Hi John,

> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf
> Of John Kemp
> Sent: Saturday, January 09, 2010 4:43 AM

> What is the actual reasoning behind this change? I don't understand why we
> would suddenly now decide to make some whole class of implementations
> non-conforming, even if there were only few deployments?

I have come to the conclusion that arguments against requiring a secure channel 
during some portions of the protocols are irrelevant. We have been making the 
case that some environments will not be able to deploy OAuth if SSL/TLS is 
requires, but we should put that claim under the same practical scrutiny as 
other requirements we consider. I did a quick survey of existing 
implementations and every single one uses HTTPS for obtaining tokens and for 
PLAINTEXT requests.

If someone if going to write internal implementations, they are free to change 
the protocol as they see fit. In such cases interop is more influenced by the 
internal nature of the system than the specification. But I can't come up with 
a single reason why we should allow, by making it a SHOULD, to implement OAuth 
in such an obviously insecure way (in an IETF RFC).

My argument is that there is little practical difference between sending token 
secrets in the clear to not validating signatures. Once you open the door for 
someone to steal secrets, nothing else matters.

My proposed language would be along the lines of "MUST use a secure channel 
such as TLS/SSL or another mechanism providing the same protections". This 
allows not using TLS/SSL when the environment provides the same protections 
some other way.

EHL

-- 
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