...
- eGov WG, after performing lot of good work for Kantara has decided to self-shutdown given that it’s not active for a while and finding it hard to get members to attend; that vote has recently passed LC.
- Mark Hapner King reminded about the eIDAS Public Consultation on digital identity and trust https://ec.europa.eu/digital-single-market/en/news/digital-identity-and-trust-commission-launches-public-consultation-eidas-regulation . It was agreed that IAWG will be generating comments on it. Ken added that as the deadline for comments is October, the plan is to start the review at the end of August. Furthermore, it was requested to promote this review work on Kantara channels.
- Tom shared a couple of standards that are being worked on developed at the OpenID and also in FIRE WG. The FIRE WG specification may result in the need for something like criteria for mobile apps at some point along distant future.
...
- Richard explained that it was adopted the multiple parties practice from the FAL2 to work into the IAL and AAL SACs, in doing so we have pulled out those criteria which relates specifically to federal agencies or might also apply to RPs, in order to take the AAL criteria to their fullest implementation responding to absolutely normative criteria.
- He has updated all the tags because it's a substantial change. There is a new contiguous set, and the old ones will be here there at least for a year or so while we transition to the new ones.
- Changes are in red text; there have been a few changes which have affected level 2 because we've been more inclusive this time with federal agencies.
- We have around 30 to 40 new discrete criteria of AAL3.
- It was decided to defer the approval for next week.
...
- The xAL3 SACs will go as a package for 45-day Public Comment and IPR Review.
- Mark Hapner asked if in terms of the overall impact there are improvements or extensions; What was the actual objective for the changes in general? Richard responded that we have to go back to the NIST requirement for that. He thinks it's a question of demanding greater rigor or in some cases denying some of the authentication techniques that might be allowed at level 2 because they were not considered to be strong enough for level 3. As you move from one assurance level to the next stronger higher level, it's all a question of removing weak solutions and increasing the rigor. Richard pointed out that it was made an extension to the user guide in that document, in order to make the point that these criteria are simply Kantara's way of interpreting the normative statements which NIST have made and we don't offer an explanation of why those criteria should be. So to understand that you have to go back to NIST SP 800-63-3 appropriate volume.
Mark Hapner: What do you think the impact on RPs and CSPs will be too actually conform to these changes? Richard responded that it depends on the individuals, they may have already gone ahead read the NIST specs and implemented something which they believe to be IAL 3 conformant. They have to review these criteria that we've produced and consider whether they can fulfill them if they were to go through an assessment process. It's a question of the maturity with regard to this standard of the any particular organization.
...
b. Although IdPs are required by 800-63C to meet stringent security requirements, none are placed specifically on RPs. He's just basically saying that there should be relevant operational security requirements placed on RPs as well as the IdPs. Richard said that there are quite a few requirements on RPs in 63c. Tom Jones added that he believe this shouldn't be in C. It was clarified that IdP is a subset of what Kantara defines as a CSP. Tom Jones remarked that the fundamental question for the group is whether or not the ideas of mirroring the CSP with the IdP are the right way to go, or whether it would be better to just take that out of 63C, which would be his suggestion. It was agreed that for version 4, the recommendation is that they NIST consistently use those the roles and functions terms when they write their normative or even even the informative requirements, in other words be clear and consistent in their use across the documents and the example is IdP and CSP. Tom Jones will create a list of roles and will send it to Ken . so we can provide an example to NIST.
Action items:
- IAWG to review the AAL3 to approve it on 2020-08-06
- Tom and Mark H. King to provide Ken text with the suggestions for Revision 4, in light of Tom Barton comments.
...