I've read over a number of document analysis reports by a variety of authors in an effort to distill the format and content of such a document.
Observation: I note that some (genealogists) include a transcription or extract and some do not. In asking some more experienced genealogists, it became evident that those persons included a transcription or extract to facilitate later publication (via a simple cut and paste). However; I not that this specific rationale is not always in play.
Question #1: Is it good practice to always include a transcription or extract when writing about a document and, if so, why?
Observation: I understand (from what I've seen) that the gold standard for a transcription is to faithfully represent what was observed in the original. However; when performing an extraction, from a "fielded" document like a marriage certificate or census, it is not always possible and/or even practical to include everything printed on the document.
Question #2: When performing an extraction; is the "gold standard" to include (as a minimum) any aspect that may have a bearing on the analysis of the document? That is; for a census, would a simple field-value table and supporting notes on relevant information contained in the header/footer and any observations be adequate?