Re: [apps-discuss] WGLC on draft-ietf-appsawg-xdash-02.txt

To add to that -- I know of one implementation that does this today (treats HTTP headers that start with X- fundamentally differently), and I need this in there to bash them over the head. Doing so is very broken.

Cheers,

On 15/12/2011, at 10:27 AM, Peter Saint-Andre wrote:

> On 12/9/11 12:03 PM, Peter Saint-Andre wrote:
>> On 12/9/11 11:54 AM, SM wrote:
>>> At 09:19 09-12-2011, Alexey Melnikov wrote:
>>>> I would like to initiate WGLC on draft-ietf-appsawg-xdash-02.txt. Due
>>>> to holiday season the WGLC is going to be a long one and will end on
>>>> January 6th. Please send any comments
>>> 
>>> Please consider the comments below as nits.
>>> 
>>> In Section 1:
>>> 
>>>  "Therefore this document deprecates the "X-" convention for most
>>>   application protocols and makes specific recommendations about
>>>   how to proceed in a world without the distinction between
>>>   standard and non-standard parameters."
>>> 
>>> In Section 2, there is a "MUST NOT" for implementers of application
>>> protocols.  The "most" (see quoted text) is not taken into
>>> consideration.  The authors could get away with a "SHOULD" instead of a
>>> "MUST".
>> 
>> Good point. Thanks for the review!
> 
> Aha, in looking at how to change the text I realize there might be a
> disconnect.
> 
> Section 2 states:
> 
>   Implementers of application protocols MUST NOT treat the general
>   categories of "standard" and "non-standard" parameters in
>   programatically different ways within their applications.
> 
> The intent of that text is to say that it's not acceptable in your code
> to search for the characters "x" and "-" at the beginning of every HTTP
> header (or whatever) and take some programatically different action
> based on the mere fact that some parameters start with those two
> characters whereas others don't. The handling of any given parameter
> needs to be based on the semantics of the parameter, not the mere
> presence of "x" and "-" at the start of the name.
> 
> Peter
> 
> -- 
> Peter Saint-Andre
> https://stpeter.im/
> 
> 
> _______________________________________________
> apps-discuss mailing list
> apps-discuss@ietf.org
> https://www.ietf.org/mailman/listinfo/apps-discuss

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