Archaeological Documentation Is Only as Good as Its Queryability
A find sheet filled out in the field is table stakes. Every project produces them. The question is whether, three excavation seasons later, you can answer the question: "How many pottery vessels with Nile clay, medium texture, and a red slip were recovered from Shaft IV contexts across tombs QH 30 through QH 36?" If your documentation system requires manual review of hundreds of scanned sheets to answer that, your documentation is not really a system—it is an archive of labor with no retrieval mechanism.
The Qubbet el-Hawa Archaeological Database is built around the premise that field data has to be query-ready the moment it is entered. Every attribute that could vary by find—context, material, clay type, vessel shape, slip colour, bioarchaeological designation—is a structured field, not a prose note. Prose lives in the Details and Analysis fields, where it belongs. Classification lives in choice fields, where it can be filtered.
The Context Architecture and Why Shaft Filling Is Its Own Designation
The multi-select Context field covers twelve distinct provenience types: Corridor, Cult Chamber, Atrium, Shaft, Shaft Filling, Room with Burial, Coffin Chamber, Passageway, Interior, Courtyard, Tunnel, and Atrial Shaft. Shaft Filling is enumerated separately from Shaft for good reason. Finds recovered from shaft fill represent a mixed, often disturbed deposit—their stratigraphic relationship to the primary burial context is uncertain by definition. Lumping them with securely stratified shaft floor material would corrupt comparative ceramic sequences.
The Shaft and Tunnel fields use Roman numerals and Greek letters respectively, matching the site's established nomenclature. Coffin Chamber designations run α through η, again following standard Qubbet el-Hawa recording conventions. These are not invented by the template; they map directly to the published tomb descriptions.
What this structure enables is cross-contextual filtering that would take hours to replicate manually. Filter for Object Type: Pottery, Context: Coffin Chamber, Shaft: II—and you have a clean corpus for ceramic analysis of a specific burial subunit without pulling a single physical file.
Clay Texture, Slip Colour, and the Ceramic Variables That Actually Matter
The ceramic description fields form the most technically dense section of the record. Clay Type distinguishes Marl from Nile clay—the primary technological divide in pharaonic ceramic production. Marl clays originate from Upper Egyptian limestone deposits; Nile alluvial clays are the default fabric for most utilitarian wares. Misidentifying this splits your corpus on a meaningless axis.
Clay Texture (Fine, Medium, Rough) combined with Clay Colour (Red, Brown, Light Red—multi-select) and Slip Colour (Red, Light Red, Dark Red, Brown, Yellow, Black—multi-select) gives you the Munsell-adjacent attribute set that ceramic specialists actually use for typological comparison with published parallels. The Vessel Shape field (Open, Closed, Other) is deliberately coarse at the top level because the nuance goes in Shape: Other as a text entry—a realistic acknowledgment that an Elephantine typology cannot be fully pre-enumerated.
Wall thickness measurement sits implicitly in the Height, Circumference, and Base dimensions. These three numeric fields, combined with Board Thickness for wooden objects, give the morphometric baseline needed for direct comparison against Elke Rosenow's corpus or the Edel publication referenced directly via the Edel Page Reference field.
Rösing Number and the Bioarchaeological Layer
For skeletal material, the Rösing Number field carries the morphological sex determination code from Rösing's standard method—the five-grade scale (M, M?, ??, W?, W) that maps to the Gender field's five choice options. Age from and Age to define the probable biological age range in years. These are not ornamental fields; they are the minimum dataset required for a proper bioarchaeological entry in a peer-reviewed appendix.
The pairing of Object Type: Human Remains with Gender and age range fields means skeletal material can be analyzed independently of other find categories, filtered by sex estimation and age cohort, and cross-referenced against associated grave goods without restructuring the database.
Epigraphy, Bibliography, and the Scholarly Chain
The inscription infrastructure is three fields: a boolean flag (Any Text or Inscriptions?), a verbatim Inscription field, and an Inscription Translation field. This separation is non-trivial. The verbatim transcription and the translation are different classes of evidence. They can be updated independently when a reading is revised, and the boolean flag allows the entire inscribed corpus to be filtered out of the full dataset in a single step for frequency analysis.
The Similar example field links entries within the database through a related-record relationship. A scarab from QH 34 Shaft III can be linked directly to a formally similar piece from QH 29 Coffin Chamber β. The relationship is navigable in both directions; you can move from one comparandum to the next without returning to a search interface.
Edel Page Reference and Other Sources round out the bibliographic layer. Edel's Die Felsgräbernekropole der Qubbet el-Hawa is the primary published reference for this site, and the dedicated field for his page numbers means literature cross-referencing is built into every entry, not added as a footnote afterthought.
What the Database Delivers When the Season Ends
Post-season, you have a structured corpus that survives the excavation team's dispersal to home institutions. The Entry Added/Edited by field logs responsibility at the individual record level—when a pottery specialist in Tübingen queries a vessel description that does not align with her reading of the object photographs, she can identify who entered the data and raise the discrepancy directly.
Object Image, Illustrations, and Context Maps fields attach visual documentation directly to the material record. No separate photo archive cross-referenced by hand. The find, its illustration, and its site map are a single retrievable unit. At two thousand entries, that integration is the difference between a functional comparative database and a folder of JPEGs with ambiguous filenames.