> I am trying out Mandrake 9.1 and am very unhappy. Grip for instance > freezes the machine if i have too many apps going. Redhat is quite a bit > better in this aspect. (With the same amount of apps) RedHat has more resources to spend on testing and integrating code. The downside to this is RedHat tends to "blaze their own trail" occasionally, meaning that learning how to do it on RedHat does not necessarily mean learning how to do it in linux. > My point is that this 6 month cycle generally sucks. Can't we get a new > version once a year that is near perfect instead of pushing out 2 > versions a year that are never bulletproof ? If you waited for a perfect release, you would never get a release. Hell, WinNT 4.0 was released how many years ago and their still finding bugs. (Extra credit: Try to find an armored car company that sells a bullet proof car. You'll find they all sell "bullet resistant" cars) I've yet to have any stability issues with a RedHat release except with my software (I wanted to use jfs with nfs, despite warnings on the jfs site. On the plus side, I've helped the jfs team get the 1.1 release of jfs stable with nfs (been months since the last "freeze") Sometimes I've had to compile from source, or worst case adjust a build script or similar, but quite often RedHat's popularity means that folks make sure their stuff works with the latest RedHat. Of course, this recent jump from 8.0 to 9.0 was unusual, just as the release of a 7.3 version was considered unusual. Personally, I'd wait at least a month before deploying 9.0 into production (I'm in the middle of a 8.0 rollout, and will likely remain there for a while, barring big performance improvements), but my desktop box will likely be clean installed as soon as I can get my hands on 9.0. Bottom line, if you want stability, build a QA box and hammer it. This is a fundamental rule of IT. -- Psyche-list mailing list Psyche-list@xxxxxxxxxx https://listman.redhat.com/mailman/listinfo/psyche-list
