[Suppress-Script] Initial list of 300 languages

Peter Constable petercon at microsoft.com
Tue Mar 14 19:41:27 CET 2006

> From: ietf-languages-bounces at alvestrand.no [mailto:ietf-languages-
> bounces at alvestrand.no] On Behalf Of McDonald, Ira

> > Let's work through a specific example.
> > Document tagged ga-Latg-IE. Printer defaults to en.
> >    (1) What will go wrong?
> >    (2) How is this different from a document tagged en-US, which
> >         contains a single word which, if it had been tagged,
> > would have
> >         been tagged ga-Latg-IE?
> >
> Depending on the tagged charset, various things will go wrong.
> If the charset tag is 'UTF-8' it's useless info, so the printer
> will guess the script and character repertoire and _font_ based
> on the language tag - which fall back in this example to 'en-US'.
> That font won't contain the accents necessary to render ga-Latg-IE,
> so a bunch of those annoying little black squares will replace every
> character that's not in the US-ASCII subset of the Latin script.
> Which seems serious to me, but perhaps it's just that user's
> problem, huh?

At the risk of getting further off topic:

- How common is it for print jobs not to include font-formatting
instructions but instead leave the font question entirely to the
printer? What are the scenarios?

- Why would a printer apply a font that supported only ASCII characters?
It's been my impression that network printers have fonts with character
support somewhat more inclusive than that.

This seems very odd to me. At the least at some point there is a printer
driver that should know what fonts are in the printer and how to
assign/substitute fonts. If print drivers can't figure out how to do
this for a larger range of Unicode characters than the Basic Latin
block, they're pretty broken IMO. We're at a point when content to be
printed can require complex character-to-glyph mappings; if a print
driver is still not able to handle utf-8 content just because the
content was tagged ga-Latg when there are rather bigger problems to be
solved, then it's high time the writers of print drivers came out of the
middle ages.

Peter Constable

More information about the Ietf-languages mailing list