At 2:49 PM -0800 12/18/98, Bill Frantz wrote:
>At 01:34 PM 12/18/98 -0500, shapj@us.ibm.com wrote:
>>I am considering a seemingly simple security question, and I am not
>>confident about my outcome. Please note that this question is not meant to
>>be specific to E -- distribution and proxy issues might well impact the
>>outcome, and I do not want to consider that at the moment.
>>
>>Given two capabilities A and B, I can see no negative security implications
>>to permitting an application to perform the 'eq' test (are these
>>capabilities identical?)
>
>[#] The KeyKOS answer is yes. The "EQ" key was generally available.
>
[+] See <http://www.mediacity.com/~norm/CapTheory/EQParadox.html> for my recent arguments for general availablity of EQ. All of the arguments that I know for and against distributing EQ are anecdotal. The scenarios against EQ seem incomplete to me. Synergy, <http://www.mediacity.com/~norm/CapTheory/Synergy.html>, seems to me to be the prime argument for EQ, although synergy may be had in otherways instead.
Norman Hardy <http://www.mediacity.com/~norm>