Repository navigation
🐛 tracker: say it once, in plain words — simplify hovers, menus and captions - #15406
Conversation
…aptions
Operator feedback on the activity tracker: "this is WAY too much
information … use easy to understand language and much much less text."
The 090 chip hover on hive#15315 ran to ~380 words: a seven-line question
with config keys, six buttons, a bulleted legend and the same
untrusted-author paragraph repeated three times.
Server (say it once, at the source):
- stageTrackerUntrustedAuthorWhy: one short plain sentence per trust
reason, no config keys; the how-to-trust line moves to the new
stageTrackerUntrustedAuthorFix and only ever shows inside Details.
- HumanMergeReasonText, the 090 caption and the 090 next-move line no
longer echo each other; gate-word tips and every decision consequence
are one plain line (<= 12 words).
- StageTrackerDecision gains Detail; decisionForHumanMerge builds a
14-word question and folds DCO / hold / admission into Detail through
stageTrackerSayOnce, which drops any sentence already said.
SPA:
- stageTrackerDecisionPopHtml: one sentence of state, one primary action
("Merge it"), "More ⋯" opening the row's ⋯ menu, who asked and when,
and one collapsed <details> for everything technical; no button
cluster, no legend, no stage-line echo (stageTrackerSayOnce).
- 090 chevron tip stops appending the stage description; the row
tooltip says "person / agent" instead of author/<kind> and drops
duplicate lines; the merge confirm dialogs lose "trust override",
"sweep" and "soak".
Tests: TestStageTrackerHumanMergeSaysItOnce (no repeated sentence, no
next-move echo, no config keys, no jargon across all trust reasons);
popover node test asserts <= 40 visible words, one primary button plus
More, no duplicate sentence, banned-jargon list and no a.b.c keys.
Docs: glossary in dashboard-sections.md, dashboard-copy-inventory.md with
before/after word counts.
Fixes #15394
Signed-off-by: Andrew Anderson <andy@clubanderson.com>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
Docs map: this PR adds page(s) under
A page missing from the map is one readers do not find. Please add a If the page is indexed from a subdirectory README, or is deliberately |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
|
@clubanderson is auto-promoted for contributor-lane merging on this PR: the last 5 merged PRs by this author in this hive's repositories have no revert or sentinel-alert evidence. CI, reviewer verdict, diff fences, soak, holds and branch protection still apply. |
|
stage 080-awaiting-merge-admission → 050-review-pending: required-check-pending:build-and-test |
Signed-off-by: Andrew Anderson <andy@clubanderson.com>
|
stage 080-awaiting-merge-admission → 050-review-pending: required-check-pending:golangci-lint |
There was a problem hiding this comment.
Reviewed (intent-alignment vs #15394) — the diff does what the issue's inventory asks: stageTrackerUntrustedAuthorWhy/…Fix split (stage_tracker.go:1490–1536), 090 caption and next-move no longer echo (stage_tracker.go:1730–1736, stage_tracker_actor.go:200–202), Detail + stageTrackerSayOnce (stage_tracker_decision.go:50–53, 263–303), popover = one sentence · one primary · More ⋯ · Details (tracker-actions.js:719–740), with tests covering each. Two small advisory gaps, neither blocking:
-
Policy-fallback slug still lands in the first read. For a 090 reason outside
stageTrackerHumanMergeReasonWords,stageTrackerHumanMergeReasonreturns the tip with the raw lane reason in backticks (stage_tracker.go:1469), anddecisionForHumanMergeputs that tip straight intod.Question(stage_tracker_decision.go:275–280). Issue goal 1 says internal nouns go behind Details.TestStageTrackerHumanMergeSaysItOnceruns"something-brand-new"but only banslane/stage numbers, so a realcontributor-lane:…slug passes. Suggested fix indecisionForHumanMerge:- _, _, tip := stageTrackerHumanMergeReason(it.Reason) + code, _, tip := stageTrackerHumanMergeReason(it.Reason) ... switch { case why != "": why = "Hive won't merge it alone — " + why + "." + case code == stageengineprstage.HumanMergeReasonPolicy: + why = "Hive's merge rules leave this one to a person." case tip != "": why = sentence(tip) ... details := []string{"Hive never merges this on its own; a maintainer merges here or on GitHub."} + if code == stageengineprstage.HumanMergeReasonPolicy && base != "" { + details = append(details, "Rule: "+base+".") + }
-
Fixes #15394closes the issue while part of its scope is deferred. The body says the 080/095 live-clock tips still say "sweep"/"soak", and the 095 caption/actor still read "Admitted by the automerge lane; the pre-merge soak window is running" / "automerge lane" (stage_tracker.go:1738; "lane" and "soak" are on the issue's banned list). Either change the keyword toPart of #15394, or open the follow-up issue before merge so the remaining surfaces aren't orphaned.
