Linked by Eugenia Loli on Mon 17th Mar 2003 22:49 UTC
Graphics, User Interfaces So many operating systems and so many graphical desktop environments... This article is a comparison of the UI and usability of several Desktop Environments (DEs), that have been widely used, admired and reviled: Windows XP Luna, BeOS 6 (Dano/Zeta), Mac OS X Aqua and Unix's KDE and Gnome. Read on which one got our best score on our long term test and usage.
Permalink for comment
To read all comments associated with this story, please click here.
Review based on similar hardware???
by anonymous on Tue 18th Mar 2003 03:22 UTC

How about trying to run your reviews on the same hardware wherever possible?

Obviously you should be able to run Gnome vs. KDE vs. XP on the same machine. I'm not sure if BeOS has a version that runs on a semi current system ( never used it ). You could even try running Gnome vs. KDE on more than one linux or BSD install, just to see if there's something the distro is doing that's causing things to act flaky.

What I got from the review was that all of these were tried on different machines and that's part of the problem. Obviously you don't have a choice with OSX, but comparing BeOS, or KDE, or whatever, on a dual CPU system and then Gnome, or whatever, on a single CPU system is likely to give vastly different experiences. I think laptops were even mentioned at one point. I don't think I've EVER seen a laptop that I'd consider a "high performance machine".

If BeOS won't run on the same system, why not? I'd say if you can't load them all on the same machine, then there'd better be a reason, and it should be reflected in your scoring. OSX we already know the reason and I can't see why you'd deduct points for it. But if they all won't run on the same machine, then there's simply no way to get meaningful results here. Also, without testing on more than one linux distro or BSD, you cannot say that it is KDE or Gnome that is causing the problem with consistency or performance.

Load times are quite possibly at the mercy of the underlying system. It's not quite fair to say that the load time of Gnome app Y is slower than the load time of KDE app Z if you run Gnome under Gentoo and KDE on FreeBSD stock out of the box.

It may take a bit more effort, but I think it would create more consistent results. It should at least more consistent methodology which should help remove some of the bias comments. After all, at least your base starting point is the same.

Just my .02$