|
The ARTFL Project French Women Writers Errata |
| The ARTFL Project | Database Home | Bibliography | Search Database | User Manual | Comments |
|
The h1 was used in the orginal data preparation to mark up the title of the whole document, when we scanned the title page. Thus, in some cases, we get results where clicking on h1 gets the whole of the document, or pretty close to it. It may be that there is only one h1 for the whole of the document, there are no sections or chapters, and so the only useful division is really the page numbers and paragraphs/sentences. We are deciding how to handle this in our procedures. My feeling is that we should stop using the h1 tag for the title as it appears on the title page. I think our reasons for doing this originally stemmed from our residual habits of thinking of h1 in terms of on-screen display, rather than in terms of oragnization and text structure. We might even dispense with including the title page in the scanning, since this same information is reproduced electronically in the Dublin Core, which is accessible to the user in the [header] tag from the table of contents. This would mean that for a work without chapters, sections, etc., there would be no h1 link from the results. Bibliographic Searching:
Author: At this time parentheses (( )) are not searchable in the
author field. Thus, block-copying a name including a parenthesis
will produce a "No documents found" message.
Title: at this time the following punctuation marks and symbols
produce a "No documents found" message: parentheses (( )), semi-colons (;),
colons (:), ampersand (&), apostrophes ('), single and double quotes, and
brackets ([ ]) as well as the dollar sign. The following punctuation marks
have no adverse effect on a title search and, if appearing within a string,
must be entered: period (.), hyphen (-), question mark (?), exclamation
mark (!), forward slash (/), and comma (,). In all cases, spacing must match
exactly that in the bibliography.
Date: The dates in the database range from 1545 to 1864.
In all cases, "date" specifies the dates of original composition.
For the dates of publication for the editions used in data-capture,
see the Database Bibliography.
Orthographic Considerations:Because of the range of dates in the database, the same word may be spelled several different ways in the various texts. For example, verbs in the imparfait may end in ait, the modern spelling, or oit, the now-obsolete form. Keep these possible variations in mind when searching, particularly across a range of dates.Word accentuation is not always consistent. Wildcard characters or Boolean operators can help detect such anomalies. One may enter assur.d or #atE or hono.?r or honor|honour to achieve the desired results.
Wildcard and Boolean Operator ProblemsOne can't search a #word with the Boolean operator OR. #augustinus|#claudius only gives the count for the first term by itself.There is currently a set limit of 2,000 for the number of unique forms into which a search term can expand. Searches that pass that limit will not run. We're looking into how high we can set the limit.
Formatting and DisplayItalics are preserved just as they exist in the original text.Block quotes and epigrams are marked as paragraphs, with the addition of a <center> tag. Images: Currently, the database does not contain any images. Notes: Currently, the database does not contain any notes. The only text acquired in the data capture is that actually written by the author. Therefore, no notes, or other parts of the critical apparatus, are acquired. Prefaces, avant-propos, introductions, etc., are acquired only if they were written by the text's author, and not by its editor. |
| ATILF | The University of Chicago | EFTS | Subscription Information | Current Subscribers |