# The checks, measured: what changed in v1.2.0, sgit.ai > The changes to the article "The checks, measured" in v1.2.0, paragraph by paragraph. *Source: · sgit.ai v0.7.87 · this file is generated from the same content as the page, so the two cannot drift. Every page on this site has a `.md` twin; internal links below point at them.* --- [Home](../../../index.md) / [Articles](../../index.md) / [The checks, measured](../../the-checks-measured.md) / [Versions](../the-checks-measured.md) / v1.2.0 # The checks, measured: what changed in v1.2.0 From v1.1.0 (2026-10-21, `e835280b3`) to v1.2.0 (2026-10-27, `db67c5ce1`), paragraph by paragraph. 0 paragraphs added, 0 removed, 6 changed in place, 64 unchanged. About 0 words added and 210 removed. Insertions are marked like this, deletions like this; unchanged runs are folded to one line; figures appear as their file names. [← v1.1.0](v1.1.0.md) · [all versions](../the-checks-measured.md) 2 unchanged paragraphs !shot qm-where.webp | images/ | Where each check runs. The site validator and the 18 test files runruns: in the repository and in CI. The rebuild and sgit.ai's style rules run onlyCI, on my machine. The summaries, page-width, graph, version and live checks livemachine, in this session's scratchpad;scratchpad,and the leak scan, with the wording check it runs, and the behaviour policyor in this container. | comparison 12 unchanged paragraphs, under In short {#in-short}, The question {#why}, Where a release's time goes {#time} !shot qm-release.webp | images/ | Where a release's time goes: about 8 minutes on this machine, borrowing sgit.ai's generator 246 s, this site's 30 builders 38 s, validation 9 s and the 18 test files 184 s, about 8 minutes; thenstages in CI since v0.13.7, validation 18 s alongside the test job, 217 s, then the tag 33 s and the deploy 49 s.v0.13.7. | comparison 8 unchanged paragraphs, under 123 local builds in seven days {#local} !shot qm-local.webp | images/ | 123 local builds between 15 and 21 October 2026: builds a day, 11, 9, 2, 28, 21, 25 and 27, and the median minutes fromtostartingread a build to reading its result, between 9.8 and 23.9.result. | comparison 2 unchanged paragraphs, under CI got slower as the tests arrived {#ci} !shot qm-ci.webp | images/ | Left: theThe median minutes of a push run on dev,dev by week, from 0.6 in late August to 7.6 in the week of 19 October. Right:and the median seconds of the test step,step by day, from 102 on 11 October to 348 on 20 October, and 180 on 21 October when the tests began to run four at a time.day. | comparison 4 unchanged paragraphs, under Eighteen test files {#tests} !shot qm-tests.webp | images/ | The 18 test files: their checks in the last release, from 203 for the wire to 14 for the review; the seconds each took;taken, what each drives, mostly a browser, two of them a live service; and how many CI runs each failed, 14 each for the wire and the vault viewer.failed. | comparison 7 unchanged paragraphs, under Why the CI tests failed {#failures} !shot qm-fails.webp | images/ | The first failing check in each of 44 failed CI test runs:runs,thegroupedvaultbyviewer from SG/Send 14, the wire map's 55 entities 12, front.py idempotent 7, a review finding 3, the wire's release notes 2, reading measured in time 2, and one each for the top bar, search, an sgit clone and a run with no check printed.check. | comparison 29 unchanged paragraphs, under Where each check lives {#where}, What the data says to change {#next}, What this does not claim {#limits}… [← v1.1.0](v1.1.0.md) · [all versions](../the-checks-measured.md) --- *[Site index for agents](../../../llms.txt) · [HTML version](https://newsroom.sgit.ai/articles/versions/the-checks-measured/v1.2.0.html)*