[Haskell-cafe] Re: String vs ByteString

[wren ng thornton]

Yitzchak Gale wrote:
> Bryan O'Sullivan wrote:
>> In general, Unicode uptake is increasing rapidly:
>>  http://googleblog.blogspot.com/2010/01/unicode-nearing-50-of-web.html
> 
> These Google graphs are the oft-quoted source of
> Unicode's growing dominance. But the data for those graphs
> is taken from Google's own web indexing.

Note also that all those encodings near the bottom are remaining 
relatively constant. UTF8 is taking its market share from ASCII and 
Western European encodings, not so much from other encodings (as yet).

As Bryan mentioned, Unicode doesn't have wide acceptance in CJK 
countries. These days, Japanese websites seem to have finally started to 
standardize--- in that they use HTTP/HTML headers to say which encoding 
the pages are in (and generally use JIS or Shift-JIS). This is a big 
step up from a decade ago when non-commercial sites pretty invariably 
required fiddling with the browser to get rid of mojibake. Japan hasn't 
been bitten by the i18n/l10n bug and they don't have a strong F/OSS 
community to drive adoption either.

-- 
Live well,
~wren