Re: datatype literals and lang codes

At 11:28 21/10/2002 +0300, Patrick Stickler wrote:

[...]

>Here a few that come to mind...
>
>1. Backwards compatability with existing usage.

I think we'd need a few examples to reference


>2. Consistency in the treatment of literals which promotes generic code.

That seems a bit weak.  I would have thought the potential for user 
confusion far outweighed that.


>3. M&S says that language codes are part of literals, and typed literals
>    are still literals.

That is very weak.  Typed literals are a new idea; we have flexibility there.

We don't need a lot of justification.  Just one strong reason would be better.

Brian