[cap-talk] erights wiki : ambient authority

[David-Sarah Hopwood]

Matej Kosik wrote:
> David-Sarah Hopwood wrote:
>> Matej Kosik wrote:
>>> I have still doubts concerning the article:
>>> http://wiki.erights.org/wiki/Ambient_authority
>>>
>>> What is relationship between:
>>> - ambient authority systems
>>> - systems where protection is made by ACLs.
>>> Aren't they defined in the same way?
>>
>> No, not at all. [...]
>> The fact that very little of the possible access control design space
>> has actually been effectively explored in popular systems, shouldn't
>> limit us to considering just that explored space when defining terminology.
> 
> I am not sure at the moment that I can imagine also those other options.
> Is it indeed impossible to find their concrete instances?

The Iguana API in L4 is a concrete example of an ambient request API that
does not use ACLs:
<http://www.disy.cse.unsw.edu.au/Software/Iguana/>

It has been discussed previously on this list:
<http://www.eros-os.org/pipermail/cap-talk/2005-May/thread.html#3631>.
(That whole thread is worth re-reading, I think.)

> Even though the article is still incomplete (namely the text that
> follows the definition) I think it makes sense to further simplify the
> quoted definition:
> http://wiki.erights.org/wiki/Ambient_authority
> The definition simply defines what is `ambient authority' and does not
> define different terms.
> 
> Is the definition correct in this shorter form?

Yes, this is much better.

-- 
David-Sarah Hopwood  ⚥  http://davidsarah.livejournal.com