Edward Cherlin wrote about W2K, MSIE, and display:
> ... The only way I found to see this character was to change the
> font for the Latin range.
>
> This seems to mean that Windows 2000 does not know that this
> codepoint is in the Cyrillic range, and regards all characters not in
> a range it knows about to be Latin. Is that right?
>
This behavior is not just confined to W2K, it is the behavior of
the Internet Explorer on Win 9x and Win M.E., too. Older versions
of the browser allowed the user to select a font for "Unicode (UTF-8)"
encoding, now the font the user chooses for "Latin" seems to be
also the default for Unicode encoding. The user's chosen font for
"User Defined" can't be used to display UTF-8 pages at all.
I'd worked up a "rant" on my 'utf8ornot' page, which is now a bit
out of date. (For example, Netscape DOES support CSS...)
When a third party reported this behavior to Microsoft, referencing
my page http://home.att.net/~jameskass/utf8ornot.htm , the response
from Microsoft indicated that this was mostly expected behavior and
wondered how anyone could consider this a problem. The response
went on to suggest that font repertoires shouldn't be tested in
a browser at all, but should rather be tested in a font viewer.
Hmmm, but what if you're testing a font to see if it will
display appropriately in HTML? Would we really want to
determine that a specific font displays well in a font viewer
and not test it in the browser, knowing of the browser's
eccentric font selection issues?
Microsoft Internet Explorer 5.5 resolved some of the issues,
but, unfortunately some web developers must still use
"user defined" and NCR format to work around this.
Best regards,
James Kass.
This archive was generated by hypermail 2.1.2 : Fri Jul 06 2001 - 00:17:15 EDT