Markdown Version | Session Recording
Session Date/Time: 23 Jul 2024 22:30
# sml
## Summary
This was the second SML working group session since chartering. The session covered updates on the use case, base specification, and trust and security documents. The group discussed potential approaches to representing structured data in email, including challenges with existing libraries and the need for provenance information. There were calls for co-authors for both the base specification and trust & security documents. The group also discussed the current state of the Trust and Security document and made plans for a call for adoption.
## Key Discussion Points
* **Use Cases Document:** Updated with short elaborations on potential privacy and trust issues for each use case. Seeking feedback on how to improve the discussion of these issues. Modeling guidance/best practices section proposed for the end of the use cases document.
* **Base Specification Document:** Updates included a more systematic handling of non-representation cases of structured data. Two surveys are ongoing: one to gather data on structured email usage in the wild and another to evaluate library support for custom multipart alternatives.
* **Full Representation of Structured Data:** Significant discussion on how to represent full representation of structured data, focusing on the multipart alternative approach. Concerns raised about library support for custom multi-part alternatives. Alternative proposals included using mime headers or CID.
* **Efficiency Concerns:** Discussion about parsing large HTML emails to extract structured data. Suggestion to have structured data in a separate part for efficient processing.
* **Provenance Tracking:** Discussion on the importance of tracking the source of structured data added to an email, considering cases where MTAs or other parties might add or modify structured data.
* **Trust and Security Document:** Revised to focus on describing the current state of trust considerations in structured email processing, rather than prescribing ideal behavior. Emphasis on automatic processing, and who is a trusted sender. D-KIM security validation was discussed.
* **Encryption:** Content encryption discussion about the security implications. The consensus was that current encryption is an email issue, not unique to structured mail.
## Decisions and Action Items
* **Action Item:** Hans-Yog to send the tool used for the scanning of emails around on the mailing list in the next couple of weeks.
* **Action Item:** Philip to send email with the proposal to the main list, and create a github ticket.
* **Action Item:** Ken Rickerson will find and post the IMIP draft to the mailing list.
* **Action Item:** Daniel King Gilmore will send the LAMPS draft link for email encryption to the mailing list.
* **Action Item:** Alexi and aunt to issue adoption call for the trust & security document
* **Action Item:** Update the trust & security document to include authentication results
## Next Steps
* Continue discussion on the mailing list.
* Publish the email scanning tool.
* Finalize the trust & security document and call for adoption.
* Continue the surveys on existing structured email and library support.
* Consider adding a provenance mechanism to the base specification.