Date:         Wed, 13 Sep 2006 14:04:49 +1000
Reply-To:     UNICODE-MARC Discussion List <[log in to unmask]>
Sender:       UNICODE-MARC Discussion List <[log in to unmask]>
From:         Andrew Cunningham <[log in to unmask]>
Subject:      Re: Character Repertoire Expansion TIme? (2)
Comments: To: UNICODE-MARC Discussion List <[log in to unmask]>
In-Reply-To:  <[log in to unmask]>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Disposition: inline

Lets make this as simple as possible. Assume I'm creating a MARC
record for a Russian title. the title will be romanized. I'm using a
cataloguing client that is Unicode based. The title has the character
sequence <U+0074, U+0361,U+0073> as part of the title. Many library
systems have character palettes to allow entry of diacritics. A clumsy
approach to character input in environments where a cataloguer may
have to frequently enter such characters. I'd prefer a keyboard layout
to directly type. A layout customised to the language or group of
languages I'm working with. But that requires that the client
application accepts direct Unicode input.

Secondly I'd want the character sequence to render correctly. That
requires being able to select a font that will correctly render the
combining diacritic either a Graphite font or an OpenType font with
appropriate GPOS or GSUB tables for the Latin script to handle the
placement of the diacritics relative to the base character.
Additionally the client application needs to use an appropriate font
rendering technology. On the windows platform the client application
needs to be built with uniscribe support, and you;d need to use the
application on a version of windows with the correct version of
Uniscribe (Windows XP with Service Pack 2 or Windows Vista).
Alternatively the vendor could license an appropriate version of
Uniscribe form Microsoft and ship it with the application in the
application directory. Alternatively, the Vendor could build in
Graphite rendering support into the application.

That's just looking a font display, and ignoring all the other aspects
of application internationalization. For example, you'd also need to
address cursor control. is the sequence <U+0074, U+0361,U+0073>
treated as a single unit or three separate units for cursor control
within the application. Will the application be designed to enable
selection and edition of the diacritic independently of the other
terms or as a single unit.

Andrew
-- 
Andrew Cunningham
[log in to unmask]

Vicnet
State Library of Victoria
Australia