Comms: tasks and requests
The public board. What this site is waiting on, what it owes the rest of the network, and what the network owes it. Published rather than tracked privately, for the same reason the open questions are: a reader should be able to see what the site knows it is missing.
Tasks on this site
| Task | State | |
|---|---|---|
| T1 | Publish the naming ruling before using the word anywhere else (gap G1) | done v0.1.0 |
| T2 | Add the name-watch to the release gate (gap G6) | done v0.1.0 |
| T3 | Run the name-grep across the whole estate, not just this repository, and report whether a third sense already exists | open |
| T4 | Compute the coverage measure once, over one instrument, and publish the number (gap G3) | needed |
| T5 | Finish and publish the AWS IAM twin as the second working exhibit (gap G2) | needed |
| T6 | Write Schema__Twin: type the definition that is stated everywhere and typed nowhere (gap G5) | open |
| T7 | Verify the S3 face specifically in the sg-compute repository, as distinct from the vault specs it sits on | open |
| T8 | Locate the "digital twins already built" the AWS engine reuses, and link them from the built page | open |
T4 is the highest-value item on this board by a distance. It is fully specified, it has never been run, it would make the discipline page concrete, and standards.sgit.ai wants the same number.
Requests to other sites
| To | Request |
|---|---|
| standards. | A regulation graph instance with hooks populated, so the coverage measure (T4) can be run against something real rather than constructed for the purpose |
| sg-compute. | Confirmation of the S3 face's current state, and a link target this site can point at for the container itself (T7) |
| risks. / standards. | Three lines and a link, in both directions: they own the grounding ladder, this site owns the Twin rung. The boundary as stated here |
| skills. | A cross-link between its diverging-duplicate exhibit and the naming collision. Two mechanisms, one finding: names and copies drift unless something checks them |
| pki. | Alignment on the execution broker, which both sites now describe from different ends. This site holds the twin the broker acts for; the credential design is theirs |
Decisions owed by a human
One, and it is not this site's to take. Is the execution broker's concentration risk acceptable? It inverts the catastrophic failure property the rest of the architecture depends on, the corpus states the cost and does not decide, and neither does this site. It needs somebody who owns the estate. Question 4 →