Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!newstop!texsun!playroom!playroom.East.Sun.COM!cliffs From: cliffs@playroom.East.Sun.COM (Clifford C. Skolnick) Newsgroups: comp.unix.internals Subject: Re: Unix files should have both real and effective ids for files too Message-ID: <513@playroom.UUCP> Date: 13 Dec 90 21:15:35 GMT References: <1990Dec5.135759.12508@noao.edu> <1990Dec6.005358.6336@dg-rtp.dg.com> <109958@convex.convex.com> <14526:Dec711:12:5790@kramden.acf.nyu.edu> Sender: news@playroom.UUCP Organization: Sun Microsystems, Inc. Rochester, NY Lines: 25 In article you write: > >Final observation: adding read/effective user ids to any UNIX kernel >probably requires half a dozen lines spread across a couple of headers >and in the stat(2) implementation, a few recompilations, and >reformatting and reloading filesystems. Frankly not a lot of work. Not everyone has the source to everything. What about software companies that have gone out of business? You could keep the only binary interface alive in the kernel, but you have no idea what the stat() should put in the UID field since you don't know what the application is going to do with it. This is a good example of a simple kernel change that will break many applications, too complicated. As a note, I do think this is a good idea but I would rather see the addition of access control lists to UNIX. Would provide all this and more. If your gonna change and break things, might as well make it worth it! Cliff -- Cliff Skolnick | "You give me the reason, you give me control. cliffs@sun.com | I gave you my purity, my purity you stole. Am I (716) 385-5049 | just too stupid to realize stale incense old sweat I think. I am. | and lies lies lies." -- Nine Inch Nails