KDE3 vs KDE4 debate misses the point.

English Soapbox
#1

I have to say that much of the debate about whether KDE4 is ready to replace
KDE3 misses the point entirely.

First of all, the openSUSE installer for RC1 and 11.0 points out that KDE4
is much less stable than KDE 3.5. So I have not a lot of sympathy for those
complaining about KDE4 because KDE3.5 is, and always has been available.
Besides, it’s not the Suse developers we should complain to but the KDE
ones. What really annoys me is not getting bugs in KDE3 fixed because
they’re playing with their new toys. I could care less about 3d, plasmoids,
or fancy graphics - I’d rather see well documented existing bugs in Kmail
fixed.

Second, the disrespectful flaming in the forum was totally unnecessary.
There wasn’t any call for it, and all of you could have made your points
without it, whether you were frustrated or it. And the developers need to
realize that most of us are here because we really want openSUSE to
succeed, and when people complain they’re generally (but not universally)
trying to be constructive. And those of us who called them names should
know it usually doesn’t make people more willing to listen to your point of
view.

I really really like openSUSE 11.0. It’s been my distro of choice since 9.3,
when you got the excellent manuals with your CDs. I really like the speed
with which YAST works, now, although I wish they’d just left the existing
categories of software alone.

There are two areas about openSUSE 11.0 that disappointed me, however. The
first was the lack of attention apparently paid to getting wireless drivers
that worked out of the box for common laptops. I’ve been used to installing
Suse and having wireless work for years, now. I did not expect the hassle
of trying to get wireless working.

The second disappointment was that even though there were all kinds of
reports in the forums, on mailing lists, and in bugzilla about non-working
wireless cards, no one stopped and said hold on a moment, let’s fix this
first before going to general release. I thought that’s what releasing a
release candidate was for - to find and get the bugs out before shipping.

–
bob@rsmits.ca * You are in a maze of UUCP
(Robert Smits, Ladysmith BC) * connections, all alike…

#2

This is a a point-0(.0) release, so it is expected to have bugs

Almost all of the common drivers need proprietary firmware(with the exceptions of ath5k and intel’s OSS one(can’t remember it’s name at the moment)), though it would be very helpful if there was a script that could automatically detect the card, and download the correct firmware

This is a point-0 release, it can’t be expected to be too stable ;).

11.1 should fix some of these bugs.

Also, 11 is bringing many new technologies to the table, KDE4, PulseAudio, PackageKit, new Zypper, so obviously, they have to ‘settle down’ :wink:

#3

some-guy wrote:

>
> Robert Smits;1823706 Wrote:
>> I have to say that much of the debate about whether KDE4 is ready to
>> replace KDE3 misses the point entirely.
>>
>> First of all, the openSUSE installer for RC1 and 11.0 points out that
>> KDE4 is much less stable than KDE 3.5. So I have not a lot of sympathy
>> for those complaining about KDE4 because KDE3.5 is, and always has been
>> available.
>> Besides, it’s not the Suse developers we should complain to but the
>> KDE ones. What really annoys me is not getting bugs in KDE3 fixed because
>> they’re playing with their new toys. I could care less about 3d,
>> plasmoids, or fancy graphics - I’d rather see well documented existing
>> bugs in Kmail fixed.
> This is a a point-0(.0) release, so it is expected to have bugs

Hardly. openSUSE 11.0 may have that excuse, but KDE 3.5.9 certainly doesn’t.
And nothing in 3.5 is being fixed because they writing KDE4.

>> There are two areas about openSUSE 11.0 that disappointed me, however.
>> The first was the lack of attention apparently paid to getting wireless
>> drivers that worked out of the box for common laptops. I’ve been used to
>> installing Suse and having wireless work for years, now. I did not expect
>> the hassle of trying to get wireless working.

> Almost all of the common drivers need proprietary firmware(with the
> exceptions of ath5k and intel’s OSS one(can’t remember it’s name at the
> moment)), though it would be very helpful if there was a script that
> could automatically detect the card, and download the correct firmware

It certainly would. But it’s not just the drivers. Lots of stuff regarding
networking simply doesn’t work. You can’t use scpm together with knetwork
manager if you want to be able to change nfs mounts, for example. And look
at all the trouble with network manager and not being able to do wpa
encryption properly.

