docs · stream · Tangled

zat.dev / stream

jetstream v2 in zig stream.waow.tech

26 files

at commit 536e005f

zzstoatzz · docs: uv on the hosts is a provisioning gap, not a constraint · 2mo ago

architecture.md

docs: disambiguate the four worker pools, and finish the workers*2 pattern

docs: drop two benchmarks the harness never had

bootstrap-semantic-parity.md

docs: the bootstrap audit's basis commit does not exist

docs: configuration.md pointed at the wrong file for flag declarations

configuration-parity.md

docs: config audit predates comptime-derived names, and is behind the artifact

configuration.md

docs: configuration.md pointed at the wrong file for flag declarations

deployment-runbook.md

docs: runbook named the wrong deployed artifact, and never defined the gate

experiment-6-handoff.md

docs: uv on the hosts is a provisioning gap, not a constraint

full-network-experiment-2026-07.md

experiment 6: record the second deadline extension and verified costs

docs: mark the gzip divergence fixed, record the transfer-buffer trap

grafana-dashboard.md

docs: say which dashboard panels are expected empty during bootstrap

docs: fix 5 more dead paths from the regroup, and unblock invariants.md cold

jss-format-v1.md

docs: the two jss format specs pin a different upstream commit than the rest

jss-seal-spec.md

docs: verify the bloom fix cannot invalidate the existing archive

lessons-from-zlay.md

docs: lessons-from-zlay never said what zlay is, and one condition has landed

docs: repair's steady ERROR stream is expected; and worker_count*2 is now 3 bounds

live-scheduler.md

docs: the scheduler's 32 and 64 are one constant, and "cutover" is overloaded

opentelemetry.md

docs: fix 5 more dead paths from the regroup, and unblock invariants.md cold

product-copy-context.md

docs: the homepage's headline numbers reset every 45 minutes

semantic-parity.md

docs: flag that the gate's basis commit is behind the artifact that ran

docs: map the three receipt suite names that are not command names

timestamp-import.md

docs: timestamp import never explained which timestamp, or why that is safe

upstream-bootstrap-spec.md

docs: the bootstrap spec defers to a study that is not in the repo

upstream-compaction-spec.md

docs: compaction spec's source paths are upstream Go, and it never said so

upstream-harness.md

docs: harness study had no upstream pin in its header, and a machine-local path

upstream-subscribe-protocol.md

docs: subscribe protocol never explained v1 vs v2, or what "v1 quirk" means