Re: Capability modelling tools

On 27 April 2011 00:55, Toby Murray <[email protected]> wrote:
> On 26 April 2011 19:17, Thomas Leonard <[email protected]> wrote:
>> Hi all,
>>
>> I'm doing a bit of modelling work at the moment, looking to see how
>> capabilities may propagate through a system, and the effects of
>> various components being compromised, etc. I found the Scollar and
>> Authodox tools for this, but they didn't quite fit what I wanted to
>> do.
[...]
>> In the end, I wrote a little modeller of my own. It's similar to
>> Scollar in many ways, but it also models "invocations" (not just
>> objects). An invocation is a particular call to a method, representing
>> the stack-frame that is created and the local variables of this stack
>> frame. You can then (separately) specify how the real invocations are
>> to be aggregated.
>
> Have you thought about formalising the approach your modeller
> implements, and proving whether aggregations are sound with respect to
> this (hypothetical) formal model?

After reading up a bit more about this, I realised I was basically
just doing a standard points-to analysis (although with a slightly
unusual choice of context). However, I did write an article about it,
which has now been published:

Thomas Leonard, Martin Hall-May, and Mike Surridge. 2013. Modelling
Access Propagation in Dynamic Systems. ACM Trans. Inf. Syst. Secur.
16, 2, Article 5 (September 2013), 31 pages.
DOI=10.1145/2516951.2516952 http://doi.acm.org/10.1145/2516951.2516952

It should be possible to download it from this link:

http://roscidus.com/blog/about/#the-serscis-access-modeller

However, I'm not working on SAM at the moment as I no longer work at
the university.


-- 
Dr Thomas Leonard        http://0install.net/
GPG: 9242 9807 C985 3C07 44A6  8B9A AE07 8280 59A5 3CC1
GPG: DA98 25AE CAD0 8975 7CDA  BD8E 0713 3F96 CA74 D8BA