Source

Leiden RMO — Egyptian collection

Egypt, Sudan, Nubia (find spots) · Predynastic through Coptic

Objects

Images

132

Cross-references

Browse all 12 objects → Visit source →

Sample objects

Source details

Type: museum_collection
Access: blocked
Stack: wordpress-cf-gated
License: Unknown (collection-side API not reachable; site terms unclear)
Modality: none
ID system: Leiden accession (form unknown — collection-side not reachable)
ID tier: tier-2
Inscription richness: 1/10
Cleanliness: 3/10

Form breakdown

papyrus: 8
statue: 3
jewelry: 1

Notes

## Discovered schema (Leiden RMO via access probe, surveyed 2026-05-23) ### HEADLINE: Pre-flight false positive — Leiden is essentially access-blocked Phase 1.5 pre-flight tagged Leiden `open` based on the JSON response from `/api/`. **That was a 90-byte WordPress "JSON API" plugin root**, not the actual collection API. The Leiden collection is not directly addressable: - `/api/` (WordPress JSON API plugin) — returns plugin metadata only; `core/get_posts` returns 17 BLOG POSTS, not collection objects; `taxonomy/get_taxonomies` returns HTTP 500 (broken). - `/wp-json/wp/v2/` (WordPress REST API) — works for built-in types (posts, pages, media) but the collection-side custom post types (`topstuk`, `aanwinst`, `3d_object`, `tijdelijketopstukken`, `collectie_categories`, `collectie_plaatsnaam`) are registered with `show_in_rest=false` and **return HTTP 404** on the REST endpoint. - `/collectie/` (Dutch collection search page) — **HTTP 403 Cloudflare Turnstile** on every UA. Same gating pattern as British Museum (#4). - `/collection/` (English) — returns 200 HTML intermittently before Cloudflare starts returning empty bodies. The 128KB HTML is a SPA shell whose search backend URL we could not extract before getting throttled. - `hdl.handle.net/21.12126/RMO` (handle from pre-flight) — HTTP 404. The probes have been stored as `_probe:*` samples on this source. ### Stack identification (Test 1 from the brief) **WordPress + Cloudflare + WPDM plugin.** Confirmed by: - Cookie `__wpdm_client` (WP Download Manager plugin) - Server header `cloudflare` - WordPress REST API at `/wp-json/wp/v2/` - JSON API plugin (older, by Dan Phiffer) at `/api/` **NOT Axiell**, NOT KE EMu, NOT CIIM, NOT Adlib. Leiden does not run on any commercial collection-management system that exposes a public API — it's a CMS-managed site whose collection data lives behind a Cloudflare- gated PHP search backend. Implication for the "Axiell = transcription-rich" hypothesis: **Leiden doesn't disprove or confirm it.** Hypothesis remains at N=1 (Fitzwilliam). ### MdC transliteration format (Test 2 from the brief) **Untestable here.** Since we cannot reach Leiden's inscription field (if any), we can't confirm the format. The transliteration-fuzzy-match cross-ref path stays at one confirmed museum source (Fitzwilliam) for the time being. ### Wikidata fallback (the standard BM/BKM pattern) - P195=Q1813814 (Leiden RMO collection): **0 items.** - No dedicated Leiden artwork-ID property exists (cf. P8910 for Fitz, P4740 for BKM). Wikidata simply doesn't cover Leiden. - Egyptian-origin items: 0. - Egyptian-find-spot items: 0. - TM cross-refs: 0. TLA cross-refs: 0. ### Europeana fallback - `DATA_PROVIDER:"Rijksmuseum van Oudheden"`: **0 items**. - `who:"Rijksmuseum van Oudheden" AND Egypt`: 1 item. - `Leiden Rijksmuseum van Oudheden Egypt` (free text): 3 items. Leiden does not contribute to Europeana directly. ### Five-way museum comparison (museum pattern N=5) | metric | MET (#3) | BM (#4 via WD) | BKM (#5 via WD) | Fitz (#6 direct) | **Leiden (#7)** | |---|---|---|---|---|---| | total surveyed | 77 | 57 (13 Eg) | 4 (4 Eg) | 109 (80 Eg) | **0 (probes only)** | | has inscription field | 0/77 | 0/57 | 0/4 | 6/109 | **unknown (gated)** | | with transcription text | 0/77 | 0/57 | 0/4 | 6/109 | **unknown (gated)** | | with TM cross-ref | 0/77 | 2/57 | 0/4 | 0/109 | **0 (via Wikidata)** | | with TLA cross-ref | 0/77 | 1/57 | 0/4 | 0/109 | **0 (via Wikidata)** | | with image | 77/77 | 48/57 | 4/4 | 48/109 | **unknown (gated)** | | with structured find spot | prose | no | no | 61/109 | **unknown (gated)** | | with find-spot coordinates | no | no | no | 50/109 | **unknown (gated)** | | with structured period | partial | prose | prose | 71/109 | **unknown (gated)** | | direct API access | yes | no (CF) | no (Vercel) | yes (key) | **no (CF on /collectie/)** | | stack | proprietary | proprietary | proprietary | Axiell + CIIM | **WordPress + Cloudflare** | | coverage on direct path | 100% | <0.1% | <0.01% | 100% (~16,801) | **0% (collection gated)** | ### Flag interpretation - Flag 1 (object-vs-inscription modeling): UNTESTABLE for Leiden directly. - Flag 2 (transcription presence): UNTESTABLE. Cannot confirm or deny. - Flag 3 (cross-ref gap): CONFIRMED (Wikidata returns 0 for Leiden in every Egyptian-likely filter we tried). - Flag 4 (image licensing): UNTESTABLE. Image URLs not reachable. ### Implication for the museum pattern after N=5 Three of five museums are now Cloudflare-gated (BM, BKM-on-Vercel, Leiden). The pattern is solidifying: most modern museums sit behind anti-bot WAFs. Only MET (publicly-funded US institution with explicit open-data policy) and Fitzwilliam (Axiell+CIIM with key-required access) supply direct collection data. The next museum to survey from the `open` tier (OI Chicago / ISAC) should clarify whether public-funded archaeological institutes follow the MET pattern or the WAF pattern. ### Pre-flight lesson learned The pre-flight classifier saw `/api/` return JSON+200 and tagged Leiden `open`. The classification was correct in that an endpoint at /api exists, but the endpoint isn't a collection API — it's a generic WordPress plugin root with no collection data. **Lesson: the next time we run a pre-flight, probe one or two follow-up paths under any JSON- returning root (e.g., /api/objects, /api/collection, /api/search) and require at least one to return collection-shaped data before tagging `open`.** Logged in decisions.md. ### Open questions / follow-ups - The English `/collection/` HTML page is intermittently reachable. A targeted scrape using a real browser (Playwright/Selenium) with Cloudflare Turnstile bypass might extract object data — but that's out of scope for Phase 1 surveying. - Leiden has a longstanding partnership with Egyptian universities and may publish data dumps through academic channels (Zenodo, Open Science Framework, OpenContext). Worth a one-off search. - The Pelagios Commons gazetteer historically references Leiden objects; could be a route to find spot data for Leiden items via cross-ref. - Direct contact with Leiden's data team — they have an active digital humanities program; an academic-use access request may succeed.

Leiden RMO — Egyptian collection

Sample objects

Statue (bone)

Seated statue of Anubis

Statue/vase of the god Bes

Papyrus; Demotic; 8 lines

Papyrus; Demotic; 8 lines

Papyrus; Demotic

Papyrus; Demotic

Papyrus; Demotic

Papyrus; Demotic

Papyrus; Demotic

Papyrus; Demotic

Scarab; Horus, standing