We’re way past the age when we should have to fiddle for hours to find a
solution for these kinds of problems with a general release. I can
understand this happening with a beta candidate, but it shouldn’t go to
general release until it’s ready.

>> The second disappointment was that even though there were all kinds of
>> reports in the forums, on mailing lists, and in bugzilla about
>> non-working wireless cards, no one stopped and said hold on a moment,
>> let’s fix this first before going to general release. I thought that’s
>> what releasing a release candidate was for - to find and get the bugs out
>> before shipping.

> This is a point-0 release, it can’t be expected to be too stable ;).

It’s not about stability, it’s about WORKING. Having non-functional wireless
is not just unstable, it’s broken. This isn’t just affecting a handful of
people with obscure equipment, it’s affecting a LOT of people with current
Toshiba and Lenovo laptops, for example.

In my view, far too much emphasis is being placed on speed of getting out a
release, and not enough to ensuring that the parts that are released are
working properly.

> 11.1 should fix some of these bugs.

I should hope so.

>
> Also, 11 is bringing many new technologies to the table, KDE4,
> PulseAudio, PackageKit, new Zypper, so obviously, they have to ‘settle
> down’ :wink:

Some of these are working very well. The increased speed of Yast alone makes
it worthwhile. But it shouldn’t be released when basic connectivity isn’t
there.

Nor am I enamoured of change for change’s sake, either. The “new” kDE menu
is one example I really don’t care for - it’s less functional than the old
one, but at least I can still switch to the old style.

I say to hell with gadgets of questionable use and would rather see
improvements in reliability, stability, and cleaning up known bugs.

–
bob@rsmits.ca * You are in a maze of UUCP
(Robert Smits, Ladysmith BC) * connections, all alike…

#4

some-guy schrieb:
> This is a a point-0(.0) release, so it is expected to have bugs

No, it isn’t. Dot-zero releases are not beta releases.

What’s more, this dot-zero release did have a beta phase, which
did reveal many of the issues that are causing complaints now.
Somebody must have decided to cut the final release with these
known issues unsolved.

Of course, every software has bugs, and if you wanted to resolve
all issues before shipping a release, you would never ship
anything. So the real question is: which issues are serious
enough that you shouldn’t ship a final release without fixing
them. But deciding that requires putting oneself in the shoes of
Joe R. User, which is obviously very difficult. A forum like
this one might be able to help getting the point of view of
Joe R. User on these issues across to those who have to make the
decision, provided everybody manages to suppress their urge to
flame.

–
Tilman Schmidt
Phoenix Software GmbH
Bonn, Germany

#5

I agree that more time to get Wifi working on install would be helpful. Even if only an option to do a scripted install of ndiswrapper so that wifi works somehow, native or not, after the install.

It is very hard to find out information on fixes of minor issues or updates without an internet connection. Especially since there may well come to be a time very soon where the average home user has only wireless access to the internet without a wired port at all.

I also agree that rational discourse, perhaps going so far as to express sour grapes in the form of, “What were they thinking?!” is a far better way to accomplish meaningful solutions than flaming. (I’m away for a week and all sorts of fun stuff happens - gotta get caught up on the flame wars! :smiley: )

#6

I am so sad that KDE4 is so unfriendly.

I’ve tried to create a desktop shortcut and after 10 minutes I gave up.

Why did they broke up the GUI usability so much?
The KDE3 was nice, stable and VERY productive.

KDE4 is the “vista” of desktop words. It’s also slow.
I used Stresos - operating system stress test to run some tests with different desktops (Gnome, KDE3, etc) and now I’m a Gnome user.

I am angry and sad about this.

#7

I’ve tried to create a desktop shortcut and after 10 minutes I gave up.

One hardly will find out how to meet certain needs by just fiddling around (not with highly configurable environments as KDE), but rather read the manual, search via google or simply ask. You can have classic icons on the desktop by setting the desktop-activity to “Folder View”.

Why did they broke up the GUI usability so much?

Because the KDE-team had a vision of a different desktop. KDE4 is not really a follower of KDE3, it’s something different and new.

KDE4 is the “vista” of desktop words. It’s also slow.

I use KDE4 fully since about half a year, yet I can not accomplish much when being forced to do something on a Vista desktop. They have nothing in common except the usual “X on top right to close a window” etc. (and it reminds me of people claiming KDE3 was like WinXP…). KDE4 runs pretty smooth too for me…

Please realize KDE4 is a project based on unsolicited work. If the KDE-team has some new vision about how a desktop should behave and look, they should feel free to turn that vision into reality. They have no duty at all to fulfill any users needs.

