odd_parity

Ever since KDE team announced their first version of KDE 4, I have been wondering where they're heading.

I am a hardcore user of KDE. I've been enjoying it for years, and even did Croatian translations back then in '98. for one of earlier KDE incarnations (you can find me on KDE Credits page). That would, I hope, fix my position as KDE user for a number of years and a number of versions, and give a perspective for my rant as well.

I have been enjoying good work done by KDE developers for years. It wasn't always a smooth ride, there were some hiccups every now and then, but nothing that would drive me away from my choice of desktop. For years, I have enjoyed pretty stable desktop implementation, that has come to mature as one of the best Linux desktops.

I like my KDE 3.5. It's good. It's quite stable. I can rely on it. I have been using versions of KDE 3 for six years, and that's a lot of time. What I do admire the most is the peace of mind I had while I was using a familiar, reasonably ergonomic KDE 3 interface. I was confident that I can concentrate on my daily tasks. That's important – to think about your work, not about your computer. (and I'm still happily using it today because of that virtues of his)

As the years passed by, there came Mac OS X and Windows Vista. Shiny. Shiny, shiny! I mean, look at those colors! Look at those widgets! It's moving! And the playful tunes emanating from speakers!

It seems that KDE people had to follow.

It was touted as better KDE than KDE – a new version that would blow your mind and make all other desktops pale in comparison. Something that would make Steve Jobs gnash his teeth, and force Bill Gates into early retirement.

It turned out to be – a bad copy of Windows Vista!

Please keep in mind that my desktop is my business tool. It has this particular importance for me. I see it as a tool to help me being more productive. Therefore, I am not one of those “ooh, shiny” people. It is nice to have visually attractive renderings on your screen, but they shall not be the most prominent thing you are looking at.

I do understand that things have to change over time, that KDE 3 is a mature, therefore conservative and out-of-fashion desktop, and that young people want something better. I believe there are lots of people that just like Plasma (and even I find it interesting concept), but it is too obtrusive for me. Plasmoids are nice drawn, they have those colors and you can turn them off, but still I find my trusty gkrellm from the Cretaceous Period - if not that beautiful as Plasma - much less obtrusive.

I am used to see the bottom end of my screen filled with buttons, virtual screens and window lists, and even that copy of Windows system tray. KDE4 is trying to mimic it, but it isn't done well. Can't say much about Phonon and Solid, for I haven't turned my eyes on them. It's because of that feeling that there's something under the hood that just keeps hummin' and brummin' and you can rely on it.

The first time I have tried KDE 4, it was just released as a new cool thing. There was a beta testing period, sure, so this KDE 4.0 thing was given away as a polished, final version of new Linux desktop.

And what a steaming pile of garbage it was!

I have tried it at two computers, one at work and one at home. The one at work kept giving me all sorts of strange things. Home version just wouldn't stretch to 1920x1200. Plasma would crash and burn often. My room was full of smoke(*). Not just once there was just a blank screen in front of me, and all the windows were gone to /dev/null.

KDE 4.0 - background dragged with mouse and Icons in state of disorder; all in all, not the most stable KDE release...

KDE 4.0 was released way prematurely. I believe that people who coded that thing thought that they're releasing something that is going to take over the world overnight. I remember reading developers blogs full of happy stories of finally releasing the next big thing since the sliced bread. I remember cheers coming from the KDE team.

But, that KDE 4.0 bit the dust, real hard. I was quite unsatisfied with the difference between the marketing and the result. Before the release it was all how Qt4-ish it is, how it is going to replace old technology, how shiny, wobbly, exploding and generally 3D-ish it is going to be... in general, KDE 4 is going to be The Desktop, the best experience user could possibly have with that state-of-the-art technology.

Well, tell me about the user experience! Bad, unstable, crashing, burning – and that for me has just one meaning: not suitable for work. And indeed, KDE 4.0 would produce issues at such rate that I just couldn't feel easy about it and concentrate on my work. I gave it a few working days, just to move back to KDE 3.5.

And I wasn't alone, as it seems. Critique poured over every side of the world, and that was the word a developers should fear the most – word from your customer, your user. A lot of people simply stated that KDE 4.0 was way too immature to be called production ready. In any smart company the management would listen up. KDE developers seemed defensive in their position of KDE 4.0 being a new way of thinking, something of such advanced concepts that people can't fully understand it, and that is the cause of their dissatisfaction. And of course, once they get to understand how great their work is, everyone is going to just love it. Something like Steve Jobs distortion field. But, there's just one Jobs.

People did not see the inner beauty of new KDE, what they saw was unstable futuristic release with just a small subset of functionality they already had in KDE 3.

This is where KDE marketing failed big time. Developers were so self-satisfied with what they have done, that they decided to rush out with half-baked product. I think it might be the collective ego thing at work. They could release something like KDE 4 Technology Concept, KDE 4 alpha, or something like that. They could have say: “Look, people, we have something so great and shiny! It's not for the average user yet, but we're working on it! Please, take a look!” They would get great feedback if they did that.

