[secdir] [new-work] REVISED WG Review: Web PKI OPS (wpkops)

[IESG Secretary <[e-mail]>]

A new IETF working group has been proposed in the Operations and
Management Area. The IESG has not made any determination yet. The
following draft charter was submitted, and is provided for informational
purposes only. Please send your comments to the IESG mailing list (iesg
at ietf.org) by 2013-02-20.

Web PKI OPS (wpkops)
------------------------------------------------
Current Status: Proposed Working Group

Chairs:
  Tim Moses <[e-mail]>

Assigned Area Director:
  Ronald Bonica <[e-mail]>

Mailing list
  Address: [e-mail]
  To Subscribe: https://www.ietf.org/mailman/listinfo/wpkops
  Archive: http://www.ietf.org/mail-archive/web/wpkops/

Charter of Working Group:

The Web Public Key Infrastructure (PKI) is the set of systems,
policies, and procedures used to protect the confidentiality,
integrity, and authenticity of communications between Web
browsers and Web content servers.  The Web PKI is used in
conjunction with security protocols such as TLS/SSL and OCSP.

More specifically, the Web PKI (as considered here) consists of
the fields included in the certificates issued to Web content
and application providers by Certification Authorities (CAs),
the certificate status services provided by the Authorities to
Web browsers and their users, and the TLS/SSL protocol stacks
embedded in web servers and browsers.

The Web PKI Operations (wpkops) working group will work to
improve the consistency of Web security behavior.  It will
address the problems caused by the many hundreds of variations
of the Web PKI currently in use:

- For end-users (i.e. relying parties), there is no clear view
  of whether certificate "problems" remain once they see an
  indication of a "good" connection.  For instance, in some
  browsers, a "good" indication is displayed when a "revoked"
  response has been received and "accepted" by the user,
  whereas other browsers refuse to display the contents under
  these circumstances.

- Many certificate holders are unsure which browser versions
  will reject their certificate if certain certificate profiles
  are not met, such as a subject public key that does not
  satisfy a minimum key size, or a certificate policies
  extension that does not contain a particular standard policy
  identifier.

- Certificate issuers (i.e., CAs) find it difficult to predict
  whether a certificate chain with certain characteristics will
  be accepted.  For instance, some browsers include a nonce in
  their OCSP requests and expect one in the corresponding
  responses, not all servers include a nonce in their replies,
  and this means some certificate chains will validate while
  others won't.

The working group's goal is to describe how the Web PKI
"actually" works in the set of browsers and servers that are in
common use today.  To that end, the working group will document
current and historic browser and server behavior.  For each
this will include:

- The trust model on which it is based;
- The contents and processing of fields and extensions;
- The processing of the various revocation schemes;
- How the TLS stack deals with PKI, including varying
  interpretations and implementation errors, as well as state
  changes visible to the user.
- The state changes that are visible to and/or controlled by
  the user (to help predict the decisions that will be made the
  users and so determine the effectiveness of the Web PKI).
- Identification of when Web PKI mechanisms are reused by other
  applications and implications of that reuse.

Where appropriate, specific products and specific versions of
those products will be identified, but recording the design
details of the user interfaces of specific products is not
necessary.

Only server-authentication behavior encountered in more than 0.1
percent of connections made by desktop and mobile browsers is to
be considered.  While it is not intended to apply the threshold
with any precision, it will be used to justify the inclusion or
exclusion of a technique.

A number of activities are outside the immedaiate scope of this
working group, but might be considered in future re-chartering
activity or included in the work of other working groups:

- The working group will not work to describe how thw Web PKI
  "should work.
- The working group will not examine the certification
  practices of certificate issuers.
- The working group will not investigate applications (such as
  client authentication, document signing, code signing, and
  email) that often use the same trust anchors and certificate
  processing mechanisms as those used for Web server
  authentication.

Given the urgency of the required developments and the scale of
the task, it is agreed that adherence to the published
milestones will take precedence over completeness of the
results, without sacrificing technical correctness.

Milestones
==========
1. First WG draft of "trust model" document (4 months).
2. First WG draft of "field and extension processing for
   certificates, CRLs, and OCSP responses" document (12 months).
3. First WG draft of "certificate revocation" document (8 months).
4. First WG draft of "TLS stack operation" document (8 months).
5. IESG submission of "trust model" document (16 months).
6. IESG submission of "field and extension processing for
   certificates, CRLs, and OCSP responses" document (24 months).
7. IESG submission of "certificate revocation" document (20
   months).
8. IESG submission of "TLS stack operation" document (16 months).
_______________________________________________
new-work mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/new-work