#8

gropiuskalle is correct. I have no problem putting icons on my desktop, although I chose to stick with the new way of having the little folder there and drop the icons in it. I just find it to look more tidy. Set to Desktop, it allows me to put icons in that folder and on the desktop as well.

Setting to Folder View, makes it look the old way with normal desktop icons. If that is the way you prefer, it will let you have it your way.

I loved 3.5, but I realized it is dead and I made up my mind to move on. It took a little tweaking, but I got 4.4 working quite nicely. I don’t have any plasma crashes and everything seems to work, including search. In fact, the new search is simply amazing. :open_mouth: I was playing with it last night and it is blazing fast. I was stunned. I don’t understand how they actually did it, but they did a great job.

If you have been using previous versions of 4.X and have not tried 4.4, you really should give it a shot. A lot of the little things have been fixed. I won’t call it perfect, but it has come a long way from the 4.0 days when this thread was started. It has come a long way from the 4.3.5 days of a few weeks ago. :slight_smile:

If you have issues, ask for help. Switching to Gnome and feeling sad doesn’t get the problems fixed. :wink:

#9

I say good riddance to desktop icons, but if you absolutely need them,
they can still be made. What I don’t understand is, if I find desktop icons
to be useless (mostly because I never see my desktop as it is covered with
windows, then how am I to make any use of ‘widgets’?? Answer is I don’t.
But I imagine for someone that really likes desktop icons that widgets
should be an awesome thing.

There is a certain learning curve moving from KDE3 to KDE4 and most of it
has to do with creating/moving icons and manupulation of the panel bars.
Not something you will likely figure out in 10 minutes, particularly if you
haven’t asked for help. Now, should it be that way? Probably not, but it
is so just ask for help. Once you learn it, it makes sense and I think
you’ll find it’s not really that fussy.

#10

I think the title of the thread says it all.

#11

Feh plasma is is a real pain in 4.4…
And the new widget thing is a pain in the a double s

#12

Just joined, one post… i smell a troll

#13

I do agree.

#14

Was it really neccessary to go necroing a two-year-old thread? :silly:
Look at the OP post date!

#15

No plasma crash here, and for me widget thing is fine.
Did you start new .kde4? hope that could solve it.

#16

Dead thread…

#17

You know, summarizing that KDE4 is the vista of KDE is exactly what I feel, there are so many bug and instability with dolphin crashing, with the fact that for example half of the time you have to open konqueror instead of dolphin, and this plasma desktop is a pain,a complete failure, what is the point of turning around icons?? and you cannot even select multiple icons!

KDE people should be ashamed, I LOVED KDE3.5 now I am forced to choose gnome!

KDE devs, if you read this, Shame on you!

#18

voltumna wrote:
> KDE devs, if you read this, Shame on you!

they won’t and if they did they wouldn’t care…

they are absolutely certain they made the right decision…

(you might notice [below] i’m still running KDE3–so, why not you?)

–
DenverD (Linux Counter 282315)
CAVEAT: http://is.gd/bpoMD
posted via NNTP w/TBird 2.0.0.23 | KDE 3.5.7 | openSUSE 10.3
2.6.22.19-0.4-default SMP i686
AMD Athlon 1 GB RAM | GeForce FX 5500 | ASRock K8Upgrade-760GX |
CMedia 9761 AC’97 Audio

#19

→ This would be quite a good reason why not:

WARNING: The KDE:KDE3 repository is currently not fully maintained and may not even update and build correctly. We also do not accept any bugreports for it. Use at your own risk - although currently (April 2010) it is still being regularly updated.

I for example experience heavy bugs with the KDE3-version of Gwenview once the kipi-plugins are installed, I suppose sooner or later more and more apps will be broken. Yet I agree with DenverD that as for now a KDE3-desktop is still usable (and will be at least for some more months). On the other hand: KDE4 is mature and stable - I really do not understand how one can still be unsatisfied with it general functionality (while the general concept is a matter of taste, of course).

#20

Here’s a risk too: one day there will be a reason to switch. Most of the times I see this in comparible situations, the switch to something new must be made, forced by non-solvable problems. I prefer to switch at a moment in time that’s my choice. That’s also the reason why I switch version, as soon as discontinuation is near.