Re: Last Call: <draft-nandakumar-rtcweb-stun-uri-05.txt> (URI Scheme for Session Traversal Utilities for NAT (STUN) Protocol) to Proposed Standard

[Harald Alvestrand <[e-mail]>]

On 08/15/2013 04:20 PM, Peter Saint-Andre wrote:
> On 8/15/13 8:10 AM, Harald Alvestrand wrote:
>> On 08/15/2013 04:05 PM, Graham Klyne wrote:
>>> Harald,
>>>
>>> Briefly:
>>>
>>> 1. Thanks for the reference,
>>>
>>> and
>>>
>>> 2. I misunderstood what you meant by "This is a format for a piece of
>>> data".  In light of your clarification, I withdraw my comments 3 & 4. 
>>> Identification of the STUN service would appear to be a perfectly
>>> reasonable use.
>>>
>>> ...
>>>
>>> So the remaining issues from my questions are whether the intended
>>> highly constrained use of these services justifies allocating a URI
>>> scheme.
>>>
>>> If the community consensus is that it is of sufficient value, I might
>>> suggest an annotation to the scheme registration along the lines of:
>>>
>>> "This URI scheme is intended for use in very specific NAT traversal
>>> environments, and should not be used otherwise on the open Web or
>>> Internet."
>>>
>>> Would such a comment run contrary to your expectations for its use?
>> I would prefer to run the comment as "This scheme is intended for use in
>> specific environments that involve NAT traversal. Users of the scheme
>> need to carefully consider the security properties of the context in
>> which they are using it."
>>
>> Echoing the warning in the STUN scheme - "use this when you know what
>> you're doing only".
>>
>> Frankly, like Hadriel indicated, I have no idea whether it will be
>> useful in other contexts or not, 
> I tend to think not.
>
>> and I'm hesitant to put language that
>> seems to claim that we've evaluated all possible contexts 
> Agreed.
>
>> and say that
>> there aren't other contexts in which it can be useful.
> Too many negatives. :-) You are hesitant to say that it won't be useful
> in other contexts, or you would prefer to say that it was designed for a
> specific contexts and probably wouldn't be useful outside that context?

I'm hesitant to say that it won't be useful in other contexts - that is,
I'd prefer to say nothing about whether it will be useful elsewhere or not.


Others understand other contexts better than I do; if they come forward
(as Hadriel just did) and say "This is useful to me", I don't want the
draft to say "Sorry, but we decided you can't use it".

>
> Peter
>