Inside a Request Pipeline
Follow a synthetic request from ingress to a response.
Inside a Request Pipeline
This is the first article in the M7 After synthetic showcase. The system described here is fictional: it is a reading aid, not a claim about a production topology.
Follow one request
A visitor asks for an article. The ingress passes the request to the public frontend, which reads publication-safe data from the API. Each step has a different responsibility.
page 1 of 1Large diagrams are paged; the complete source below includes all connections, including cross-page edges.A fictional read request, from visitor to publication-safe response.flowchart TD
A[Visitor] --> B[Ingress]
B[Ingress] --> C[Public frontend]
C[Public frontend] --> D[Publishing API]
D[Publishing API] --> E[Published response]Separate responsibilities
- Receive the request at the ingress.
- Read the current public representation.
- Keep owner-only state out of the response.
- Preserve the article's stable identity.
- Render a useful page before optional interactive features load.
The public route is /blog/{slug}. An article title may change; its identity should not depend on the title's current spelling.
request -> public read -> safe document -> HTML response
A small reading exercise
Try reading this page without activating anything: the explanation and diagram should already be useful. Resize the browser and inspect the nested list. Use the heading links to return to a section.
For the same ideas in a visual release story, continue to A Deployment in Pictures. For formatting examples, visit Readable Technical Notes.
What this example does not prove
It does not measure throughput, define a deployment operation, or authorize a network change. It simply makes the distinction between presentation, public data, and request routing visible.