A Document Model for Zaima Chant

A Pragmatic Model Using Existing Standards

Ruby and Zaima both share a requirement for the layout of half-scale interlinear text bound to one ore more letters in a line of base text. A distinctive feature of Zaima notation that immediately distinguishes it from Ruby annotation is the regular occurrence of multiple interlinear lines, or “rows”, of text. No upper limit is set on the number of rows a line of base Zaima may have, eight rows however have been observed in corpus (አቋቋም፡ዘወርኃ፡በዓላት። Page 113).

Multiple levels or Ruby annotation are possible but irregular. Nested Ruby markup can be used to display additional annotation levels, this approach assumes some implicit association between the levels. In Zaima notation, the interlinear rows are fully independent from one another and the positioning and alignment between the base text and a given row cannot be a function of any other interlinear row. Attempting to apply the nested Ruby technique to format Zaima notation very shortly leads to complications with overlapping markup.

A model for Zaima notation that is able to avoid the overlap problem treats rows as residing on separate document “layers”. Each layer will contain the same base text and marked up with the notation of a single interlinear row. The notation line offset from the base text is incremented in each layer. When the layers are viewed together in a projection, the base text of each layer is superimposed upon itself and the notations appear in successive “rows”.

	Row 3: <ruby> <rb>አጽ</rb><rt></rt> <rb>ነ</rb><rt style="color: red;">᎒</rt> <rb>ነ</rb><rt style="color: red;">᎐</rt> </ruby>
	Row 2: <ruby> <rb>አ</rb><rt></rt> <rb>ጽነነ</rb><rt>ባሕራ</rt> </ruby>
	Row 1: <ruby> <rb>አ</rb><rt>ቡበ</rt> <rb>ጽነ</rb><rt style="text-align: right;"></rt> <rb>ነ</rb><rt>᎓</rt> </ruby>

The W3C Cascading Style Sheets standard provides the means to specify and position layers. An example applicataion appears in Figure 4:

Technical note on Figure 4: the markup sample raises the notation by shifting the ruby base bottom position and ruby text along with it. A more desirable approach is to change only the bottom position of the notation text (<rt>). Moving only the notation text is supported in Safari v11 and Chrome version 60 and earlier but requires <span> tags within the <rt> tags to apply the bottom property style to (applying bottom to the <rt> is ignored, thus the need for the inner <span>). Both appraoches have been applied in use case documents (apply bottom property to notation line only , apply bottom property to base). Recommendations for CSS experts are needed here.

The chant element

Categories:

Flow content

Phrasing content

Palpable content.

Contexts in which this element can be used:

Where phrasing content is expected.

Content model:

See prose.

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The chant element allows one or more spans of phrasing content to be marked with chant notations. Chant notations are short runs of text (symbols or letters representing notes) presented over base text, primarily used to score hymns before the advent of staff lines. Staffless chant scoring is practiced in a number of traditions, notably Byzantine, Gregorian (older form), Hebrew, Znamenny, Ge’ez, Syrian and others.

The content model of chant elements consists of one or more of the following sequences:

1. One or more phrasing content nodes.

2. One or more elements whose semantics are defined only as children of chant:

   • Universal chant elements: cbase, notation or notationGroup.

   • Zaima chant specific elements: beit, medgam, sereyu or haleita

     .

The chant element is somewhat analagous to ruby but with a different context for interlinear text.

The cbase element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The cbase element marks the base text component of a chant notation. When it is the child of a chant element, it does not represent anything itself, but its parent chant element uses it as part of determining what it represents.

An cbase element that is not a child of a chant element represents the same thing as its children.

The cbase element is short for "chant base" (the simpler base is an element with other semantics in HTML) has the same role as the rb element in Ruby annotation.

The notation element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant or a group element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The notation element marks the notation line component of a chant notation that will be laid out with respect to text within a preceeding cbase element. The element should display as a block type (?) and permit the position of its text contents to be positioned with the bottom and top properties.

When notation is the descendant of a chant element, it does not represent anything itself, but its ancestor chant element uses it as part of determining what it represents.

An notation element that is not a descendant of a chant element represents the same thing as its children.

The notation element has the same role as the rt element in Ruby annotation. By default the notation text font size will be 50% the size of the cbase text but may be set to another size at the discression of the content author.

[TBD: discuss the nested notation use case and provide example. ]

The notationGroup element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The notationGroup element associates two or more lines of chant notation against common base text. The individual lines of chant notation will be contained with separate group elements. Each group elements are children of a notationGroup and must contain identical sequences of chant base text. The base text within each group will later be superimposed so as to appear as a single base line. The cbase and notation pairings within a group is expected to be different for each group. With the chant base text with a group overlaid, the notation text will appear in the interlinear space between the base text and the preceding text line. The notation text within a notationGroup does not superimpose line the base text, but stacks in successive rows above the base. The layout of two group children of a notationGroup would be identical to nested rubies where the inner ruby text appears at the higher level (the position of the first group child of the notationGroup).

