When I try to run yast2, I get > sudo yast2 Error: "/var/tmp/kdecache-law" is owned by uid 5013 instead of uid 0. --- So I think...ok...that's weird, it wants law's kdecache to be owned by root.. ok, I'm game (changed ownership)... Then I go to do a kernel build (another kde using app)...which I usually build as normal user... scripts/kconfig/qconf Kconfig # # using defaults found in arch/x86/configs/x86_64_defconfig # Error: "/var/tmp/kdecache-law" is owned by uid 0 instead of uid 5013. Error: "/var/tmp/kdecache-law" is owned by uid 0 instead of uid 5013. Error: "/tmp/kde-law" is owned by uid 0 instead of uid 5013. Error: "/tmp/kde-law" is owned by uid 0 instead of uid 5013. Um.... I think yast2 is the one at fault here... It shouldn't expect that /var/tmp/kdecache-law would be owned by other than law!... or is that just silly? This worked on 11.4. I think 12.1 should be declared null and void as far as moving the support dates for previous releases 'forward', since from my own experience, and others, it doesn't seem it was read to be released, yet was because "It was the date". Maybe we should move more toward a linux kernel model -- aim for 3-4 months, but if it isn't ready then, then it won't be shipped until it's quality!.. even if it means a 1-month delay -- and that DOES push the next release out by that amount. It is **impossible** -- i.e not rational, to expect that one can know how much work something will take and fit it to exactly a preset time period. This is a delusion mostly pushed down from marketers and managers who want control over a chaotic process that is -- sorry, becoming increasingly chaotic as the number of software packages grows and the number of new programmers who have no experience enter the field and generate whole knew sets of popular packages with bugs that were fixed in older software 40 years ago. I need to start keeping a log of how many times a "new idea" comes up or a new methodology" comes up that is just a repackaging of a previous --- not that new ideas are impossible, but since most of the people writing software have no software background, they don't know what has been done before, so of course it is new! It's rare that I find a computer science graduate actually doing software engineering these days...which is fine -- humans are very adaptable creatures, but it does mean many of them will re-invent wheels and re-invent the same mistakes that were done a generation ago. There is little "institutional knowledge" that is carried over from generation to generation due to software engineering not being treated like those requiring similar training and operating in fields of similar complexity. Nor are there any standards for who is qualified to write code nor for code quality. It's not like you can call a building inspector to inspect to see if your code meets 'Code', and it's not like you have to be vetted by some process (journeyman training with plumbers/electricians), or lawyers and doctors studying under seniors -- all of those complex professions have requirements of working under senior partners to make sure you get trained. Software engineering doesn't have that. As a result -- the chaos at each generation (and each release) is inescapably more complex than the previous. To expect each to take the same finite amount of time is irrational. That doesn't mean each release HAS to take longer, but the scope of what is done may need to be narrowed to fit the time allotted. This isn't being done and quality is going down the tubes (noticed first with 11.4 in terms of ease of upgrade)... This wasn't meant to be about these larger issues -- just that yast2 is broken in expecting a user-specific kde dir to be owned by root. Feel freel to discuss the larger issue -- but maybe the subjects should split to help people who want to discuss either (please prune my verbosity appropriately). Linda -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org