Re: [RAI] Global Service Provider ID - draft-pfautz-service-provider-identifier-urn-01

["PFAUTZ, PENN L" <[e-mail]>]

Jason:
Because of the issues alluded to, of finding you've locked someone out you'd like to be in, we were inclined *not* to set up specific qualifications.
So although the focus was service providers, it clearly went beyond ITU recognized operating authorities or even US OCN-eligible parties.
I understand the concern about being too short but not the basis for 10 digits (10 billion) not being enough as the thinking was entities, not end points or individuals in general. ITADs at 2**32-1 will top out below 10B. I'm not sure there's a direct limit on PENs in terms of OID syntax but I'm guessing SNMP implementations might constrain it to the same range as ITADs.

I remain concerned about being too big because of the desire of some parties to do prefixing with them or to incorporate into domain name frames.


Penn Pfautz
AT&T Access Management
[phone]

-----Original Message-----
From: Livingood, Jason [mailto:[e-mail]]
Sent: Monday, September 26, 2011 1:32 PM
To: Worley, Dale R (Dale); Hadriel Kaplan; PFAUTZ, PENN L
Cc: [e-mail]
Subject: Re: [RAI] Global Service Provider ID - draft-pfautz-service-provider-identifier-urn-01

Along these lines, who qualifies for a SPID? Could it be a small business
with their own SIP proxy / softswitch? If so, a 10-digit-length registry
will be wholly insufficient.


Jason

On 9/26/11 11:41 AM, "Worley, Dale R (Dale)" <[e-mail]> wrote:

>> From: Hadriel Kaplan [[e-mail]]
>>
>> So I asked some people about why ITAD wasn't what they wanted to use
>> for SPIDs, and they said exactly what you said above:
>>
>> > because the existing ones were intended for other purposes and
>> > already had many registrations
>>
>> It's the latter part (that they already have many registrations) that
>> I think is really why people don't want to use ITADs, and that's what
>> worries me in a weird way... because it implies they think some new
>> SPID registry won't be populated with junk.  It will be.
>
>I think there are two overlapping flavors of this distaste.  One is
>the symbolism of membership in a registry.  There is going to be a
>visceral desire that the registry is "for service providers, but not
>limited to traditional telephony".  However, if there are enough
>identifiers that we don't risk running out, it really doesn't matter
>who or what is allowed into the registry.
>
>The other flavor is what Hadriel says, "we don't want the registry
>filled with junk".  But the reality, the *meta-design principle*, is
>that if you have a system that can actually keep the junk out, you
>will also be keeping out something you want to let in.
>
>Dale
>_______________________________________________
>RAI mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/rai