Internet Threat Model (model-t)

Program Name Internet Threat Model
Acronym model-t
State Active
Dependencies Document dependency graph (SVG)
Personnel Chairs Jari Arkko 
Stephen Farrell 
Mailing list Address model-t@iab.org
To subscribe https://www.iab.org/mailman/listinfo/model-t
Archive https://mailarchive.ietf.org/arch/browse/model-t/

About

RFC3552/BCP72 describes the threat model that has been used in Internet protocol design. Since that was written however, the world has changed in terms of the threats experienced and in terms of how protocol endpoints are implemented and deployed. It is therefore timely to re-consider the threat model used in the design of IETF protocols, and there appears to be sufficient interest in doing so to justify an IAB program on this topic.

The model-t program provides an open venue for analysis of the Internet threat model and environment and has as a goal to produce a potential update to BCP72 that defines an Internet threat model that better matches today’s reality and/or better describes the current threat environment. Specifically, the intent is to document why an update is needed and provide a suggested update that could be considered by the IETF. A potential BCP72 update would likely extend the set of threats considered. Reducing the protection offered by current comsec mechanisms is a non-goal. Similarly, re-consideration of the parts of BCP72 that are outside the very narrow part on Internet threat model and the current threat environment are not in scope. (Should there be a desire within the program to do more, then consideration could be given to extending the scope and phasing the work, but initially, the scope is limited as described here.)

The potential update to BCP72 would be text offered to the IETF for consideration – it is not within the IAB’s remit, nor that of an IAB program, to directly modify a BCP. Nonetheless, an IAB program offers a good venue for this work, as it perhaps allows for more focus on the evolving architecture aspect within this space, compared to an IETF working group that aims to meet more specific milestones.

Updating BCP72 requires text that can be used by anyone developing an IETF specification. This imposes significant constraints on that text in terms of length, simplicity and the background knowledge required for the text to produce useful results. The program is therefore also likely to produce other documents (as RFCs or otherwise) that motivate or provide background materials to justify or further elucidate the proposed update to BCP72. Those background documents don’t have to adhere to the same constraints as an update to BCP72.

There is relatively broad agreement that the roles and trustwortiness of different endpoints in a protocol exchange are important considerations. In specific cases there may be quite different perspectives on the tradeoffs involved, e.g., to what extent privacy, information centralisation, or the needs of enterprise networks should be considered as the highest priority. And no doubt, other equally valid perspectives exist. The model-t program is not a venue for attempting to determine the one true perspective from which to tackle specific priorities. Our hope is that, regardless of what one considers highest priority, there is joint interest in taking these issues into consideration in protocol work, so that consensus for including this consideration in an updated BCP72 could be reached (in the IETF).

Participants of the program are drawn from the IAB and the rest of the community. Participation in this program is open to all interested parties, and considered analysis of diverse inputs is important. The IAB selects chairs for the program that are expected to facilitate discussions. Specific individuals may be invited to participate on particular topics from time to time based on their expertise without needing to commit to a long-term engagement.

Work is carried out on the model-t@iab.org mailing list that is open to any subscriber. Virtual meetings of the program are announced on the mailing list and physical meetings during an IETF meetings will, as appropriate, be published on the IETF meeting agenda (possibly as side-meetings).

The program is created for an initial six months, after which the IAB will review progress.

Expected outputs from the programmme include:

  • RFCs and other documents proposing, justifying and explaining new threat models and the current threat environment
  • RFCs and other documents analysing how BCP72 matches (or doesn’t match) today’s reality
  • Text to offer to the IETF as a possible update to BCP72