[gnso-rds-pdp-wg] authoritative

[Sam Lanfranco]

David & Andrew,

I can agree that the Source (or Sources) of the Data (Data of Record) is 
a secondary consideration, answerable in part by technical solutions. 
The key is that it is sourced in such a way that it is recognized as 
Data of Record.

Once there is agreement on what should be the Data of Record (the DoR 
fields) for (say) Thin Data, with (say) unconstrained access, there is 
then the question of which access window(s) [locations (SoR) or 
processes (blockchain)] provide DoR. As for "can be sure the data is 
correct", that is the validity issue and separate from the Data of 
Record and Sources of Record issues.

How good is the Data of Record? All one can be sure of is that the data 
is Data of Record is retrieved in a process that has been agreed upon.  
There is nothing in that process that can validate that the actual data 
is correct. We hope it is and strive for near perfection, but that is a 
separate issue and the responsibility primarily of whomever collects the 
data in the first place (ISP, Registrar). A different set of processes 
need to be in place to "be sure (or maximize the probability) that the 
data is correct".

Consider: I got my cash from the ATM. There is gated access since I have 
to verify that I have a right to access the cash. The machine is the 
source, but the machine cannot guarantee that none of the bills are 
counterfeit, That is the responsibility of those who stock the machine.

Sam L.

On 5/7/2017 3:26 AM, David Cake wrote:
> I agree with Andrew. We should not be specifying source of data at this point. Maybe we’ll get to it in Phase 2, but even then we may decide that the system should not specify all possible methods of sourcing data.
>
> Davud
>
>> On 2 May 2017, at 9:47 pm, Andrew Sullivan <ajs at anvilwalrusden.com> wrote:
>>
>> Hi,
>>
>> On Tue, May 02, 2017 at 08:12:51AM -0400, Sam Lanfranco wrote:
>>> * Where should DoR reside? Centralized vs. distributed Source of
>>>    Record (SoR)?
>> I'm still not sure I understand why we care about this.  The lesson of
>> DNSSEC and Tor and blockchain and every actually-scalable file
>> distribution system we've built since at least the 1990s is that you
>> don't need to care about where you get it from, _if_ you can be sure
>> that the data is correct.  One way to do that in _implementation_ is
>> to worry about the source whence you obtain the data, but that's not
>> the important question in terms of ensuring that you have the right data.
>>
>> I think worrying about the SoR is therefore specifying implementation
>> or architecture rather than specifying policy, and therefore I think
>> we shouldn't do that.
>>
>>> * Whatever it is called, DoR from SoR is "the data"
>> But we don't actually need to define it this way, as I posted
>> yesterday, and that gets us out of needing to specify the SoR.  I
>> really like your idea about the DoR, but I think the SoR is a mistake.
>>
>> Best regards,
>>
>> A
>>
>> -- 
>> Andrew Sullivan
>> ajs at anvilwalrusden.com
>> _______________________________________________
>> gnso-rds-pdp-wg mailing list
>> gnso-rds-pdp-wg at icann.org
>> https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg