Suggestion for adding a Proofreader role

[Chu-Ren Huang]

Dear All:

We would like to suggest the addition of a Proofreader role to OLACMC.
This suggestion was originally made in a paper (Ru-Yng Chang and Chu-Ren
Huang. 2002. OLACMS: Comparisons and Applications in Chinese and Formosan
Languages. Proceedings of  The 3rd Workshop on Asian Language Resources
and International Standardization, A Post-COLING2002 workshop.)

Since all digitized material must be proofread to ensure quality, it is
important to know who the proofreader is. A proofreader should be able to
refer to an individual, a team, or a standard procedure. This is very
crucial for heritage data, since a competent proofreader may require some
reading skill of a language that is no longer in use.

Take classical Chinese text (or Latin) for example. To construct an
archive based on classical Chinese texts, input and proofreading cannot be
avoided. The standard procedure here at Academia Sinica involves inputting
the text twice (manually or automatically, but by different teames), run
an automatic proofreading program on the two versions, then use human
proofreader to go over the differences for at least three more runs (while
proofreading the identical parts quickly.) Such a standard procedure can
ensure data integrety and accurracy to a very high standard (less than 1%
of error remains.)  The mor traditional way require 7 runs of
proofreading, with one proofreader reading the text backwards [to avoide
automatic self-correction in reading)>  In other words, a text proofread
by the AS team should be highly reliable. While a text archived by a
Chinese graduate student, input and proofread by him/herself may not reach
the same level.

Chu-Ren