New accent modifier types?

Brian McMahon bm at iucr.org
Wed Mar 7 11:03:57 GMT 2007


> When I first looked into accents, I thought the Latin1 and Latin2
> character groups should be sufficient, but Latin3 and Latin4 are needed
> to include some characters accents like over-bar.

Again, the existing markup has been adequate for the requirements of
Acta C (and E) for 15 years, but it won't cover all conceivable
possibilities. (The greatest need for accented characters is
in authors' names; and, as an international journal, we need to
accommodate authors from anywhere in the world, including Iceland.)

However, the more feature creep you allow (in terms of accommodating
possible markup), the greater the overhead. TeX, for example, seems
the natural choice for math markup; but even so, many math symbols
aren't defined in vanilla TeX, and one needs to be able to support
TeX extension packages - and to support them in a portable and
archival way. HTML and XML approaches can be followed, but also
require support of DTDs, schemas, entity tables etc. Unicode
offers a very large library of glyphs, but even that won't cover
all requirements. Sure, there are mechanisms for extending Unicode,
but they also require support and systematisation.

How far down this road should CIF go? At this stage, I wouldn't
care to commit myself. I will say, though, that the discussion
so far is encouraging, in that it begins to offer a clean way
(through MIME types and associated external delegates or handlers)
to support existing standard mechanisms for text processing
that should sit nicely alongside the image-handling mechanisms
of imgCIF.

Brian


More information about the comcifs mailing list