Mondrian: Decisions during eDITem project

These are decisions taken during the eDITem project that have to be worked into the rest of the documentation. For the moment, we document them here.

We will not be using sic and corr, but orig and reg. We only keep the changed (added, deleted) string within the element and keep the unchanged part out of these elements. We move the @type from the sic to the reg (not to the orig: not allowed).

Organizations:

We create a separated .xml file for organizations. In the text we use the following encoding: <rs type="org" ref="org.xml#xxx">

Sample of the Organizations file:

  -listOrg
       org
         @ana (value: museum, gallery, artistassoc, firm, org) and @xml:id 
        orgName
        place
        country 
        settlement
        desc

Layout:

For now we regularise layout of postcards the way we do letters.

We display date/place always full left, on a single line, followed by one white line. We change the current encoding for address and date in dateline (to be added by a stylesheet):

<opener>

  <dateline> … </dateline>
  <salute> … </salute>

</opener>

We display salutation always full left, beneath the date/place after a white line and the first line of the letter immediately below the salutation, set full left.

Paragraphs are indented except for the first line of the letter text and after a white line.

Postscripts are shown after the signature and after one white line.

Superscript (and general rule for unusual abbreviations): keep encoding exactly what’s on the manuscripts and then the editors will decide how to regularize it.

Es: <choice><orig>N<hi rend="super underline2">o</hi></orig><reg>No.</reg></choice>

Unusual abbreviations: to be coded as <choice><expan>. The usual abbreviations will be described in a list.

White lines: keep them in xml as they are encoded (vertical spaces) and display them as one white line when they occur within the body of the letters. Only after <dateline> use the @rend=“norend” encoding, since we don’t want to display the white line/lines here. (The @rend=“norend” will be added through a stylesheet)

Quotation marks: still under discussion.

Comma and full stop: to be encoded with <orig> and <reg>

Single underlining: encode as underlined and display in italics

Multiple underlining: encode as multiple underlined and display in italics. Annotation when the underlining have a specific meaning.

Atypical writing, combinations, repeated quotes, quotations, lines and dots: not relevant or not present

Parentheses: all parentheses are rendered as ( ) And also transcribed as ()?

If not, eg: <choice><orig>[</orig><reg>(</reg></choice>

Uncertainties: encode with <choice><unclear><unclear>. Display content of first unclear element. Always add note?

Additions in margins: encode with <metamark> with @function and @place and show (and encode) them where they are supposed to be insert by Mondrian. It is also necessary to keep into account the direction of writing? Not sure whether we really need @function on metamark. @place on <add>

Appendices: encode as a separated division with @type="appendix", and a new <pb/>

We make explicit what is preprinted or pre-written by using the @rend with "preprinted" or "prewritten" values. We do that only when Mondrian fill in preprinted/prewritten address or forms and not for fixed parts of the letters (as letterheads).

Table: encode as such when they occur (with , <row> and <cell>). Introductions: not provided at this stage: most of the letters don’t require introduction and when pieces of information are needed, they are provided within annotation. This point could be discuss again in the future. We will make the XML freely available for users If Steering Committee agrees.