Re: [salud] New version of the ABNF-syntax

[Paul Kyzivat <[e-mail]>]

On 2/18/13 9:24 AM, Laura Liess wrote:
> Dale,
>
> Thank you for adding the rules. I like them :-). I have following
> proposals on the two open items.
>
>> 2) date:  <date> syntax is a subset of ??? in RFC 3339.
>
> I would propose following text, which is a modification of the
> corresponding text in the rfc 4198:
>
> "<date> is a date in ISO 8601 Extended Format ([CC]YY["-"MM["-"DD]]),
> and MUST correspond to a specific day on which the organization
> allocating the URN owned the domain name specified in the provider-id.
>   If not included, the default value for MM and DD is "01". "

I've been trying to explain why this sort of wording doesn't work for 
me, several times.

The problem is that the URN is a *reference*. Typically it won't be 
written by the owner of the domain. Rather it will be written by someone 
who wants to take some special alerting action. So it isn't the 
allocation of the URN that is restricted - it is the *meaning* of that 
URN. For example, it may be that example.com was owned by a startup in 
2013, but then they go bankrupt and cease owning the domain. Then in 
2014 example.com is bought by a different company. I may have a business 
relationship with this second company, and so define some special 
alerting for calls from them, tagging my alert with "[e-mail](2014)".

Of course somebody needs to insert the alert-info header with the URN in 
the call. It *could* be the caller (example.com) in which case the 
language you have above would apply. But it could also be a server on my 
receiving side, that authorizes incoming calls and assigns alert URNs. 
This server is free to use whatever URN it wants, but the meaning will 
vary. IMO this usage is much more likely - I am not likely to trust 
Alert-Info URNs assigned by the caller. Or if I do I will most likely 
screen/filter them.

	Thanks,
	Paul