[cap-talk] EQ all over again (Re: Is this attack feasible in any real ocap systems?)

[Jed Donnelley]

On 4/4/2008 4:07 PM, Raoul Duke wrote:
>>  I'll let the anti-EQ people try to come up with a fix for the
>>  KeyKOS box that doesn't use EQ. :-)
> 
> the Grant Matcher pages make mention of the fact that distributed EQ
> is i guess fraught with some kinds of performance and/or security
> perils (i am mostly clueless about them). as another approach to the
> whole "avoid EQ" meme, are there ways one might improve on distributed
> EQ so that it wasn't so evil to make use of the feature?

I don't think so, because the problem with EQ is its power,
not its implementation.  The above would only make sense if
you were suggesting some sort of a weakened "EQ".  We have
some of those already (e.g. my "MyCap?"), but seem to keep
coming back to questions surrounding EQ - I think partly
because often an EQ-based solution to a problem is the
most immediate one that comes to mind.

--Jed  http://www.webstart.com/jed/