The WWP encodes acrostics so as to enable the words formed by the acrostics to be searched, counted, and otherwise processed explicitly. To make this possible we have created two new elements, <acrostic> and <acrostics>.
The <acrostics> element goes inside the <hyperDiv>, and is a container for one or more <acrostic> elements. Each <acrostic> element carries a target= attribute, which points to the id= of the smallest element containing the entire acrostic. The word formed by the acrostic is encoded as the content of the <acrostic> element, with phrase-level tagging as necessary if the word is a proper name, foreign word, etc. No renditional information will be encoded within <acrostic>, since that information is already encoded in the actual acrostic in the text. All words in the <acrostic> element will be entered in modern mixed case (only proper nouns capitalized), regardless of the capitalization of the original. Original typography will be encoded if appropriate within the <acrostic> element, since in some cases letter ambiguity may make it inappropriate to encode it in the main text. (See for instance the acrostic in Speght’s Mouzell for Melastomus, in which the acrostic spells “Joseph Swetnam” but the initial word of the poem is “If”.)
In the main text, we do not tag the individual letters which make up the acrostic word (although this is possible and is provided for by the DTD).