From 0bfa7c17c36b35bb8ce2a5c9d5f620f586fae1bb Mon Sep 17 00:00:00 2001 From: Steven Roberts Date: Thu, 6 Aug 2026 15:04:22 -0700 Subject: [PATCH 1/2] docs: link Histology Databank Explorer from data management plan The Histology Data Management Plan described filling out the databank sheet and storing images on owl, but never pointed at the Explorer that joins the two. Add that pointer after the owl storage convention, noting the "Unmatched images" backfill view and that the Explorer is a rebuilt snapshot rather than a live read of the sheet. Co-Authored-By: Claude Opus 5 --- docs/Data-Management.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/Data-Management.md b/docs/Data-Management.md index 8aac1ac0..cb3a3c28 100644 --- a/docs/Data-Management.md +++ b/docs/Data-Management.md @@ -319,6 +319,8 @@ _Taxa Representation in Nightingales_ All images should be stored in the proper species directory at http://owl.fish.washington.edu/hesperornis/ +To browse what has already been catalogued, use the [Histology Databank Explorer](histology-explorer/index.html) — it joins the databank sheet to the images on owl so you can search samples by species, project, year, tissue, and researcher, and view the slides. Its "Unmatched images" view lists specimens imaged on owl that have no databank row yet. The Explorer is a snapshot rebuilt from a CSV export of the sheet, so new entries appear only after the next [rebuild](https://github.com/RobertsLab/resources/blob/master/docs/histology-explorer/README.md). + --- From 57f0fe4c0248c6727ea0e779112fae7d98d0cd67 Mon Sep 17 00:00:00 2001 From: Steven Roberts Date: Thu, 6 Aug 2026 15:04:22 -0700 Subject: [PATCH 2/2] docs: correct stale coverage numbers in explorer README The README claimed 56 of the 100 imaged samples had a browser-viewable image, but data/qc_report.json from the last build reports 100. The gap was the JPEG derivative run: build/derivatives_manifest.csv has 352 rows, and every imaged sample in data/samples.json now resolves to a viewable image, 44 of them via generated derivatives. Update the coverage figures and add the matched-image count so the 918/651 split accounts for all indexed files. Also reword the .tif note, which was stale in the same way -- it implied archival sets like O_angasi still render as "TIF only" placeholders, when that state now only applies to a .tif with no derivative. Co-Authored-By: Claude Opus 5 --- docs/histology-explorer/README.md | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/docs/histology-explorer/README.md b/docs/histology-explorer/README.md index 399fa190..9d0ed033 100644 --- a/docs/histology-explorer/README.md +++ b/docs/histology-explorer/README.md @@ -43,8 +43,10 @@ histology-explorer/ Many images (all of `O_angasi`, the 2026 geoduck gonad set) are archival **`.tif`** files, which browsers cannot display. The explorer shows browser-viewable `.jpg`/`.png` inline and -offers `.tif` as a download link. A sample with only `.tif` images shows a "TIF only" card. -To give those real thumbnails, generate `.jpg` derivatives (see next section). +offers the `.tif` as a download link. For `.tif` files with a generated `.jpg` derivative, the +derivative is displayed and the original stays downloadable — 44 of the 100 imaged samples +currently display this way. A `.tif` with no derivative shows a "TIF only" card; to give those +real thumbnails, generate derivatives (see next section). ## Generating JPEG derivatives for `.tif` images @@ -76,9 +78,10 @@ updates `build/derivatives_manifest.csv`. Then: ## Current coverage (from the last build) - 1,308 sheet rows → **940 unique samples** -- **100 samples** have at least one image; **56** have a browser-viewable image -- **918 image files** indexed; ~651 are "orphans" — imaged on owl but **not yet entered in the sheet** - (e.g. `O_angasi` 45–146, `O_lurida` 241–386, the Larken clam set). See `data/qc_report.md`. +- **100 samples** have at least one image, all of them browser-viewable +- **918 image files** indexed; **267** matched to a sample and **651** are "orphans" — imaged on owl + but **not yet entered in the sheet** (e.g. `O_angasi` 45–146, `O_lurida` 241–386, the Larken clam + set). See `data/qc_report.md`. ## Viewing locally