— hive: agent=ci-maintainer backend=copilot model=claude-fable-5.1 copilot=1.0.88
There was a problem hiding this comment.
Correctness review (head ed2ecc7)
1. The hover's single "recommended" button can be a red Close as superseded — static-src/js/tracker-actions.js:728
primary = options.find(o => o.enabled) || options[0] promotes the first enabled option. For a held 090 PR by a person, the server list (stage_tracker_decision.go:291-321) is merge-now ✗ (hold), queue ✗ (hold), delegate ✗ (decisionDelegateOption(held…), line 217), direction ✗ (human author), close-superseded ✓ (danger)… — so the one primary action a maintainer sees is the destructive close. Same for a held waiting-on-author row (stage_tracker_decision.go:514). The old cluster didn't have this problem because it showed every button. stage_tracker_decision_popover_test.go:111 only covers the held case with options.slice(0, 1), so the full list is never exercised.
const primary = options.length ? (options.find(o => o.enabled && !o.danger) || options[0]) : null;
(and add a held case with the full option list to the harness asserting the lead is not a decide:close-*.)
2. Next-move line contradicts the reason for admitted heads — stage_tracker_actor.go:202
090 now always says human — a maintainer merges it, while stage_tracker.go:2085 sets the reason text to @login approved it; hive merges it on its next pass for the same row. Before, the actor why carried that text, so the two agreed.
if stagelabels.AdmissionLogin(it.AdmittedBy) != "" {
return set(StageActorHive, "", "hive merges it on its next pass")
}
return set(StageActorHuman, it.HumanApprover, "a maintainer merges it")
Minor: stageTrackerSentences (stage_tracker_decision.go:355) splits on every .; a hold reason like "see 1.2" in decisionHeldBlocker comes back as "see 1. 2". Low-impact; noting only.
Could not run go test/the Node harness here (blocked on agent hosts); Go/JS logic read by hand.
— hive: agent=adjudicator backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
@clubanderson this PR is held at Why:
Next: the hive does not push to your branch. Push a new head (or reply here); CI and the review swarm re-run automatically and the PR returns to the normal path. |
|
merged by Hive trusted-author auto-merge: author @clubanderson holds role owner and GitHub permission admin on repo; CI green at ed2ecc7. |
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
…et hive decide Two open PRs declaring the same Fixes #N (hive's #15402 and #15407 for per closing issue (green head first, then earlier-opened) and holds the other with stage reason merge-train:duplicate-of#<winner>; the tracker shows merge_gate.state duplicate-of:<n> ("Duplicate of #15402 · needs a pick", needs-decision) and Let hive decide takes the existing supersede path (rule one-pr-per-issue). Also maps every merge_train state slug to plain words (#15406 read bare "front") with a never-leak test. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…et hive decide Two open PRs declaring the same Fixes #N (hive's #15402 and #15407 for per closing issue (green head first, then earlier-opened) and holds the other with stage reason merge-train:duplicate-of#<winner>; the tracker shows merge_gate.state duplicate-of:<n> ("Duplicate of #15402 · needs a pick", needs-decision) and Let hive decide takes the existing supersede path (rule one-pr-per-issue). Also maps every merge_train state slug to plain words (#15406 read bare "front") with a never-leak test. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…et hive decide Two open PRs declaring the same Fixes #N (hive's #15402 and #15407 for per closing issue (green head first, then earlier-opened) and holds the other with stage reason merge-train:duplicate-of#<winner>; the tracker shows merge_gate.state duplicate-of:<n> ("Duplicate of #15402 · needs a pick", needs-decision) and Let hive decide takes the existing supersede path (rule one-pr-per-issue). Also maps every merge_train state slug to plain words (#15406 read bare "front") with a never-leak test. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
…et hive decide Two open PRs declaring the same Fixes #N (hive's #15402 and #15407 for per closing issue (green head first, then earlier-opened) and holds the other with stage reason merge-train:duplicate-of#<winner>; the tracker shows merge_gate.state duplicate-of:<n> ("Duplicate of #15402 · needs a pick", needs-decision) and Let hive decide takes the existing supersede path (rule one-pr-per-issue). Also maps every merge_train state slug to plain words (#15406 read bare "front") with a never-leak test. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
… warning (#15427) * ✨ tracker: merge queue position on the chevron, stable wording, stall warning Position chip "In merge queue · 2 of 6 · ~4m" (front: "Next to merge · 1 of 4"), label stable across sweeps (sweep = spinning ⬡ + "Checking now…"), Merge queue popover in the sticky bar, countdown to the member's turn anchored server-side (merge_train.eta_s/eta_at/eta_basis/late), stall warning with governor.merge_train_stall_warn_s (default 3600) and an hourly activity line, and a words table so no merge_train state slug (front, queued-behind, …) can reach a chip again. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com> * ✨ merge queue: one PR per issue — hold the duplicate, supersede via Let hive decide Two open PRs declaring the same Fixes #N (hive's #15402 and #15407 for per closing issue (green head first, then earlier-opened) and holds the other with stage reason merge-train:duplicate-of#<winner>; the tracker shows merge_gate.state duplicate-of:<n> ("Duplicate of #15402 · needs a pick", needs-decision) and Let hive decide takes the existing supersede path (rule one-pr-per-issue). Also maps every merge_train state slug to plain words (#15406 read bare "front") with a never-leak test. Fixes #15395 Signed-off-by: Andrew Anderson <andy@clubanderson.com> --------- Signed-off-by: Andrew Anderson <andy@clubanderson.com>
Fixes #15394
Operator feedback on the activity tracker (verbatim): "this is WAY too much information. I dont even know where to start. You need to look at all information displayed in activity tracker and its associated hovers and dialogs and simplify them. use easy to understand language and much much less text."
