[Mageia-dev] new fonts?

Mon Sep 24 20:12:53 CEST 2012

On Mon, 2012-09-24 at 18:34 +0100, brian.smith at blueyonder.co.uk wrote:
+
+> I assume any default font should have the complete UTF8 Character set?
+
+This isn't really meaningful, because of unification - fonts implement
+glyphs, not characters, and there are a great many Unicode characters
+that represent multiple glyphs, mostly as a result of CJK
+(Chinese/Chinese/Japanese/Korean) unification.
+
+So the right answer is to have a set of fonts that overall provides the
+necessary coverage for scripts and languages that Mageia's users
+actually use.
+
+> Are the Liberation Font's UTF8 Complete?
+Strictly speaking UTF-8 is an encoding, a way of representing 32-bit
+integers in as few bytes (octets) as possible while still working with
+text tools; those 32-bit integers are Unicode "codepoints", indices into
+a table of characters published by the Unicode Consortium. Currently the
+table actually only needs I think 18 bits (more than 16 and less than 24
+at any rate), but is still growing.
+
+Wikipedia says,
+[[
+The Liberation family supports only the Latin, Greek, and Cyrillic
+alphabets, leaving out many writing systems. Extension to other writing
+systems is prevented by its unique licensing terms
+]] [1]
+
+However, asking for a single typeface to have a unified design for
+Vietnamese, Hindi, Traditional Chinese, Simplified Chinese, Japanese,
+English, French, Czech, Cree, ... is a tall order. You'd also have a
+rather large font file (probably between 500M and a gigabyte).
+
+So, the best approach is to choose smaller fonts that work OK with each
+other and give the best coverage.
+
+Hope this helps. I responded to more than your question to try & help
+others see the issues too :-)
+
+Liam
+
+[1] http://en.wikipedia.org/wiki/Liberation_fonts
+
+-- 
+Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
+Pictures from old books: http://fromoldbooks.org/
+Ankh: irc.sorcery.net irc.gnome.org freenode/#xml
+
+