At the risk of going slightly off-topic, does anyone know what the
English for Geminationsstrich *is*? Since being disabused of the notion
that it was 'macron', I've been struggling to come up with something
better than 'that line that people call a macron but shouldn't'.
But thanks to Anne I can now at least say it in German.
John
On 21/02/2012 17:59, James Cummings wrote:
> As others have said, the use of it as Geminationsstriche seems to me a
> clear case of it being a marked abbreviation with a well-understood
> expansion. The choices are:
>
> a) Expand it silently
> b) Leave it encoded in Unicode as an m with a combining macron
> c) Use <g> to point to a <charDecl> with more information
> d) Encode just the <am/> marker
> e) Encode just the <ex> supplied text
> f) Encode a <choice> with an <abbr> and <expan>, the first containing
> the version with the <am> the latter containing the version with the <ex>
>
> I would argue that a) and b) are poor practice (sorry Paul) because
> silently expanding things is just wrong and it is a suspension mark
> not a macron. c) is ok, but probably not really what you want. d) and
> e) are also ok, just an editorial decision. My favourite, obviously,
> would be to have it fully marked up as in f).
>
> Perhaps something like
>
> <choice>
> <abbr>So<am>m̄</am>er</abbr>
> <expan>So<ex>mm</ex>er</abbr>
> </choice>
>
> Part of it comes down to whether editorially you view the 'm with a
> bar above it' as a single abbreviation marker which is replaced by two
> characters, or whether you view the first 'm' as being there and just
> marked with a bar above it. In the example above I'm treating the m
> with bar above it (apologies if your mail client doesn't render that
> unicode properly) as a single abbreviation marker that is replaced by
> the editorially-supplied expansion of two m's.
>
> Depending on the nature of your data, an XSLT stylesheet that reads in
> your existing XML, finds any m's or n's that have been marked this way
> and wraps the orthographic (whitespace or punctuation separated) words
> into a markup structure as above would be fairly straightforward to
> write (depending on how much interspersed markup there is). That kind
> of automated up-conversion should usually be done before adding lots
> of other manual markup.
>
> -James
>
|