>> I assumed we _were_ talking about both (as opposed to only markup).
>> Having only markup would be a big problem, because in those cases
>> where the author wanted to use a foreign quotation as-is (with the
>> original "foreign" quotation marks and quoting style), there'd be no
>> way to transmit it. If there were only one choice, it'd have to be
>> entities only, to avoid this problem.
>
>What? One of us has misunderstood something here. What about
><q lang=fr>Fous le camp, quitte vite et plus tôt que cela
>Nos honnêtes Ardennes.</q>
>
>Since the context is obviously French, the quote will be
>enclosed in "<<" and ">>"...
But in this case, the "<q lang=fr>" is an exact synonym for "<<", isn't it?
It doesn't buy you anything. I thought the original intention of <q> was
to make documents language-independent, which is something that can't be
conveyed with entities alone. What your example says is "Substitute French
quotation marks for the <q lang=fr>." OK, so why not just type in the
French quotation marks yourself? The only advantage I can see is that if
there's some client or computer that can't display those actual characters,
it can substitute other ones. But why can't it substitute another
character for the _entity_?
>Only problem is that an editor should no well enough to add
>a language attribute when importing fragments from different
>language documents.
I assume you mean the human kind of editor. So you mean "I'm an American
keying in this French document and I don't know what the French quoting
convertions are, so I'd like the clients to take care of it for me"? That
seems more reasonable, I guess. Still, that's not the way I understood the
proposal, which (again) I thought was intended to make documents
language-independent.
Regards and bonne année.
--Dave.
-- Dave Mandl dmandl@panix.com http://www.wfmu.org/~davem