Instead, they decided to release it as a final product.

They – didn't impress.

And some users got emotional. That, in turn, made some developers get emotional, with the result of a small drama on developers mailing list with some of the developers getting so angry by some user rants, that they declared “who needs users,anyway?”

Ooops.

That was emo thinking, Ok. But that was the most stupid thing a developer could say. In such a big project as KDE is, users matter. KDE is not a personal itch you scratch, it is a huge undertaking that is being done for the community – read: users. A developer who wrote some small python script to solve some everyday task might not care about users, but KDE developers can't. If they aren't playing nice with users and don't listen to what they have to say, their users will switch – to Gnome, for example. Or any other desktop they're feeling comfortable with. KDE is just as strong as their user base is. The reason for floating on the top of the desktop niche is that lot of people use it. And that is the only way to stay afloat.

So KDE 4 did remind me so much on Windows Vista. Vista, too, was lauded as the best shiny thing a PC owner could have, and it was lauded by the very same people who wrote it. Users, on the other side, didn't find the marketing on par with their expectations (inflated, in part, with aggressive marketing done by Microsoft) and generally gave Vista bad scores. What Microsoft did was to deny the voice of users and stand ground on the terms of the best OS world has ever seen. But, that didn't help, and that caused quite a trouble in Redmond. People kept rejecting Vista because they felt that they're being ripped off for just a few shiny windows and nothing substantially new. Even today, sales of Vista are satisfying probably only because of OEM licensing. XP is still very, very strong.

But, if you ask Microsoft, they'll tell you that XP is an old technology, and that you really should switch to Vista, because they say it's better.

KDE 4 is looking like Open Source Vista – a product that isn't quite mature, but it was prematurely released; later on it was marketed as developers only version, but the damage has already been done. Lots of people installed it and found it lacking in many ways. The unwise decision to market it as a stable version instead of straightforward developers release or technology preview produced lots of attention, but that backfired. Emo statements on developers mailing list didn't help, either. Enthusiasm on developers blogs continued, but users reviews weren't that much enthusiastic. It is still being heavily developed, but the stupid misconception with release versions did quite a bit of damage, as it seems that the release was too early to be stable, but it was pushed by someone, probably developers eager to show the world their masterpiece, a Klassik. It turned out to be a Kraken.

Luckily, the story doesn't end here. Developers are a stubborn bunch of people, so they didn't give up on first negative reviews. They continued their work, steadily improving KDE 4. After some time, KDE 4.1 got out, and it was much better than previous version. I still had serious issues with it, but at least I wasn't able to drag the desktop background around. :-D

KDE 4.1 kept crashing on me, sorry. I had other serious issues, like big screen support. I gave it another few work days to see if I can fit in, but it wasn't the case. KDE 4.1 was still too buggy.

Then, there came KDE 4.1.1. - a bugfix and performance update. This is, at the moment, latest in KDE 4 technology. And it is slowly coming together. It has much more features of KDE 3 built in, and most of my old productivity tricks were replicable in this release. However, I still think that I do want icons on my desktop, and not in the folder view; Plasmoids are still randomly crashing, and the most annoying thing – inability to move icons and plasmoids in bottom bar, coupled with the strange quirk of disappearing buttons, makes it still way too unstable for the casual user.

On the other hand, improvements are clearly visible. I do believe that KDE 4.2 (expected in early 2009.) might actually be usable in production environment. In the meantime, however, developers might have to bear the title of “makers of Open Source Vista”.

(update 30.04.2009. - checked KDE 4.2.2 - nope, still not there!)

P.S. One of the reasons for this rant is that some people and me are writing a book about using Linux in SOHO environment. Among the decisions to be made in preparation of the book was the question of what will the default desktop manager be? The book should come out in fortunate time of release of several interesting applications, namely OpenOffice.org 3.0 and Firefox 3.1, so the book should be quite up-to-date with newest on FLOSS software. However, since the book is aimed at small businesses, we have decided to take somewhat conservative approach with Ubuntu 8.04 LTS, as we reckon that business will not want to upgrade their distro every few months; therefore LTS version.

We hoped that the book would take KDE4 as the chosen desktop manager, but KDE4 is way too immature for business, so we turned to KDE 3.5 as the desktop of choice. It is mature, really stable and familiar to most business users who have/had to deal with Windows XP on a daily basis. Switching them to KDE 3.5 is much easier venture, because the interface is much more familiar and once they're there, I don't think they would switch to another desktop manager. After all, they need something that will help them being more productive, not distract them or frustrate them. For that, I still see KDE 3.5 as the best choice. And that might stay so for the next few years.

(*) There is a nice and useless Plasmoid for someone to make – if a process crashes, smoke should fill the screen and the dreaded “smoked window” should float from the background with an error message. I bet neither OS X nor Vista have anything like it.