TEI Physical Bibliography Workgroup: Summary of Activity November 28, 2004

<collation>

TEI-MS includes a collation element to ‘specify exactly how the leaves or bifolia of a manuscript are combined into a quire, etc.’. Use of the <formula> is recommended to encode ‘appropriate notational convention’. There is no provision, however, for encoding such notational convention.

The PB workgroup has been discussing possible encodings of traditional bibliographical collational formula. Two models have been pursued. One, the ‘concise’ method, aims to follow the syntax of collational formula in which the physical arrangement and relationships among leaves and gatherings are tersely and implicitly represented. For example, the formula for an octavo consisting of 7 gatherings signed ‘A’ through ‘M’ (‘J’ omitted) might be encoded as in this example: collation_concise_01.xml. ‘Modifications’ to the implied regular patterns, such as cancel leaves, could be encoded using a <mods> element. For example, an extra leaf inserted following A4 could be represented as in collation_concise_02.xml. There has been some discussion about using the <mods> element to record the omission of letters used as signatures in ‘ranges’ of gatherings. It might be preferable to simply note in an attribute value or element what alphabet, numeric sequence, and/or writing system was used in a gathering range.

A second model has also been proposed in each leaf and gathering is explicitly represented by an element. Thus, the same 7 gathering octavo book would be represented by 7 gathering elements each with 8 leaf elements as in collation_verbose_01.xml. As has been pointed out, this method is not so much a formula as a mapping and thus inappropriate as content of a <formula> element. Such encoding would certainly be easier to use than the ‘concise/implicit’ method, but also might be more difficult to encode, particularly from existing collational formulae. Both methods are still under discussion, neither recommended over the other. More thought and discussion is needed to insure that all the information contained in bibliographical formulae (e.g., cancels; inferred signing; etc.) can be adequately encoded.

<page>

While the PB workgroup has paid most of its attention to the encoding of collational formulae, some discussion has taken place about a <page> element. This element can be understood as a non-empty <pb> element which can contain richer identification and physical descriptive information as well as relevant content (e.g., pointers to image surrogates; transcriptions of forme work such as running titles, signatures, press figures, etc). This element could be pointed to and grouped into leaves, formes, gatherings, via <join> elements elsewhere in a TEI document, thus facilitating analysis and navigation of the text by the physical structures of the book in which it is printed, as well as the book itself.