've forgone features or made the really difficult "policy 
decision" when it is best to do so. 

it's all about ballance, as others in this thread have noted.

> This has got to be the most rediculous thing I have ever heard.  To think
> someone would even think that not adding configurability is a "Good Thing"
> 'cause it might step on some poor newbies toes is absurd.

it isn't about stepping on some poor newbies toes, but about creating a system 
that is predictable, learnable and useful. most people use computers to get 
things done, not simply for the joy of using them (even though probably you 
and i both do =). when decisions to make something configurable are made by 
developers that hurt the ability for people (of any skill level) to use and 
learn the system, we are not improving KDE but rather we are degrading it.

we don't win based on feature count, we win based on the quality of a user's 
experience. configurability is deffinitely a part of that; but it isn't 
anywhere near being even a large part of it.

>  Create a button and label it "Advanced."  Problem solved.

there are often other approaches that can be used first, such as rethinking 
wording, lay out, choice of widgets, grouping and, yes, removal / hiding of 
options. the advanced button/tab should really be a last resort, it isn't a 
magic hammer.

> If the newbie gets his poor
> underused mind in a twist 'cause he clicked the "Advanced" he has no one to
> blame but himself. 

please don't blame the users. for anyone who hasn't done this and designs UIs: 
watch some really bright people use computers for an extended peroid of time. 
go to your University and watch some of the bright professors use their 
computers; it becomes rapidly obvious that this isn't a "stupid users" 
problem. it's a "stupid programmers" problem when we don't make tools that 
work properly and elegantly for the intended purpose.

> Forcing users who want to and do know what they are
> doing to resort to code modification to get desired changes is NOT a "Good
> Thing."

some ideas for configuration are patently absurd and should be avoided. just 
because we can code something doesn't necessarily mean we should.

on the other hand, there's no point to lock things down too much. a power 
tweaker's tool is very possible and will, i'm guessing, make those 
tweak-hungry people (the vast minority, btw) happier. not every option needs 
to be thrown in every user's face by placing it in KControl. what would you 
think about such a tool? (you can search the kde-usability archives for 
kconfedit if you want to see the dicussions about it)

> Configurability and ease of use do not have to be mutually
> exclusive. 

agreed! they don't have to be, but they can be. and that's the entire trick =)

> Give the users multiple levels of abstraction and allow them to
> select the one they are comfortable with. 

this gets suggested often and is often one of the first suggestions offered 
for this problem. it's even been tried in the real world. the reason why you 
don't see user levels used everywhere is: 

                                                           it doesn't work.

- -- 
Aaron J. Seigo
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

"Everything should be made as simple as possible, but not simpler"
    - Albert Einstein
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE+M4PB1rcusafx20MRAtAvAKCB4XjLQLbWyl/tytmj2nZ22xSpOACfV7Xh
tpHcAMSeTC2YtMBunEH2LcU=
=WKCz
-----END PGP SIGNATURE-----

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]