We have to remember, *anyone* can create relationships from or to a piece of data, not just the original producer. The original producer may not even know those relationships exist or have access to that information... and even if they do, they don't have permissions to update it. In a successful relationship model, people would be creating relationships everywhere, making a "web" of connected threat intelligence. However, If every time I publish an update to an object, all of it's relationships break (relationships which by the way I certainly do not have permission to update as I am not the producer, and which I may not even have access to viewing), this "web" is not going to happen, instead we will just have many disconnected threads. The only reasonable solution for this problem would be to have TAXII servers and other intel-repositories assume the job of updating all relationships transparently in the background whenever a new version comes along. But if we are assuming that - then why are we not just using implicit relationships in the first place?
We're moving a huge burden downstream. I also see no real benefit in this - as I pointed out in slack, because everything has a timestamp, even if we have implicit relationships it is not hard for a repository to support querying the object that existed when the relationship was first created if that is your aim (I still think this will be the far minority of actual real-world use cases)
-
Jason Keirstead
STSM, Product Architect, Security Intelligence, IBM Security Systems
www.ibm.com/security | www.securityintelligence.com
Without data, all you are is just another person with an opinion - Unknown
Trey Darley ---03/21/2016 08:49:30 AM---On 18.03.2016 23:36:16, Marlon.Taylor@us-cert.gov wrote: >From: Trey Darley <trey@soltra.com>
To: <Marlon.Taylor@us-cert.gov>
Cc: <cti@lists.oasis-open.org>
Date: 03/21/2016 08:49 AM
Subject: Re: [cti] Question Gathering: Relationship Preservation in Versioning (Implicit vs Explicit)
Sent by: <cti@lists.oasis-open.org>