Darren J Moffat wrote: > jan damborsky wrote: >> Bart Smaalders wrote: >>> Jason Zhao wrote: >>>> By adding this, it will be convenient for administrator. >>>> Especially, when >>>> there is some packages were added by "pkgadd", but there is no >>>> index for >>>> these packages in /var/pkg, and most of important the administrator >>>> know >>>> what he needs to do and what he is doing. It will be easy to only >>>> add one >>>> single package, otherwise, it will be time consuming to add them all. >>> > >>> >>>> Anyway, it brings flexible works. >>> So does using a binary editor on your raw disk to modify your >>> filesystem. It still doesn't mean it's a good idea. >> >> But it definitely doesn't imply it is a bad idea. >> If I know what/why/how, I should be allowed to do this. >> I think that following statement better explains >> the philosophy I prefer as far as taking >> restrictive versus liberal approach is concerned: > > I agree with Bart here this is just not functionality that should be > in the pkg command by default. If you want it that badly implement > it yourself and keep it as a set of local patches. > > I personally think it is far too dangerous. > > Compare this to what what ZFS does. You can if you really don't are > about your data turn off checksuming for it on a per dataset basis, > but even if you do that ZFS doesn't allow you to turn it off for > metadata. > >> "/UNIX was not designed to stop its users from doing stupid things, >> as that would also stop them from doing clever things./" > > IMO that philosophy is old and out dated.
Yes, it is old, but I might disagree it is out dated. > Users don't put their hands up and say "sorry I did a stupid thing" > they shout and scream at the vendor and blame the system for not > protecting them. Well, but since other tools on the system allow to do "stupid things", I can imagine, that if/when user runs into scenario which is not allowed to be solved by pkg itself, other ways will be found and the result will be even worse, since you will need to just guess, what was actually done. On the other hand, if you provide a way to do it using standard tools, you have control over how it will be done. And when user starts to scream, you have records about where/what/how and you can start asking why. In the former case, you might be out out of luck, since there are no records/history you might inspect. > > We have do have to stop users doing stupid things because stupid > things can lead to data corruption, security vulnerabilities and > pissed of people. I agree this that the default behavior should take care of this. But I am not convinced it should be the only one implemented. > > Having a deliberately corrupted package database (which his what you > have if you ignore dependecies) could lead to all sorts of problems > down the line. > _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
