Re: [urn] Request for oneM2M URN Namespace registration

[Peter Saint-Andre <[e-mail]>]

On 11/21/22 4:24 PM, Dale R. Worley wrote:
>> The namespace is now registered:
>>
>> https://www.iana.org/assignments/urn-namespaces/urn-namespaces.xhtml
> 
> Thanks, Peter, for pushing this through.  I was going to say that I
> apologize for not engaging with this proposal, but I see that very early
> I recommended fast-tracking it with no objections.
> 
> However, there is one place I think the wording could be improved.  I
> doubt the registration template can be changed easily at this date, 

Actually it's fairly straightforward to do so.

> but
> the web page
> https://wiki.onem2m.org/index.php?title=OneM2M_URN_Namespace could be
> updated to clarify the intention.  Specifically, it says
> 
>      As general rule, the NSS-restoftree is limited to printable ASCII
>      character set, unless otherwise specified by the sub-node namespace.
> 
> The NSS is part of the URN, and a URN is required to be a sequence of
> printable ASCII characters.  So this sentence, read literally, imposes
> no new constraint.
> 
> Of course, what you mean is that NSS-restoftree is (usually) limited to
> not *represent* non-ASCII characters, that is, it does not contain what
> are commonly called %-escapes, '"%" <hex> <hex>'.
> 
> The question is what is a more accurate, but still clear, statement of
> the intention.  Looking at RFC 8141, I see that it refers to RFC 2141 to
> specify the syntax, and in RFC 2141 section 2.2 are the rules:
> 
>     <NSS>         ::= 1*<URN chars>
> 
>     <URN chars>   ::= <trans> | "%" <hex> <hex>
> 
>     <trans>       ::= <upper> | <lower> | <number> | <other> | <reserved>
> 
>     <hex>         ::= <number> | "A" | "B" | "C" | "D" | "E" | "F" |
>                       "a" | "b" | "c" | "d" | "e" | "f"
> 
>     [definitions of upper, lower, number, other, and reserved omitted]
> 
> So you could accurately say that NSS-restoftree is (usually) limited to
> <trans> characters.  But nobody would understand that without a direct
> reference to RFC 2141!
> 
> Perhaps a better wording is
> 
>      As a general rule, unless otherwise specified by the sub-node
>      namespace, the NSS-restoftree does not contain %-escapes, and thus
>      represents only the printable ASCII character set.
>      
> Here, I've moved the qualifying "unless ..." earlier in the sentence to
> connect it more with "As a general rule".

We could definitely fix this in the template. Miguel, do you agree with 
Dale's proposed clarification?

Peter