Re: [EPP-discuss] Draft EPP Specification rev. 1.0 and invitation to meeting/workshop

From: Patrik Fältström <patrik_at_frobbit.se>
Date: Fri, 26 Oct 2012 11:10:05 -0200

On 26 okt 2012, at 10:37, Jonas B. Nielsen <jonasbn_at_dk-hostmaster.dk> wrote:

> On 26/10/2012, at 14.29, Patrik Fältström <paf_at_frobbit.se> wrote:
>
>> A few comments:
>>
>> Generic comment that I tell every epp registry ;-)
>>
>> - I am disappointed that this ccTLD registry do have private extensions.
>
> No comment

Of course not!

As I said, it is a generic comment. Any extension increase the cost and complexity not only to us registrars but also to the registrants that for their domain names have to enter and give the superset of all information that all registries in all TLDs they have domains.

>> Specifics:
>>
>> - The response with tracking number for create domains must be machine parseable. What is proposed today is not good enough.
>
> Due to an implementation issue I had to roll-back to a simpler implementation, the initial proposal, which we still aim for looks some like this:
>
> <?xml version="1.0" encoding="utf-8" standalone="no"?>
> <epp xmlns="urn:ietf:params:xml:ns:epp-1.0"
> xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
> xsi:schemaLocation="urn:ietf:params:xml:ns:epp-1.0 epp-1.0.xsd">
> <response>
> <result code="1001">
> <msg>Create domain domain.dk pending. Sequence ID = 2468.</msg>
> </result>
> <resData>
> <trackingNo>1234</trackingNo>
> </resData>
> <msgQ count="2" id="ABC-1234-XY"></msgQ>
> <trID>
> <clTRID>300c8288dbe6e15e837733983143e63d</clTRID>
> <svTRID>5B1D5990-0988-11E2-A187-AFB5A63584A5</svTRID>
> </trID>
> </response>
> </epp>

Was it intentional the sequence it is 2468 and the trackingNo 1234 in your example, or was it supposed to be the same?

> We regard this issue as open, we however wanted the specification to reflect what was in the available test environment at the time of disclosure.

Good!

>> - Creation of hosts is also pending, and should use the same mechanism for poll as creation of domains
>
> That is our intention, if this is not clear in the specification draft, I will emphasize this in the next revision.

Let me emphasize that it should be "the same" mechanism first of all. Then secondly I like what you displayed above is better than what was in the draft.

>> - User type should be standardized via IETF so that same mechanism is used on all registries using similar mechanisms
>
> You have to be more clear on this.

My point is that other TLDs do have extensions for basically the same issues. It is more and more complicated to tell the registries what kind of entity the registrant is, so I would like to see harmonization.

If you are not the first to have a flag for "Individual" I want you to use the flag someone else is using. If you are the first, I want others to use the same flag as you use.

I have not myself implemented epp to enough TLDs to be able to say whether you are first or not.

>> - For DNSSEC I sure hope you accept the DS interface (not clear in this spec)
>
> That is our intention, if this is not clear in the specification draft, I will emphasize this in the next revision.

Excellent.

Let me put it this way, when I did read the document, I only saw references to the RFC. I might have missed something. The RFC give the ability to support DS, DNSKEY or both.

I want to (for many reasons) continue to use DS, so as long as you support a DS interface, I am happy. If you also support DNSKEY, you will make others happy I guess.

The important thing is that you end up being aware (which I am sure you are) that there is no agreement among registries on what interface to use.

You "just" have to be clear on what you use, and then *my* personal feeling is that DS is the right way of going.

  Patrik
Received on Fri Oct 26 2012 - 15:10:05 CEST

This archive was generated by hypermail 2.3.0 : Fri Feb 06 2015 - 11:39:05 CET