Re: [tlug] Re: WWWJDIC backdoor issue

["Stephen J. Turnbull" <stephen@example.com>]

Jim Breen writes:

 > I should have said "there's nothing *workable* in the HTML
 > standards".

What's unworkable about it?  The evidence presented so far is just
that it's not widely implemented.

I suspect it's mostly a YAGNI, and that's why they didn't implement.
But if we have a workable standard, we should complain that browsers
don't conform to the standard, not that the standard is deficient.

 > When I experimented with "accept-charset", a few years back, the
 > browser(s) I tried went on sending in strings in the coding of the
 > form (e.g. Shift_JIS),

What do you mean by "coding of the form"?  The coding of the document
containing the form?  As in:

 > "Microsoft Internet Explorer 4.0 and Netscape Navigator 4.0
 > ...... both these browsers ignore the ACCEPT-CHARSET value and
 > consider the form charset the same as the document charset"

This of course the recommended default behavior if you don't know what
ACCEPT-CHARSET is.  It should work almost all the time; even the
provisions that XML has for changing character sets are very rarely
invoked.