On Jan 8, 2009, at 1:27 PM, Hans Granqvist wrote:

>
>> It seemed to me that the original request might be more easily (if
>> only partially?) satisfied by creating an "implementation guidelines"
>> document, to which the different library implementers might submit
>> their "gotchas" and where we could agree "best practices" (not yet
>> mandatory) such as the ones suggested by JR, and help those who are
>> implementing their own libraries.
>
>
> I'm all for pragmatism, but this is slightly disheartening.  If the  
> spec is
> tight enough, why the need for "implementation guidelines" and  
> "gotchas"
> and "best practices"?

Because the spec. details the minimum needed for interoperability, not  
"everything". If we find that indeed something which appears to be a  
best practice is something that should be mandated, then by all means  
it should affect a future version of the spec.

JR's examples were, I think, good ones - things that affect specific  
implementations but not mandatory sections of the spec.

>
>
> If we're too far from a RI, then that IMO points at the spec not being
> good enough from this point of view, and we're looking for quick and
> dirty fixes.
>
> An attempt at an RI would certainly help IETF-bound people and
> organizations, too.

I'm not disagreeing that an RI would be nice.

- johnk

>
>
> Hans
>
> >


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"OAuth" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/oauth?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to