A process having euid as its effective user id and egid as its effective group id is authorized to perform an operation requiring a privilege p provided (7.9) holds.

What are the consequences of replacing (7.9) with the following. (p PrivsF .owner euid = owner F ) (p PrivsF .group egid = groupF ) (p PrivsF .other ).

For reference I am posting 7.9 question,but I want the answer for the first question

7.9)Consider a fixed set {P1, P2, . . . , Pn} of principals, which also serve as the sole objects in the system. Suppose there are only two kinds of privileges. Privilege pp (for propagation permitted) does not propagate beyond its initial assignment in Auth; privilege t (for token) can propagate according to the following commands. grantToken(P, P0 ): command pre: invoker (P) hP, P0 , ppi Auth hP, P, ti Auth action: Auth := Auth {hP 0 , P0 , ti} takeToken(P, P0 ): command pre: invoker (P 0 ) hP, P0 , ppi Auth hP, P, ti Auth action: Auth := Auth {hP 0 , P0 , ti} There are no other commands.

(a) Is the privilege propagation problem for t decidable for this system? If it is, then give an efficient algorithm for deciding whether some sequence of grantToken and takeToken commands can cause hP, P, ti Auth to hold for any given initial value of Auth and principal P; if it is not decidable, then give a proof.

(b) If additional privileges may not be introduced, then are there commands that can be added to the system and have the answer to (a) change? If so, then describe those command(s); if not, then explain why.

(c) If additional privileges may be introduced, then are there commands that can be added to the system and have the answer to (a) change? If so, then describe those privileges and command(s); if not, then explain why.