Policies - HCMID/germanicus GitHub Wiki

TBA:

We need to explicitly define how we are recording graphics and relating them to the text

The scribe does not always separate words as distinctly as a modern print edition. You should use white space to separate word tokens. Punctuation tokens may either immediately follow a word token or be separated from it by a space. Use a hyphen to show the boundary between enclitic -ve, -que or -ne from the word it is attached to.

Usage of white space

Examples:

Transcription:

tibi sacra fero

Transcription:

celsa~

Transcription:

musis ~

enclitics (-que, -ne, -ve)

Example:

Transcription:

docti-que

high stop: transcribed as ~ (Unicode high stop is evil)

Example:

Transcription:

auctor ~

There is no distinct "j" or "v" character for consonantal values of "i" and "u".

Example:

Transcription:

obuersa

The scribe uses a "y" in some Greek names. The form can be highlighted with an overdot. We take this simply as part of the letter form.

Example:

Transcription:

cynosura

Note the following common variant forms and ligatures. We do not treat these as abbreviations.

final m

Example:

Transcription:

cancrum

ae ligature

Example:

Transcription:

ae

et ligature

Example:

Transcription:

et

que ligature

Example:

Transcription:

Sidera-que

Paired within choice:

We use sic/corr to mark places where the original text has been corrected. In Bern 88, the scribe likes to mark a single character to remove with an underdot signifying "delete this"; the correction can be written directly over the "erroneous" text.

<choice><sic>adsis</sic><corr>assis</corr></choice>

When he deletes a longer sequence of characters, he marks the text to delete with an underscore, often "fenced" on one (as here) or both ends.

<choice><sic>donabat</sic><corr>donauit</corr>

A second hand has compared the text to other versions of the Aratea and noted alternate readings. These are often written immediately above the text to replace, and signalled with a symbol that looks like a "+" or "t". We record the original text with orig and the alternate reading with reg. We do not add a "+" character to our transcription because it is not part of the text: the TEI reg element encodes its meaning.

<choice><orig>conor</orig><reg>cogor</reg></choice>

<l n="34"><choice><sic>quid</sic><corr>quod</corr></choice> fidae comites~ prima incunabula magni </l>
<l n="35"><add>Fouerunt <persName>iovis</persName> attonitae cum furta parentis~ </add></l>