Steve Lamb wrote: >Kent West wrote: > > >>1. Training oneself not to run things as root is one benefit of sudo, so >>that you don't mess up when you go to another machine. >> >> > > One presumes when you go to another machine you won't have root. > Hmm; I've got two machines here in the house, one of which is more or less single-user (which I use most of the time), the other is multi-user. In addition, I have a half-dozen other machines which I admin. So, I go to lots of machines on which I have root, some of which are "multi-user" and some of which are not.
I'd rather have a consistent habit across the machines. But that's just me. > Training >oneself to not run things as root is not a benefit of sudo, it is a benefit of >training one's self to not run things as root. > Which implies that you're firing up an xterm, "su"ing and doing your command, then exiting out of "su" after each command, never leaving a terminal window running as root most of the time. The converse would be that you tend to "run things as root" (such as a terminal window), which kind of indicates that you're ignoring your own "good practices" training. > Oddly enough it was a practice >in place well before sudo existed. However did they survive and train >themselves to do it? > > By inventing "sudo"? (It's just a joke ;-) ) > > >>2. Not logging into X as root is another benefit. Running a single X >>client/app as root is different than running all of X as root. >> >> > > Which does not require sudo. rxvt, su... > > Granted. But the original claim which I dispute was not that "other methods provide similar benefit to sudo"; the claim was that "sudo provides no benefit on a single-user machine", with which I disagree. >>3. Logging, provided by sudo, is not merely for the sake of knowing who >>did what; sometimes it's for who did what when, etc. >> >> > > Which was implicit in my statement that it provides logging; generally >timestamps are invovled. > > So, I'm confused. Are you saying that the logging capability of sudo provides a benefit on a single-user machine (my claim), or not (the original claim)? >>I'll grant that there may be considerably less reason to use sudo on a >>single-user machine, but to claim that there is "*NO* benefit of sudo" >>is simply incorrect. >> >> > > No, it is an opinion contrary to yours. > I stand by my statement. See below. > That alone doesn't make it >incorrect. However given the above you've not provided compelling arguments >that sudo provides any benefit to a single user who is de facto admin of his >own personal box. Good behaviors are good behaviors regardless of environment >and simply don't count. > > > If 99.9% of the nix-using population finds that sudo does not provide any benefit on a single-user machine, then I'd concede that sudo has no benefit on a single-user machine for most people. However, if one person, anywhere, finds sudo to be of benefit on a single-user machine, then the claim that there is "*NO* benefit of sudo" is simply incorrect. (I myself am such a person, so this is not just a hypothetical possibility.) Of course, the whole point of these recent posts is a "wrangling of words", being as I objected to the "absoluteness" of the original claim that sudo provides "*NO* benefit" on a single-user machine. If you wish to continue arguing that the benefit I find in using sudo on a single-user machine is in reality no benefit, then I humbly bow out for the sake of peace, and concede that I could very well be wrong. -- Kent -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
