>
>> I agree that hiding this is wrong, but given all the consumers out
>> there, I'm worried adding this inference will cause trouble. All we
>> need is one more person to load a URI, and for them not to understand
>> these semantics and bad things happen. Of course, if you can insert
>> this layer, and document the semantics for future users, great!
>>
>
> I agree with you here. HTTP might just have to call into the MIME
> service. I will leave this question with Darin.
>
>
HTTP currently uses the MIME service in two cases: 1) before a
transaction with the server has produced headers, and 2) if the server
did not set the Content-Type header. So, even if we skip case 1) by
returning an error, we would probably still want case 2).
Darin
