David Powell wrote:
James M Snell writes:
Well answered! Thanks for the patience, you've convinced me.
Good to hear.
Constructive criticism now - could you explain this:
Implementers are reminded that unexpected results could occur when
using both the "dir" attribute and the Unicode control characters
within a single document.
Isn't using both the expected, and only, way to mark-up bidi text when
the Unicode heuristics won't do a good enough job? It seems a bit odd
to warn against using the feature as intended?
No, both should not be used.
Third, neither Unicode or the W3C recommend using the Unicode bidi
formatting codes are not recommended in markup, for a variety of
reasons.
True, but I'm not sure I'd describe a plain text element content or
attribute value as markup. Isn't markup something that could
alternatively contain tags, and be "marked-up"? Given that these
locations can't, the Unicode bidi codes will need to be used
regardless, so I'm not sure whether those recommendations apply here?
The markup in question here are the Atom elements. In other words, the
bidi formatting characters should not be used in XML, period, not just
within specific kinds of elements.
Fourth, even with the Unicode formatting codes, there are existing
deployed feed readers that get it wrong.
And will continue to.
Yep.
- James
