Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 4 Nov 2008 17:16:53 -0700
From:      Chad Perrin <perrin@apotheon.com>
To:        freebsd-questions@freebsd.org
Subject:   Re: Is KDE4 usable on FreeBSD?
Message-ID:  <20081105001653.GA2742@kokopelli.hydra>
In-Reply-To: <120974.86441.qm@web56801.mail.re3.yahoo.com>
References:  <490CADB3.4070902@lazlarlyricon.com> <120974.86441.qm@web56801.mail.re3.yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help

--7JfCtLOvnd9MIVvH
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Nov 01, 2008 at 01:09:18PM -0700, mdh wrote:
> --- On Sat, 11/1/08, Rolf G Nielsen <listreader@lazlarlyricon.com> wrote:
> >=20
> > If I need to (re)configure the behaviour of som app or part
> > of the system, I edit the appropriate config file, which
> > takes about a minute or two...
>=20
> Unless you've never modified the configs for that app before, in which ca=
se you have to learn the configuration format.  It also sometimes occurs th=
at these formats and locations and whatnot are changed between released by =
the developers.  Above and beyond that, some apps have good configuration d=
ocumentation and are a breeze.  Others, less so. =20
>=20
> I'm not advocating a user interface for configuring everything, but for c=
ertain things which are inherently extremely complex, such as window manage=
r layout and behavior, it's my opinion that it really is a time-saver. =20
>=20

For heavy-weight GUI environments like KDE and GNOME, and even
"feature-rich" but kinda medium-weight alternatives like WindowMaker
(possibly with GNUstep to make it a complete "desktop environment"), I
agree: a GUIfied configuration utility is a dire necessity.

For something at the lightweight end of the spectrum (assuming halfway
decent design), such as AHWM or wmii, such a tool would just get in the
way.


> >=20
> > If a user of some fancy desktop with lots of whistles and
> > bells wants to do the same, he/she has to browse through an
> > extensive hierarchy of categories and subcategories to get
> > to the setting he/she wants to change. That hierarchy is
> > more than often far from intuitive, so that very same task
> > may take ten minutes or more.
>=20
> I find KDE's configuration interface to be intuitive and generally quite =
sane.  GNOME's isn't lacking in that area either, imho, it's just lacking a=
 lot of options that I feel ought to be tunable parameters (most of which a=
re, but require extensive config file hacking...) =20
>=20

My very vague recollection of KDE 3 is that it was much easier to find
what I needed for configuration purposes than it is with KDE 4.  Version
4 seems to either lack a lot of configuration options or hide them really
well for some perverse reason.


> The simple fact is that I can configure my KDE desktop quicker than someo=
ne can, seeking the same granularity of modification, configure something w=
hich has no UI for configuration. =20
> This isn't too big a deal for me, or you, or likely many of the folks on =
this list, but for someone who is new to FreeBSD and has never hacked a win=
dow manager config file before, it likely is.  They'd have to spend quite s=
ome time learning the format and locations, and finally doing the tweaking =
to get what they actually want from their system. =20
>=20

You make a good point here.  Maybe, if I ever get around to picking up
AHWM maintenance (since its creator abandoned it), I'll create a GUI
configurator.  Of course, I don't really have much need for it -- but it
would be absurdly easy to do, I think.


> Part of the reason a lot of folks use FreeBSD is for its flexibility.  On=
e can do a great deal with a FreeBSD system.  It doesn't have to be taxing.=
  There's no sense in giving out "hardcore points" to people who expend tim=
e and energy doing something that can be done more efficiently through a UI=
 and without the learning curve. =20
>=20

"More efficiently" and "without the learning curve" are not correlated,
in my experience.  In fact, I find that usually they each get in the
other's way.  Exceptions include things like Web page design.

There's a far more significant learning curve for basic use of wmii than
for KDE, for instance, but once one gets past the learning curve wmii is
a far better productivity enhancer than KDE for many types of activity.
The same goes for Vim vs. Notepad, tcsh vs. DOS, and Mutt vs. Outlook
Express.


