RE: conference call - March 29

Title: 

Hal,

 

    I agree, essentially what we get 
down to is an authentication mechanism based on a shared secret. This is fine 
for bilateral relationships but does not scale.

 

    When it comes to authenticating the 
principal in a multi-domain setting the only way I know that scales is to use a 
credential that does not require the secret to be revealed or pre-shared to 
demonstrate knowledge of it, i.e. public key.

 

        
Phill

 

Phillip Hallam-Baker
Principal Scientist
VeriSign 
Inc.

781 245 6996 x227

  
-----Original Message-----
From: Hal Lockhart 
  [mailto:]
Sent: Friday, March 30, 2001 
  2:31 PM
To: 'Philip Hallam-Baker'; ''Security-Core 
  (E-mail)'
Subject: RE: conference call - March 
  29

  
In 
  the 0.3 version of the document, I am unclear about the box at the top of page 
  11 that starts "Alternate means of identifying the assertion" Are these design 
  alternatives or are you proposing to support a number of different ways of 
  identifying the same assertion?

  
 

  
Without regard to the answer to that question, have you considered the 
  alternative of using tickets which are simple references and contain no 
  information in themselves? For example, create a random number X, use the hash 
  of X as the assertion id, put the domain and X in a cookie (or 
  wherever).

  
 

  
The 
  problem with using a reversible encryption of the ticket is that given that 
  SAML will be used for interop between a large number of independant orgs, the 
  key will be a) potentially known to a lot of people and b) hard to change. 
  Unless we invent some key management protocol this will make it easy for 
  attackers to spoof tickets.

  
 

  
Hal

  
    
-----Original Message-----
From: Philip Hallam-Baker 
    [mailto:]
Sent: Wednesday, March 28, 2001 4:04 
    PM
To: Philip Hallam-Baker; ''Security-Core 
    (E-mail)'
Subject: RE: conference call - March 
    29

    
All,

    
 

    
    I am still working on the 
    attached trying to get it into some sort of shape. Problem is trying not to 
    spend time writing stuff that should flow from the use cases and 
    requirements.

    
 

    
    I'll try to get a version 0.4 
    for tommorow.

    
 

    
        
    Phill

    
 

    
Phillip Hallam-Baker
Principal Scientist
VeriSign 
    Inc.

781 245 6996 x227

    
      
-----Original Message-----
From: Philip Hallam-Baker 
      [mailto:]
Sent: Tuesday, March 27, 2001 2:41 
      PM
To: ''Security-Core (E-mail)'
Subject: FW: 
      conference call - March 29

      

March 29th 
12pm EST 

Running time :appx. 
      1 hour 
Confirmation # 8532838 
      
Dial in number : 1.904.779.4702 

      
 

Phillip Hallam-Baker (E-mail).vcf