The 090 chip hover on #15315 ("Ready for human merge · untrusted author") ran to 421 words: a seven-line question with config keys, six buttons, a bulleted legend, and the same untrusted-author paragraph three times. It is now ≤ 40 visible words, one primary action, and a collapsed Details.
Before / after — the #15315 hover
New hover text, verbatim:
What changed
Server — say it once, at the source (
stage_tracker.go,stage_tracker_decision.go,stage_tracker_actor.go,delegated_decision_human_merge.go)stageTrackerUntrustedAuthorWhy: one short plain sentence per trust reason, no config keys (314 → 68 words across the ten reasons). The how-to-trust line moved to a newstageTrackerUntrustedAuthorFixand only ever shows inside Details.HumanMergeReasonText, the 090 caption ("Ready to merge — a maintainer merges it") and the 090 next-move line ("a maintainer merges it") no longer echo each other. Gate-word tips are one plain line each (227 → 101 words).StageTrackerDecisiongainsdetail.decisionForHumanMergebuilds a 14-word question, folds DCO / hold / admission intodetailthroughstageTrackerSayOnce(drops any sentence already said), and every option consequence is ≤ 12 words — those consequences are the ⋯ menu item hints.SPA (
static-src/js/tracker-actions.js,tracker-row.js,css/tracker.css;static/index.htmlregenerated)stageTrackerDecisionPopHtml: head · one sentence · one primary button (Merge it/Let hive merge it/Send back, else the lead option) ·More ⋯(opens the row's ⋯ menu anchored on the chip) · who asked and when · one<details class="ham-decision-details">holding blocker, detail and the stage line, deduplicated with the JSstageTrackerSayOnce. No button cluster, no legend, no stage-line echo. Existing tokens only.(person)/(agent)instead of(author/human)/(author/clanker)and drops duplicate lines; merge confirm dialogs lose "trust override", "sweep", "soak".Tests
TestStageTrackerHumanMergeSaysItOnce— across all trust and policy reasons, no sentence repeats between question / blocker / detail / next-move, the next-move never echoes the reason text, noa.b.cconfig keys and no jargon (sweep, soak, sentinel, audit marker, trust override, lane, requires_human, stage numbers).TestStageTrackerDecisionForHumanMergeasserts the exact question, Details content, empty first-read blocker and ≤ 12-word consequences.TestStageTrackerDecisionPopoverContent— ≤ 40 visible words outside Details, exactly onedata-ham-actbutton plusMore ⋯, the greyed-lead case, a say-it-once case where the stage tip repeats the question, banned-jargon list and the config-key regex.Docs
src/docs/dashboard-copy-inventory.md(new): every tracker-visible string builder with before/after word counts. Totals: server builders ≈ 1,350 → ≈ 620 words (≈ 110 of those only inside Details); 🐛 fix: stop filing the pre-restart kick-undeliverable warning as a fleet issue #15315 hover 421 → 29 visible.src/docs/dashboard-sections.md: operator glossary (Agent / Contributor / Author / Reviewer / Merge; "clanker" only as the relay's product noun) and the hover budget.changelog.d/fixed-tracker-say-it-once.md.Left for a follow-up under the same issue: the 080/095 merge-gate live-clock tips still say "sweep" / "soak".