Marc’s Mind

I’ve been running Linux as my primary desktop since about 1996 (switching from OS/2 when IBM started dropping support for it), and running Linux servers since early 1994 (my first Slackware system was running Linux Kernel 0.99pl15). Over the years I’ve tried many of the various distributions including Slackware (which is what I started with), Caldera, Debian, SuSE, and of course RedHat. Back around RedHat Linux 3 I switched and have been using RedHat distros for my desktop almost exclusively since then. There have been a lot of frustrations with RedHat over the years, but with the Fedora Project things have gotten much better. The switch to Yum for package management and 3rd party repositories such as Livna have made things very easy to upgrade and maintain. Every 6 months I watch to see what’s new in the latest Fedora release. Typically I’m a version or two behind since I want my computer to be stable. Right now at home I’m on Fedora 7 and Fedora 8 at the office.

I’ve switched window managers quite a few times over the years, too. The first one I really started using was fvwm, then I found AfterStep. I used AfterStep for several years. KDE was still too bloated back then and not quite feature rich enough, Gnome 1.x was in the same boat as KDE at the time. AfterStep stayed out of my way and was customizable-enough to let me get my work done. I like keyboard hot-keys for just about everything and AfterStep fit the bill. The only thing really lacking in AfterStep was a good way to interact with the desktop. I still missed a lot of the features that I had in OS/2’s Workplace Shell. When KDE 3.x and Gnome 2.x came out I switched back and forth between the two as each started advancing beyond the other. About 2 years ago I finally switched and stayed with KDE. My biggest reason was one that may sound silly, but when you’re managing dozens of servers and have everything hot-keyed it makes a big difference: Smart Window Placement. With Gnome 1.x I ran sawfish as the underlying window manager. With the 2.x series they made it very difficult to run anything except Metacity which has almost brain-dead window placement.

Back in February I looked at the road map for Fedora 9 and noticed that KDE 4 was going to be in there. I started looking at the some of the cool new features, specifically Plasma, the framework for allowing active widgets (plasmoids) to be running on the desktop or the panel. Its a big step beyond the little docklets that Gnome, KDE and even Windows have now. From what I understand its a bit more Mac-like (I’ve never used OS X so I can’t say for sure).

So I waited for about 2 weeks after Fedora 9 was released to let the first round of patches and bug-fixes to make it out. On Thursday I took the plunge and upgraded to Fedora 9 and KDE 4. I finished the install and started using it by mid-afternoon, exploring the new aspects of it and trying to figure out how to re-implement some of the things I use the most with KDE or Gnome. On Friday at about noon I re-installed Fedora 8 with KDE 3.5.

KDE 4 is not ready for widespread use. I probably would have left Fedora 9 on there and used Gnome instead, but Metacity is still retarded about its window placement.

KDE 4 is gorgeous to look at. Very smooth and clean with lots of eye-candy and desktop effects — when they work. I crashed various KDE components a number of times. Many of the crash issues I could live with since the apps restarted. It would have been an annoyance, but I could have suffered through a few more update cycles.

What I couldn’t live without, though was a way to either hot-key or mouse click to get the list of active windows. I hate the taskbar window list. Its one of the first things I turn off. I prefer to either hit a hotkey or more often, middle click on my desktop or click on the window list icon in my panel and then pick the window. Its how I’ve worked for the last 10 years and its something I can’t live without. In KDE 4 there is no way to show the open windows except for the taskbar. There are no plasmoids that will provide this same feature, and with the architecture of KDE 4, I don’t think there ever will be. The concept of the desktop has been turned into a second workspace for running programs on. The only difference is that application windows will always cover them and there is only one place where plasmoids run for all of your virtual desktops. The concept is promising, but I think they’re headed in the wrong direction.

I don’t have any problem with improving the appearance of the desktop. Far from it, I like the cool widgets and the flash. But not if its going to hamstring functionality.

With how different KDE 4 is from its predecessors, and how much it is lacking in features compared to the much more mature KDE 3.5, it shouldn’t have been called KDE 4. I expect a new desktop project to emerge from this, either renaming KDE 4 or a fork of KDE 3 by another group, but this won’t be embraced by business or power users/programmers like myself (I’ve been programming since 1982).

There were a long list of annoyances that drove me to going back, both with the distro itself, KDE 4 and Gnome. I know that my productivity would have dropped just by running this on my desktop, and thats bad.

In the 14 years I’ve been running Linux on my computers this was the first time I’ve ever downgraded to a previous OS. I won’t run Fedora 9 and won’t recommend it to anyone. I hope that the Fedora, KDE and Gnome teams can learn from this and not just trudge forward.

marc Rants, Reviews

Okay. I know that Seattle doesn’t top the list for the worst drivers in the country, but on some mornings it sure feels like they’ve all moved here and decided to take the same route to work that I do.

I have to know something, and I’m sure the question has been asked before, but why can’t people make the connection between slow traffic and their own behaviour? Since I-405 is usually a parking lot once you get within 2-3 miles north or south of Bellevue I tend to take the back roads in from Bothell. Most mornings are just fine. Its a nice easy commute only a couple of spots where there are any real slowdowns and those are usually from all of the road construction that happens during the period between April and September.

But I digress. Construction Season (that season that falls between Winter in the Seattle area) is a whole different rant.

My question is this: Why can’t people make the connection that when the right lane ends, and they try to zoom past the considerate people in the left lane to get that extra car length ahead, and then everyone has to slow down or stop to let them in that they are the ones causing the slowdown? Once you get past the point where the two lanes turn into one, things start moving again. Where there are two lanes, its touch and go for a mile. *sigh* If these people would just get into the left lane a mile, or even a half-mile before the right lane ends, traffic would go smoothly and not stop.

Of course, being in stop and go traffic like that does have its high-points. I’ve learned to keep myself entertained during these commutes. You can, for example, watch people gesture with their hands while talking on their cell phone. You can watch those very talented individuals put on eyeliner while sitting an inch away from the rear view mirror and trying to go around a turn at 10-15MPH. You can watch people bee-bopping to their favorite songs and feeling the beat of the tire driving over the bumps separating the lanes. And though I’ve heard it in a joke before, I actually have seen someone shaving, talking on the phone, and drinking coffee all at once.

Oh, my favorite part is that these people — the ones causing the slowdowns — are the ones who have a tendancy to get most upset about the slowdowns.

I’ve changed my mind. If I were to understand why these people do what they do, I’d probably end up with a brain aneurism.

marc Rants