<p dir="ltr">I am afraid we are going in too much detail...too EARLY for that !</p>
<p dir="ltr">Best Regards<br><br></p>
<p dir="ltr">Best Regards<br>
--ff--</p>
<p dir="ltr">Mail sent from my mobile phone. Excuse for brievety.</p>
<div class="gmail_quote">Le 5 juil. 2016 16:49, "Rob Golding" <<a href="mailto:rob.golding@astutium.com">rob.golding@astutium.com</a>> a écrit :<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
registrar<br>
</blockquote>
Only in TLD that have a registry/registrar model. (And we are only<br>
concerned about gTLD in this WG, right?)<br>
</blockquote>
<br>
Yes, as a WG we're essentially only concerned with gTLDs under ICANNs remit, so they all have a Registry/Registrar model (even if the Registrar on a brand-tld could be the Registry)<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
auth-code<br>
</blockquote>
Also note that this is not baked into DNS technology or anything like<br>
that, but is only an artifact of a current registrar transfer system.<br>
Fundamentally this is a way to authenticate the registrant, which is<br>
what the registry really needs.<br>
</blockquote>
<br>
It's less about authentication of the Registrant, and more about ensuring authorisation has been gained for the Transfer.<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
status<br>
</blockquote>
What does this mean?<br>
</blockquote>
<br>
Domains are (currently) required to have "status" code(s) relating to locks, udrp etc - but on further reflection it's not absolutely necessary for the domain name to exist or function (although will be there "internally" at the Registry)<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
And for it to be of some "functional" use:<br>
Optional<br>
nameservers<br>
</blockquote>
<br>
And also optionally:<br>
DS record (or DNSKEY or other cryptographic information which can be<br>
used to produce a DS record for DNSSEC).<br>
</blockquote>
<br>
Is the most-effective Denial-Of-Service methodology for domains actually needed for it to "work" though (leaving aside how many dont work because of dnssec) ? Probably.<br>
<br>
So ...<br>
<br>
To Exist =<br>
* domain-name (what it is)<br>
* registrar (who manages it)<br>
<br>
To Function [optional] =<br>
* nameservers<br>
* ds-stuff<br>
<br>
To Manage / Needed for Internal Systems =<br>
* transfer authentication [optional dependant on registration model]<br>
* expiry date<br>
* status<br>
<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
As a final note, if there are no registrars then everything except for<br>
the expiry date can be obtained from the DNS itself.<br>
</blockquote>
<br>
In principle I agree - I'm sure we can argue the minutiae of detail about whether it needs to be somewhere else for DNS to work though<br>
<br>
Anyone else have items to add which are _required_ rather than one-or-more of wants/nice-to-have/needed-for-my-business-model etc ?<br>
<br>
<br>
Rob<br>
_______________________________________________<br>
gnso-rds-pdp-wg mailing list<br>
<a href="mailto:gnso-rds-pdp-wg@icann.org" target="_blank">gnso-rds-pdp-wg@icann.org</a><br>
<a href="https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg" rel="noreferrer" target="_blank">https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg</a><br>
</blockquote></div>