Back to matrix

E04-NS: Yes

Rationale

Yasuhiro Matsumoto publishes elapsed seconds and executable scripts for five sequential publish, ID-query and deletion-request cycles against named public WSS relays. Credit is restricted to external strfry relays, not his own relay entries. strfry is separately maintained by hoytech. The shared mechanism is Nostr event publication and subscription through WebSocket relays. The disclosed client workload, endpoints, implementation family and timing boundary meet revised E04 as a small external quantitative observation.

Assessment method

Revised E04 documentary assessment of bounded client-workload timings

Source context

Matsumoto external relay timing report and scripts; strfry upstream

Outstanding qualification

The scripts time complete nak command sequences, including client and network overhead. They do not robustly assert every returned event, acceptance or deletion; measurements are workload elapsed times, not verified successful-event throughput. Deletion requests are additional workload, not a selected-profile capability. Hardware, relay versions and TLS versions are unspecified; remote configurations vary. No ranking, isolated server latency, TLS 1.3 cost, selected nostr-rs-relay memory configuration or delivery guarantee is inferred. Author-operated relay results are excluded; no exhaustive contributor audit is claimed. Scripts were inspected but not executed.

Public sources

• https://github.com/mattn/nostr-relay-benchmark
• https://github.com/hoytech/strfry

Editorial evidence references

The following identifiers refer to research records not included in this public package; they are not public links.

• round626.html
• round626-sources.json