search

rbms-bsc / DCRMR

Other
12 stars 5 forks source link

Indicating when basic vs. normalized transcription was used #200

Open rbms-bsc opened 1 year ago

rbms-bsc commented 1 year ago

Indicate whether basic or normalized transcription was used and consider making this a required note. This will likely be dependent on recommendations from LC and other groups regarding recording data provenance.

Note: This issue was originally posted by @elizhobart on 13 October 2021 as a comment on issue #36, which was created as a catch-all for issues related to future development. For ease of tracking, the comments have been moved to their own individual issues and tagged as "future work."

rbms-bsc commented 1 year ago

The Editorial Group decided to remove the instruction at 0.4.12.2, "Optionally, when converting uppercase to lowercase, add diacritical marks that are not present on the source of information in accordance with the pattern of usage in the text."

Concern regarding this instruction was raised during the DCRMR public hearing on Dec. 8 and in a thread begun on Dec. 9, 2021 on DCRM-L. @deborahjleslie noted in response:

This option was added during early stages of the of what we were then calling DCRM2, on analogy with ijuv. If memory serves, it was along the lines of "well, if we do this with IJUV, shouldn't we do it with diacritics too?" Conceptual rather than practical or experiential grounds. Bob makes a really good case for removing the option.

The method of transcription is really significant in how far it is used to identify variants and the like, and is even more significant with original vs normalized punctuation. I would like to see it become a required note in every record identifying it. It may seem tedious, but DCRM(G) requires that every record identify the source of the title, and serials cataloging gives a "Description based on" note. If we were to go this route, this community could come up with normative wording that is brief but descriptive enough to alert catalog users to the parameters of transcription.

The previous issue posted on Oct. 13 (referenced at the beginning of this comment) also addresses similar issues surrounding transcription and data provenance. When we revisit these issues and how to address them in DCRMR, we can consider whether to re-insert 0.4.12.2.

Note: This comment was originally posted by @jess-grz on 13 December 2021 as a comment on issue #36, which was created as a catch-all for issues related to future development. For ease of tracking, the comments have been moved to their own individual issues and tagged as "future work."