Annotation and indexing (Mondrian)
Contents
Annotation at the level of the letter
Annotation for which the teiHeader provides a natural place is stored there. Besides that, each letter or manuscript contains a <div> containing annotation, structured as follows:
- a number of typed notes with fixed subjects (now: source of transcription, dating and remarks)
- zero or more of ongoing topic notes
- zero or more notes to specific locations in the text
An example:
<div type="annotation">
<div type="typednotes" xml:lang="nl">
<note type="transcrSource">Manuscript.</note>
<note type="dating">Poststempel.</note>
<note type="remarks">Beeldzijde: watermolen bij Delden, nabij Hengelo; cf. ongoing topic.</note>
</div>
<div type="ogtnotes">
<note target="MEP_0011.xml#verblijfplaats">Mondriaans verblijf in Hengelo bij Hulshoff Pol.</note>
</div>
<div type="notes" xml:lang="nl">
<note xml:id="openingTeekening">De tentoonstelling <ref target="exhibitions.xml"><hi rend="italics">Aquarellen, Teekeningen en ... </note>
</div>
</div>
Explanatory notes
To create a note
- Put into the transcription or translation a pointer (<ptr/>), with @target consisting of '#' and a short keyword (start with letter, no spaces) and @type=“note”.
- Create a <note> into the <div type=“notes”> with the text of the note. Its @xml:id must must be the keyword used in the pointer's @target.
- Put all notes in a single language in the same div element.
- Order the notes by their referred location (technically irrelevant, but helpful to the human eye)
For example:
<p>This is an example<anchor target="#nootje" type= "note"/> of how to create a note</p> <div type="notes" xml:lang="en"> <note xml:id="nootje">This is the note</note> … other notes in English… </div> <div type="notes" xml:lang="nl"> <note n="nootje">Dit is de noot</note> … other notes in Dutch … </div>
Recurring annotation
See Recurring notes to persons and other subjects.
<rs>
Referencing string. Into <rs> are encoded names, with attributes, so they can be easily identified (i.e. You have different spellings of the same person's name and you want that they point to the same person), retrieved, indexed.
- @type kind of entity encoded (i.e. Person, exhibition, museum, journals, location, etc). The available types have been defined in the schema
- @ref to indicate an external url (RDKarchive url)
- @key a coded value that identifies the entity encoded (i.e. a number, a normalised spelling, etc.) We prefer to use @key rather than @ref.
It is possible to use multiple values in the key or ref attributes.
In the postal data (transcription of envelope) we don’t use <rs> to index the addressee, as his role is known from the letter metadata. Similarly, we don’t index persons saluted at the beginning or end of a letter, if these are addressees. (But we do tag ‘Jan’ in ‘Give my love to Jan’ if Jan is not an addressee).
When encoding a referring phrase as <rs>, we only tag the noun. This avoids potential problems with other words that need their own encoding.
Not:
<rs>the woman who lives next door</rs>
but
the <rs>woman</rs> who lives next door
In the case of multiple references to a person within a single paragraph, we encode each of them.
References
The element <ref> points to another resource and it is used for instance when a letter, a painting or another resource is mentioned in the text. @target provides the resource location (local or not).
<ref> is also used for pointing to the bibliography. For example:
... as mentioned in <ref target="biblio.xml#Andersen1861">Andersen 1861</ref>, we ...
Here, 'Andersen1861' should be an entry in the bibliography.
