Re: GET/HEAD support "MUST"

One quick comment -

On 31/01/2009, at 5:33 AM, Robert Brewer wrote:
>
> I think I'd still prefer the more-specific "405" in most cases to  
> the catch-all "404". But I can see how 404 is a valid response, and  
> might be preferred for maximizing interop with clients, some of  
> which may not gracefully deal with 405.


There's a very fine (sometimes invisible) line between accommodating  
the realities of implementation (i.e., making things work), and  
dumbing down the protocol to the lowest common denominator.

Trying to anticipate where clients *may* not handle something  
correctly errs too far on the side of the latter, I think. Without  
hard and compelling evidence, I'd be inclined to go with your  
preference.

Cheers,

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

Received on Friday, 30 January 2009 23:31:02 UTC