> >=20
> > In what way is the latter easier than the first? I see
> > none...
>=20
> The fact is that your opinion (and mine, for that matter) are fairly subj=
ective.  I've done things both ways - I was using FreeBSD before KDE and GN=
OME were at all widely used, and if you wanted a decent looking desktop tha=
t functioned the way you wanted to be most productive, you had to hack a co=
nfig file. =20
> That said, I just don't see how KDE's configuration system (as this is th=
e topic at hand in this thread) is at all counterintuitive. =20

My memory of KDE 3 is pretty sketchy, so I'll stick with KDE 4 on this
one:

I found KDE's configuration interface(s) very unintuitive recently.  So
called "intuitive" design is, to a significant degree, predicated upon
assumptions of familiarity.

Given a lack of familiarity, the time spent finding the options I needed
to customize the configuration of a KDE4 GUI last week by stumbling
around clicking on various things to see if that's how I find the way to
adjust behavior foo was at least comparable to, if not greater than, the
time I spent learning how to hack AHWM's configuration file.

In fact, I'd say that "easy" is much less subjective than "intuitive".
It's "easy" to create a new keyboard shortcut in AHWM's configuration
file using Vim, even if I had to get past an initial learning curve
before it became easy -- and I find hacking configuration files quite
intuitive, though part of the reason for that is, of course, the simple
fact that I do it a lot.


>=20
> As far as KDE4 being untested, I'd send you over to the KDE folks to let =
them set you straight on that.  The short of it is that you're just flat-ou=
t wrong. =20

It may be heavily tested, but in my experience, it is not *thoroughly*
tested.  It was . . . problematic, trying to get things to work properly,
in my case.  Turning off the desktop folder view was the only way to work
around the display problems with that widget last week, for example.  I,
personally, don't like desktop icons anyway -- but the computer I was
working with was for someone else, and the lack of desktop icons would be
kind of a burden on the person for whom the computer was intended.  Since
the folder view thing is KDE4's official way to do the desktop icon
thing, this seems like kind of a big deal to me.

I've never had problems of that kind with KDE's version 3, nor with MS
Windows.  Of course, I'd never trade that problem for the kinds of
problems I have had with MS Windows -- but this seems like just one more
piece of evidence of a step backwared from version 3.

For my purposes, KDE4 is beta software.


>=20
> At the end of the day, when you find bugs in closed-source software, you =
call the vendor and file a ticket.  With open-source software, since you ar=
en't paying anything, you ought to deal with bugs through the community.  B=
ug trackers for KDE exist.  So do mailing lists.  There's a community there=
 with people - usually unpaid volunteers - who are willing to help debug th=
e software, just as commercial software vendors have paid support staff for=
 such issues.  If you don't like free UNIX-like systems, you can buy a nice=
 Sun box and get Solaris support from Sun.  In fact, Sun's support has been=
 really good in my vast experience, so I'd even go so far as to recommend t=
his if what you want is that level of support.  Even Sun releases bugs some=
times though.  This is why they, like those of us in the open-source world,=
 release patches. =20
>=20

Indeed.  I agree with that -- as far as it goes.  KDE4 seems to have some
bigger bug problems than what I'd expect from supposedly release-worthy
software, though.


> This whole argument just strikes me as a lot of meaningless complaining i=
n lieu of actually productively trying to identify and fix bugs. =20

If you want to get involved in bug fixing, using a beta version is a
great idea.  If you don't have the time or inclination, a supposed
release version that feels like beta test software is not the answer.

--=20
Chad Perrin [ content licensed PDL: http://pdl.apotheon.org ]
Quoth Nat Torkington, on Perl internals: ". . . an interconnected mass
of livers and pancreas and lungs and little sharp pointy things and the
occasional exploding kidney."

--7JfCtLOvnd9MIVvH
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (FreeBSD)

iEYEARECAAYFAkkQ5fUACgkQ9mn/Pj01uKXZeQCfRQyXuSV20ix8MJ+XYIOxPMYh
CAwAoO471EQY8oDMKOmhNMPfyhWt0vTY
=wOmx
-----END PGP SIGNATURE-----

--7JfCtLOvnd9MIVvH--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20081105001653.GA2742>