From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bedros Hanounik <2bedros@gmail.com> Subject: trusted processes Date: Thu, 12 May 2005 17:19:00 -0700 Message-ID: Reply-To: Bedros Hanounik <2bedros@gmail.com> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_806_24116259.1115943540506" Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com List-Id: To: reiserfs-list@namesys.com ------=_Part_806_24116259.1115943540506 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline last year, I asked a question on this list whether we could have file=20 permissions for programs (or processes) in addition to users and groups. we= =20 need this feature to reject malicious code from accessing system files. Microsoft has a tech paper about having what they call trusted processes.= =20 you can find it here=20 http://msdn.microsoft.com/mobility/default.aspx?pull=3D/library/en-us/dnppc= gen/html/wmsecurity.asp I don't know for how long they've been working on this, but I bet I came up= =20 with this concept first. what do you think guys of implementing such feature; should it be in the=20 file system, kernel, or both.=20 How hard is it to implement, besides the complexity of authentication=20 management. -B ------=_Part_806_24116259.1115943540506 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline last year, I asked a question on this list whether we could have file permissions for programs (or processes) in addition to users and groups. we need this feature to reject malicious code from accessing system files.

Microsoft has a tech paper about having what they call trusted processes. y= ou can find it here

http://msdn.microsoft.com/mobility/defa= ult.aspx?pull=3D/library/en-us/dnppcgen/html/wmsecurity.asp

I don't know for how long they've been working on this, but I bet I came up= with this concept first.

what do you think guys of implementing such feature; should it be in the fi= le system, kernel, or both.

How hard is it to implement, besides the complexity of authentication manag= ement.

-B
 
------=_Part_806_24116259.1115943540506--