The child node index of the group children of notationGroup indicates its interlinear row position counted from top to bottom (e.g. first child appears at the top position, last child at the bottom). The interlinear spacing rules will follow the regular line spacing rules that would be applied to text at the notation font size (by default 50% of base text size).

A notationGroup element may also be a child of another notationGroup. This is useful for identifying the inner notationGroup element as providing the scope of a medgam element.

An notationGroup element that is not a child of a chant element represents the same thing as its children.

The group element

Categories:

None.

Contexts in which this element can be used:

As a child of a notationGroup element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The group element encompasses a collection of cbase and notation representing a single interlinear line of chant notation. A group element may be empty which indicates that no text should be presented at its row index position. The group element should display as a block type (?) and permit the position of its text contents to be positioned with the bottom and top properties.

[Issue: bottom or top as a style attribute of group should only apply to the contents of the notation element. Should a notation-bottom element be added (top seems non-essential)?]

An group element that is not a child of a notationGroup element represents the same thing as its children.

The beit element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Flow content

Phrasing content

Palpable content.

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The beit element marks the scope of the Beit (ቤት) context of a portion of Zaima chant. The Zaima Beit context is given with an identifier that is the abbreviation of the full name of the Beit. The identifier will appear in the left side margin of the chant text adjacent to the first line where the context begins. The abbreviation text may be stylized and will appear within the required marker child element. The full name of the beit will then be given in the optional child dfn element which should not appear in the document by default (display: none). The web author is left to decide how to display the full name of the beit.

A beit element that is not a child of a chant element represents the same thing as its children.

The marker element

Categories:

None.

Contexts in which this element can be used:

As a child of a beit, medgam or haleita element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The marker element identifies a mnemonic “key” that is the referent to the definition value of its parent element.

A marker element that is not a child of a chant element represents the same thing as its children.

The medgam element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The medgam element is used to set the Medgam (መድጋም) counter for a line of chant. The counter is placed in the required marker child element which may also be stylized (typically set to a smaller font-size value).

The medgam scope applies to two or more interlinear lines of notation which may be fewer than the total number of interlinear lines above a chant line. The medgam scope will also be disjoint from its open and closing elements. The suggested convention for binding the medgam to its scope is to give a medgam element an id attribute and apply the id value to the class attributes of the notation, group, and notationGroup elements that collectively form the scope of the medgam. It is left to the author to determine if the medgam should apply to just additional notation line only, or also include the primary notation line. Example 2 demonstrates the later case where the primary notation line of the mode (silt) is included for clarity since a notation line in another mode is also present.

A medgam element that is not a child of a chant element represents the same thing as its children.

The sereyu element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The sereyu element is very similar to the medgam element in that it has a disjoint scope that will again be referenced by applying its element id as a class type. The sereyu element also has a required child marker element. By default the sereyu body specifies the components of a miliket abbreviation and definition. A sereyu may also be used to define elements for later reference and mark the point of definition as an exemplar. Presently, only the haleita element requires defintion in this style (See Example 3).

A sereyu element that is not a child of a chant element represents the same thing as its children.

In the following example sereyu elements encompass the full names of the ለሔዋን and ተፈሥሒ. milikets, identify the letters that form their respective abbreviations, place the contiguous abbreviation in dfn tags and position the ረዩ marker following in the expected location. The marker itself is not notation, but by default should be positioned as if it were. However any notation text that may be present will have to flow around the marker -left, right or above [TBD: review rules for marker positioning.]

The following example applies the sereyu tags to define a haleita reference from an exemplar. The haleita would later be inserted into a document through an idref type attribute. For example <haleita idref="ዖደ"/> in a document would display the only the marker part inline, an event would be used to display the definition.

The haleita element

Categories:

None.

Contexts in which this element can be used:

As a child of a chant element.

Content model:

Phrasing content

Content attributes:

Global attributes

DOM interface:

Uses HTMLElement.

The haleita element marks up the Haleita (ሃሌታ) shorthand which is keyed from the Halleluya Number, Beit and Silt. The Halleluya Number will be marked up in the required marker child element which is also expected to be stylized. The fully expanded haleita may then appear in the optional dfn element which should not appear in the document by default (display: none). The web author is left to decide how to display the expanded from of the haleita. In practice the dfn child element is not expected to be used since the content within would be complex and desirable for reuse. Rather, haleita content would be displayed via an idref type attribute to the definition created with a sereyu.

A haleita element that is not a child of a chant element represents the same thing as its children.

Issue: Review if a haleita ID in the margin note is also a beit, or is just there to help dereference a haleita number. Is every beit defined in a haleita definition? Or are beits fully orthogonal and do not require a haleita? In this case the haleita is another entity that has a marker residing in the margin.

Digua Sample Applying Many Elements

An interactive rendering of this sample appears in the What is Zaima? article.