[whatwg] Requiring the Encoding Standard preferred name is too strict for no good reason

[Henri Sivonen]

In various places that deal with encoding labels, the HTML spec now
requires authors to use the name of the encoding from the Encoding
Standard, which means using the preferred name rather than an alias.

Compared to the previous reference to the IANA registry, some names
that work in all browsers but are no longer preferred names are now
errors, such as iso-8859-1 and tis-620. Making broadly-supported names
that were previously preferred names according to IANA now be errors
does not appear to provide any utility to Web authors who use
validators.

Please relax the requirement so that at least previously-preferred
names are not errors.

zcorpan suggested
(http://krijnhoetmer.nl/irc-logs/whatwg/20130325#l-920) allowing
non-preferred names for non-UTF-8 encodings. I'm not familiar with the
level of browser support for all of the non-preferred aliases, but I
could accept zcorpan's suggestion.

-- 
Henri Sivonen
hsivonen at iki.fi
http://hsivonen.iki.fi/