[trustable-software] Re Git Auditing Tool and other topics

John Lewis john_lewis at mac.com
Wed Dec 21 11:11:33 UTC 2016


Hi,

Various comments

You may find this useful - https://codescene.io/about <https://codescene.io/about>  Adam has recently undertaken an audit of Linux.

Edmund - Much of this debate is covering old ground. We had the IAM issue on Eurofighter and it was covered in the ECMA work. That covered a single project / multiple sub-projects and companies. Effective IAM is very difficult outside of tightly controlled organisations. as gov.uk <http://gov.uk/> 

Jim - Safety and Security environments the are not tightly controlled are by definition not secure (I know you agree)

John Ellis - slashdot post. yes reuse does and one difference between safety-critical and security-critical code is that many software systems (e.g. fly-by-wire) that use voting (security critical apps don’t need voting) actually require review to ensure that code is NOT re-used.

Paul - I don’t know if you have seen this http://www.bcs.org/upload/pdf/safety-v-security-report.pdf <http://www.bcs.org/upload/pdf/safety-v-security-report.pdf> 

IMHO - Some good points but safety-critical and security critical are not the same (see point above re voting) but share many characteristics. 

One problem still not solved (to my knowledge) is "additive security”, where two items of information that are relatively low security, are when combined, much more sensitive. E.g. (Item 1) the computers at each site - site A, site B etc. in a NATO CCIS and (Item 2) Site A is xxxxx, Site B is yyyy. From the two, the topology of the entire CCIS can be determined (i.e. secret + secret= top secret). As an aside, this is another example of the difference between safety and security critical - most safety critical systems run at one security level.

Happy Christmas everyone
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.trustable.io/pipermail/trustable-software/attachments/20161221/9e79b662/attachment.html>


More information about the trustable-software mailing list