Any news and updates of general interest to CTWG members

• Drummond Reed give an update on the Trust Spanning Protocol Task Force progress and said that it would be summarized in an upcoming ToIP blog post.

Brian Richter Rieks Joosten 

Brian Richter has made little (if any) progress due to business. The idea now is that (under the assumption we will reconvene in September) he'll show the ingress tools by then, i.e. the tools that enable Wiki-pages to be converted to Curated Texts (as specified by TEv2). Then, work can continue with the HRGT.

TNO has made progress on the TRRT, which we expect will be delivered (incl. documentation which is still progressing) by the end of this week. See an overview in screenshot #2 below. There are some few nifty feature requests that may need some additional work, but we're quite happy with what it should be able to do - it will be fully customizable, both in terms of TermRef syntax as well as in terms of the texts that they will be converted into.

There have been a few changes in specs, (see screenshot #1 below) the most notable one being the addition of a new tool - the MRG importer.d

Rieks Joosten has made his first Python script, which is a MRGT that only works for all curated texts in a single scopedir. It was used to make the MRG for the eSSIF-Lab stuff.

On 2023-07-11 in the ToIP Slack, Henk posted this update:

In the KERI meeting today we voted for the move of our “source of terms truth” known as the ACDC-wiki: it’s going to be WOT-terms wiki: https://github.com/WebOfTrust/WOT-terms/wiki from today. We will still be able to sync between the glossaries at
ToIP user acdc repo and WebofTrust WOT-terms repo.

The reason for the move is a practical one: more control over our own data and functionality while maintaining the symbiotic properties between ToIP and WoT. More info and acknowledgements in KERISSE section how we did.

The most recent update of the glossary in ToIP format can now be found here but will most probably also be hosted from ToIP github space (here) soon.

The key takeaway is that the authoritative source for the KERI/CESR/ACDC terms is now the WOT glossary.

Henk also showed us the KERISSE search tool. An example is at this link.

Rieks asked whether there was a manual. Henk pointed to this Docusaurus documentation.

Drummond Reed Rieks Joosten 

Discuss the best way to develop mental models for the terms in the ToIP Glossary Workspace.

Drummond said that he believes there are several clusters of terms in the ToIP Glossary Workspace. So he wondered about how to best go about developing those clusters into mental models.

Rieks advised to focus on the questions that readers will have about the relevant terms and their usage. Sometimes it develops into a picture; other times it is just a list of terms and their relationships. But Rieks notes that often it is not clear what the overall focus and purpose it.

Rieks gave an example of the Parties-Actors-Actions model: it can answers a series of questions about who is doing what action on behalf of what party—that's the key set of questions answered by the model.

Another example is the Jurisdictions model: it answers questions about who is responsible for what within a digital interaction or ecosystem. What happens when something goes wrong? Who helps? What does this help consist of?

The same applies to the Governance pattern and the Management pattern.

So the focus should be on identifying the key questions that need to be answered. These questions should be relatively easy, e.g., "How do you know that the data is valid?"

Neil: "Use Case driven. Likely going to need a collection of Use Cases, each illustrating key concepts and relationships, which when pulled together becomes a single model which spans all the use cases."

Neil said that, particularly with the APAC call discussions, it "all comes down to a basket of use cases". That's what provides the context. It makes it much easier to then understand the composite, rather than the other way around.

For example, Neil said that there are two parts to an issuer: the decision-making process about whether to issue (the core governance/human-managed process), and then the mechanical process of actual issuance.

Neil recommends not trying to do the composite picture first.

Rieks completely agreed. "Stay away from the composite picture as long as possible."

Rieks also advised to look for common contexts and patterns. For example, "issuing" and "authoring" are similar patterns answering similar questions: "What is responsible for issuing this data?" So the general pattern is that, if a verifier can determine who is the author of the data, that's what the verifier ultimately needs to know to make a decision. That's how to keep it as simple and general as possible.

Neil pointed out that this can require a delegated trust model in order to trace the provenance.

Rieks pointed out that the current definitions in the W3C VC specifications say that the role of issuer is an "entity", which means it can included a machines such as an IT sensor. That raises the question of who (what party) is actually accountable for the issuance of that sensor? You must resort to other entities. 

ACTION: Drummond Reed will look for sets of questions about clusters of terms in the ToIP Glossary Workspace that may be candidates for mental models.

We agreed that due to summer vacations, we will have no meetings in August. Our next meeting will be 11 Sept 2023.

Neil Thomson said the GSWG is also taking the same approach.

ACTION: Drummond Reed to request Michelle Janata cancel the two August 2023 meetings of the CTWG.