WG Action: Formed vCon (vcon)

[The IESG <[e-mail]>]

A new IETF WG has been formed in the Applications and Real-Time Area. For
additional information, please contact the Area Directors or the WG Chairs.

vCon (vcon)
-----------------------------------------------------------------------
Current status: BOF WG

Chairs:
  Brian Rosen <[e-mail]>
  Chris Wendt <[e-mail]>

Assigned Area Director:
  Murray Kucherawy <[e-mail]>

Applications and Real-Time Area Directors:
  Murray Kucherawy <[e-mail]>
  Francesca Palombini <[e-mail]>

Mailing list:
  Address: [e-mail]
  To subscribe: https://www.ietf.org/mailman/listinfo/vcon
  Archive: https://mailarchive.ietf.org/arch/browse/vcon/

Group page: https://datatracker.ietf.org/group/vcon/

Charter: https://datatracker.ietf.org/doc/charter-ietf-vcon/

[https://github.com/dgpetrie/draft-petrie-vcon/blob/main/vcon_charter.md]

Introduction and Group Overview

The vCon ("Virtualized Conversations") working group is about passing
conversational data. Such data is commonly generated and collected in
business environments, from chat logs to transcripts to recordings. Most
systems provide a way to store such information, but there are not many
standards or interoperability within the storage or transmission mechanisms
or formats.

The two opposing forces influencing such information passing are trying to
enforce privacy of personal data and providing the ability and interest to
use conversations in various ways, e.g., AI analysis. For the first, the key
is knowing what information exists, where it comes from, and being able to
protect it appropriately; or being able to refer to conversations without
exposing their contents or assure suitable redaction has been performed. For
the second, the key is being able to integrate data between multiple systems
(phones, chat systems, email, etc.), move data when transitioning from one
software or provider to another, and so on. There is a lot of conversational
data already being exchanged, but using proprietary formats and per-case
engineered exchange solutions, from using FTP and other legacy components, to
file naming conventions, and so on. There are also open source systems
implementing vCon which may yield input to the requirements.

The use of a standard for a conversation data container will focus on the
following data exchange scenarios:

* Communication System to Data Owner/Consumer

* Data Owner/Consumer to Analysis or Storage Services

* Analysis Services to Data Consumer

The Communication System is an application, service or system which is able
to capture the conversation metadata and the conversation. The Analysis
Services will add data to an existing conversation data container. It should
be noted that these entities are not always distinct. For example, the
Communications System may also provide some analysis data. It should also be
noted that these entities may also exist in multitude. For example, an
enterprise may have a communication system for each mode (e.g., text,
message, voice, video) or for each corporate product or division.

In Scope

The scope of the VCON working group is:

(a) Standards Track RFC Output

* Define a JSON based standard container and Media type to contain and/or
reference conversational data, including call style metadata, recordings,
data exchanged or presented in the conversations, conversation analysis,
transcriptions, translations and annotations

* Define/specify the use of an existing mechanism for proving integrity and
optionally authenticity of the conversation data

* Define/specify the use of an existing mechanism for encrypting of the
objects enclosed in the vCon conversation data container to provide privacy
of the participants and/or confidentiality of the data independent of
transport such that some parts of the vCon may be disclosed to different
parties

* Determine if there is need for defining media types and standard containers
for some small set of analysis, annotation or transcription data

(b) Informational Internet Draft Output

The Working Group may develop use cases in drafts for reference, but there is
no expectation they will be published as an RFC. The use cases should include
considerations for data minimization. The Working Group may consider the
following as well as other use cases:

* Contact Center Data Exchange

* Non-Contact Center Data Exchange with Customer Relationship Management (CRM)

* Conversations of Record including ECRIT Environments

* Message History Data Exchange

Out of Scope

The following are out of scope:

* Algorithms or methodologies for transcription, translation, redaction,
analysis or annotation of call data

* Real-time streaming or updating of conversational data

* Transport mechanisms

* Storage or databases specifications

* Key management

* API definitions

* Definition of a new object security model.  (It is expected that JOSE or
other existing IETF technology is sufficient.)

Privacy

There are privacy matters relevant to transcripts of the sort VCON is meant
to standardize. VCON itself does not attempt to address these concerns
directly, as that is intended to be handled by metadata in the layers above
VCON itself.  However, the documents produced by this working group will
discuss such considerations, and will not preclude support for conveying
important notions such as user consent.

Milestones

* Specification for a JSON based container for conversation data

* Recommendations or analysis of missing (not defined) data containers and
media types for components of the conversation data

Milestones:

  Dec 2023 - Adopt a draft to provide a specification for a JSON based
  container for conversation data (Proposed Standard)

  Dec 2023 - Adopt a draft containing recommendations or analysis of
  undefined  data containers and media types for components of conversation
  data (Informational)

  Jun 2024 - Submit draft to provide a specification for a JSON based
  container for conversation data to the IESG (Proposed Standard)

  Jun 2024 - Submit draft containing recommendations or analysis of undefined
   data containers and media types for components of conversation data to the
  IESG (Informational)