Re: Error handling in URIs

Anne van Kesteren scripsit:

> >If other specs accept invalid URIs, then those specs also need to define  
> >how to do that.
> 
> So why do we get LEIRIs in iri-bis?

Because there are a bunch of specs that define them either independently
or by reference to XLink, which is not a plausible place to centralize
them.  The original idea was to write a separate W3C document or RFC,
but since the IRI RFC was being revised at the time, the opportunity to
add LEIRIs to it was available.

> >> I think the problem is that currently no specification says how to  
> >>construct a URI from a bunch of Unicode characters while taking into  
> >>account that the path component always needs to be in UTF-8 and the  
> >>query component in the document encoding.

I agree that having such a specification is a Good Thing.  There is
no unique way to map a string of Unicode characters to a URI; IRIs use
one approach, but if browsers use another, that should be written down
in the IRI RFC or elsewhere.

-- 
In my last lifetime,                            John Cowan
I believed in reincarnation;                    http://www.ccil.org/~cowan
in this lifetime,                               cowan@ccil.org
I don't.  --Thiagi