Data model for ISMI

Conceptual objects in the database:

• (abstract-) TEXT
• ALIAS (title for text or name for person or place)
• WITNESS (aka manuscript)
• CODEX (of one or more witnesses)
• COLLECTION
• REPOSITORY (aka library)
• PERSON
• PLACE
• ms-city
• country

Objects

TEXT

an abstract text (mostly with a specific title) that exists as one or more witnesses.

Attributes:

• full_title (arabic)
• full_title_translit (transliteration, generated from arabic if not present)
• short_title canonical short title in arabic or latin (if applicable)
• language
• incipit i.e. beginning (arabic text)
• explicit i.e. ending (arabic text)
• dedication (arabic text)
• composition_date (hijra date, should generate a CE date)
  • composition_date_ce (legacy)
  • composition_date_num (legacy)
• text_type i.e. genre
  • matn, sharḥ, taḥrīr, ... (we need a full list here)
• table_of_contents (arabic text)
• text_arrangement (?) latin text description of number of books, parts, chapters, sections
• notes some text notes
• version_of_text (?) (is this the original text or a known different version/edition)
  • if there is more than one known version then also relation

Relations:

• TEXT is_commentary_on TEXT
• TEXT is_translation_of TEXT
  • is redaction (revision, abbreviation, paraphrase, ...) of text
• TEXT was_written_by PERSON
• TEXT was_composed_in PLACE
• TEXT was_dedicated_to PERSON
• ALIAS is_alias_title_of TEXT
• ALIAS is_alias_incipit_of TEXT
• ...

Problematic attributes:

• subject of text (legacy attribute genre, replaced by relation)
  • hay'a, zīj, mīqāt, astrolabe, etc
  • -> a canonical list of subjects should be present in the database
• format/structure of text (number of books, parts, chapters, sections)
• bibliographical information (secondary literature)
• number of versions [this number will generate additional forms to fill in unique details for each version; a version is usually an author's reedition but could have been made by students or someone else, even w/o author's approval]
• information whether the work has been published, translated, studied (relations)
• Text Arrangement: list of chapters, with a title in Arabic attached to each chapter.
• basis of text [the immediate text to which this is related as commentary, translation, gloss, etc.]

Problems:

• is a translation another abstract text? [no; see basis of text above]
• alias titles e.g. greek, latin linked to translations? [yes; also for persons, e.g. Avicenna=Ibn Sīnā]
• must Text Arrangement be a new definition? It is composed of several field! [yes since each component of the text arrangement corresponds to a particular page and line number in a witness; e.g. chapter 3 might correspond to f. 31a, line 2 in one witness, f. 42b, line 20 in another witness, etc.]

WITNESS

witnesses (manuscripts) are concrete written pieces of paper lying in a library.

Attributes:

• title (?) as written on the manuscript
• title_translit transliteration of title
• source_of_information (?)
• ms_number (?)
• shelf_mark
• codex_part
• incipit if different from TEXT
• explicit if different from TEXT
• colophon (arabic text)
• date_of_copying (hijra, translated to CE)
• folios total number and/or text
• page_dimensions height + width in cm
• written_area_dimensions height + width in cm
• lines_per_page
• writing_surface (predefined list)
• page_layout (latin text)
• script
  • include comments on legibility, ink, etc.
• other_features

• distinguishing features (e.g. state of preservation)
• colophon ("this was copied by") sample
• incipit/explicit if different from standard text
• information on diagrams, maps, tables
  • include number of diagrams, folios, color of ink, brief description
  • include number of tables if possible, folios, color of ink, brief description
• annotations/marginal notes
• Text Arrangement Pagination (from Text Arrangement into TEXT). This field shows folio and line number of the start of each chapter (from TEXT) in the list.
• image of the witness (new content type: Hyperlink).

Relations:

• witness is an instance (exemplar) of a text
• is part of a codex
  • is part of a collection
  • is in a library
• was copied by a copyist (PERSON)
• had a patron (PERSON)
• was written by a author (PERSON)
• can be direct copy of another manuscript
• was copied/written in (PLACE)
• annotator (PERSON)
• taught by (PERSON)
• studied by (PERSON)
• owned by (PERSON)
• studied in (PLACE/INSTITUTION)

Problems:

• alternative MS/catalog numbers?
• how to manage annotations?

CODEX

a codex is a physical, bound collection of manuscripts

Relations:

• contains one or more manuscripts
• was written in place
• is part of a collection
• is in a library
• can be direct copy of another codex
• was owned by person
• was patronized by person
• was copied by person

Information:

• dimensions
• material
• distinguishing features (e.g. presentation copy, decoration)
• table of contents (with titles and folio numbers)
• other information (including language(s) used in codex)
• owner(s) (relation)
• patron(s) (relation)
• reader(s) (relation)
  • for the above three: locations and dates if available
• information on additional material (diagrams, maps, tables, etc)

Problems:

• does every manuscript have to be in a codex?

PERSON

persons are (real or hypothetical) people that have some relation to a manuscript/text.

Attributes:

• name long arabic (or other script) name
• name_translit transliteration of name
• short_name canonical short name (in transliteration)
• birth_date birth date (hijra)
• birth_date_ce birth date (ce)
• death_date death date (hijra)
• birth_date_ce birth date (ce)
• major_influences (?)
• notes

Relations:

• PERSON is_author_of TEXT
• PERSON is_author_of WITNESS
• PERSON is_copyist_of WITNESS
• PERSON is_patron_of TEXT (WITNESS)
• ALIAS is_alias_name_of PERSON
• (...)

Problems:

• anonymous or insufficiently specified names
• alternative names (historical change?)

Anonymous authors:

• link to one generic Anonymous person (with optional notes)
• if more individuating information becomes available a new person will be created and linked as an author

ALIAS

alias titles or names for places and persons.

Attributes:

• alias arabic or latin or other text
• alias_translit semi-automatically generated transliteration if neccessary

Relations:

• ALIAS is_alias_name_of PERSON
• ALIAS is_alias_title_of TEXT
• ALIAS is_alias_incipit_of TEXT
• ALIAS is_alias_explicit_of TEXT

Place

places are places that have some relation to a manuscript/text/person/library/...

Types:

• institution (Madrasa Mosque in Cairo, etc.)
• city area (Fustat in Cairo)
• city
• region (Egypt, Iran, Maghreb, etc.)

Relations:

• place is part of other place (city is part of region, etc)
• manuscript was written in
• person was born/lived/died in
• library is in
• has alias name (at certain time)

Information:

• name of place (translit)
• original script name
• location(?)

Problems:

• linking to historical names?

Collection

a manuscript is in a collection in a library.

Relations:

• contains manuscripts
• is part of a library

Information:

• collection name
• owner and historical information
• description of holdings

Library

a codex is in a library

Relations:

• contains codices
• contains collections
• is in a place

Information:

• name of library
• more info

Secondary literature

Items:

• bibliographical items (book/article/etc)
• later separate authors with aliases

Relations:

• covers text/manuscript/codex/collection/library or person

Questions

• historical record of manuscripts in libraries?
• historical record of author attribution?
• historical record of ownership?
• have the possibility to maintain a list of Titles which were wrongly attributed to an author in secondary literature
• collect secondary literature
• format of dates
• workflow for public and non-public entries in the display?