Treatment of OD col stencils

File last modified
For other MEC files, see the  MEC Index page.



Task: regularize OD col. stencils
---------------------------------

1. Extract list of OD-col stencils using perl ex-stencl.pl *.sgm >> odcol
   Sort them by (1) running pre-sort.pl on the list, (2) running sort/uniq
   on the resultant file in TextPad, (3) running post-sort.pl on the sorted
   file.

2. Identify the stencils for which we can say *something* about the source.
   These should include:

   (1) stencils that simply list a manuscript ("MS Rawl.B.520")

   (2) stencils that list an identifiable book as the source.

3. Make sure that you do NOT include stencils that may refer to works
   already given MED stencils. These should be given new stencil
   groups or MS groups in existing entries, not new entries of their
   own.

   Also, of course, exclude from consideration all OD-col stencils 
   that have already been assigned to existing entries and given IDs.


4. Create a new HyperBib batch that includes (1) a separate entry for
   each stencil identified in step no. 2 (except those excluded under
   step no. 3); and (2) a blanket entry for all the others.

   When creating the stencil for each entry, stick fairly close to 
   the OD's own stencil, especially if you are in any doubt about how
   to interpret it, but adapt it to MED style too, where you can. E.g., in
   stencils that include MS abbreviations in the title, use the MED
   abbreviation, not the OD abbreviation (Rwl not Rawl).

   When creating the individual stencils, omit page references and the like
   from within the body of the stencil, as with ordinary MED stencils.

   Entries based on print sources will probably require some bibliographic
   research to discover the appropriate source and the relevant details.

5. Create a regularization script that, when applied to the MED, will

   (1) FOR THE STENCILS FOR WHICH YOU HAVE CREATED NEW ENTRIES:

       move the page reference (etc) out of the stencil and into the
       <BIBL> where it usually goes, at least for the stencils for
       which you have created new entries.

   (2) FOR THE STENCILS FOR WHICH YOU HAVE NOT CREATED NEW ENTRIES
       (i.e. those that are destined to be mapped to the blanket entry):

       regularize as much as possible divergent forms of the OD stencil
       (spelling, spacing, even reference style where possible).

6. Create a map file based on your new bib batch. The blanket stencil
   will be a <NRGRP>; the others will probably often be
   <PRGRP>s. Check if there are date discrepancies in the stencils
   in the dictionary. In general, if there are, we will want to leave
   them in place, though a fluke date or two we can  be forgiven
   for regularizing: the former case requires a <PRGRP>, the latter
   a regular <GRP>. If there are no date discrepancies in the data,
   use <GRP>.

7. Run the regularization script from no. 5 on the whole MED.

8. From the altered MED files (altered by no. 7), extract a new list
   of OD-col stencils, sort as in step no. 1.

9. Map the stencils extracted in no. 8 to the map file created in no. 6.