Re: PROPOSAL: i74: Encoding for non-ASCII headers

Yes, that's one path we can take, but we need to make that decision.

Separate from that, we need to determine whether omitting the C1  
controls when using iso-8859-1 was an oversight, or purposeful;  
there's more than once character in that set.

Cheers,


On 04/04/2008, at 8:34 AM, Jamie Lokier wrote:

> Mark Nottingham wrote:
>>
>> Julian -- sorry, I mean to say that we need to consider excluding C1.
>>
>> Jamie -- if we later decided to allow UTF-8, we'd of course have to
>> figure out how it fit into the overall picture. This sub-issue is  
>> just
>> about whether we should allow C1 in iso-8859-1, as the spec is
>> currently written.
>
> I see your point, but I think the reality at the moment is it's just
> permission to transport high valued octets, with %x20-7F being ASCII.
> I don't think anyone seriously treats the high values as iso-8859-1
> (except by accident).
>
> I'm thinking the current spec, plus current practice (assumed, not
> thoroughly investigated), has a nice loophole to move it to UTF-8
> without breaking anything.
>
> The key thing is "without breaking anything".  Even if there are a few
> implementations which treat the upper octets as iso-8859-1, they will
> continue working if we start sending UTF-8.
>
> -- Jamie


--
Mark Nottingham     http://www.mnot.net/

Received on Thursday, 3 April 2008 22:59:08 UTC