= January 8, 2014 Meeting Minutes = APPROVED 22-Jan-2014 by PKCS 11 TC
Roll call taken by Bob Griffin. Quorum was achieved.
- Opening remarks (co-chairs)
- Roll call (secretary)
- Review / approval of the agenda
- Review of previous meeting minutes (if available)
- Old Business
- Review comments fromV2.40 public review Planning for face-to-face Topics for next call
- New Business
- Review Action Items
- Should we make the dialin information an action item on the calendar?
- wiki for minutes
- Stef sent around a function table
Motion to accept agenda
- Tim Hudson Moves
- Chris Z seconds the motion
- No objections - No Discussion
- Motion Passes
1. Approve Previous Meeting Minutes
- not available, can't approve.
2. Dialin Information
- Bob currently sends this to everyone bcc, over private email (because putting them to the list puts them in the public reflector and calls get sabotaged), Bob forwarded an email from Robin Cover about how it would work, it seems like it would be hidden from the public and web crawlers. Bob is proposing to try this the next time around, with a one-time number.
- Valerie F likes this idea, new members should hopefully be able to easily find the number if Bob didn't have their email address.
- No other comments or objections, Bob will try it for the next meeting.
3. Minutes on a Wiki
- Tim says they have been doing this with KMIP, says this has worked well. Individual members can quickly make corrections and you can add links to proposals, making everything more clear.
- It's easy to approve the minutes the next time, as it's just right there.
- Bob suggests we start this with today's minutes, keeping PDFs for the outstanding minutes from 2013.
- Bob will take Valerie's notes and put them in the wiki.
- no objections or concerns, this doesn't need a vote so we'll try it.
4. Face to Face
- About 9 people cannot attend at all, hoping most people can make it as we have so many contributors.
- As it is the last day of the conference, some people would miss keynotes/sessions on the Friday.
- Question: Will that only be a one day meeting? Bob: yes, it will be. Too many travel constraints for people. Thursday at the conference is too busy As people are trying to fly out, would we be able to go much past 1PM?
- Sven: Many international flights leave 9PM or later, 4 hours would not be enough.
- Chris: the office can handle up to 75 people, free food all day.
- Michael St. John's: I wasn't on the call when this was discussed, did we discuss other dates? other locations?
- Bob: we did discuss other dates and locations. with IETF and other events happening right before/after RSA, this seemed to be the only real option.
- Valerie: there was also follow-up discussion on the reflector.
- Chris will need Name, Company, maybe email.
- Bob will set up a non-binding poll to allow people to reply with that information for Chris. Asked Chris to send out local street address and for recommendations for lodging.
5. Documents comments
- Bob will go through the tab comments to make sure there are not any issues of substance that we need to deal with. Suspect there will be more things like that RFC case that we don't need address in the document.
- Bob asks the editors of each document to look at the spreadsheet to review comments on their document. Make suggestions for how we can address this.
- Unfortunately we did get comments that will require that we make changes, but overall this is a good thing: people are reading it and they are interested.
- One comment on the profiles, what's the best way to address this without impacting all of the other documents?
- One comment from Anders came in after the review period, it did not get included in the comment review document. If anyone feels strongly that it should be included, let Bob know (it's more applicable to a future draft).
- Tim: perhaps someone who is a member of PKCS11 TC and also the Fido alliance could give us a briefing of the Fido alliance
- Question on comment list: is there a delta sheet between 2.30 and 2.40? Bob does not have time to produce such a document, but it would be valuable.
- Volunteers? I did not hear volunteers, but more suggestions on how to do this: summary against 2.20 or highlight changes in 2.40. Because of the format changes, this is not a simple exercise.
- Mike: could list the past resolutions without all of the details.
- Bob: did a little bit of this in reply to the comment, but did not have the time to do a complete job.
- Tim: maybe a diff of the header files? Bob: that might work as well. Anyone interested in doing this? let Bob G. and Valerie know , or send a note to the reflector.
- We'll return to this in 2 weeks so we'll have a better idea of how we want to address these comments. We can dig in deeper then.
5. Extending PKCS11 API for 3.0
- Stef gave an overview today. This is a way to build this as extensible on top of 2.X, handling vendor extensions and function API flexibility.
- Multiple function tables (Interface).
- Identifiers should be collision resistant, should not require a registry, needs to be easy to use.
- One interface would be the 3.X name API.
For modules that want to provide both APIs, we might be able to provide a small shim or header file to help make that happen. 3.x would be a superset of 2.x, so there would be an ability to map from 2.x -> 3.x.
- modules would still be able to access this via a single pointer , like they do now.
- Try not to make it any more complicated than 2.x is for simple operations.
- additional thread safety requirements, which we knew would be coming.
- Valerie's only note at this time: instead of just "not more complicated", let's aim for "more simple".
- Stef said this is just scaffolding and he agrees in general to keep things simpler. Stef would like to suggest that 2.x calls can still be executed, but we do want to simplify.
- Bob R. : we do need to allow old implementations to continue to work.
6. Next Agenda
- document from Wan-Teh
- Review comments on 2.40
- How successful is having the constant document as the normative version of the header files? Bob will send a note out about this. Wan-Teh needs the .h files, we can resolve this hopefully in the next call.
- We will want to fix the manifests, for the problems found. We do want to make sure we are taking the right approach with 3.0, Stef, so keep that in mind.
Dina: C_GenerateRandom: does anyone have any thoughts or feelings on extending the interface to give parameters, like on quality? Please send Dina email and she can summarize for the list.
- Question on where are the 3.X documents? Bob clarified that we're just starting to take proposals and there are no documents available. we aren't even ready to request templates from OASIS.
Motion to Adjourn
- Tim moves to adjourn
- Stef seconds
- no objections or abstentions.
- Adjourned 1:51PM PT