> Em 22 de set de 2016, à(s) 17:39:000, Andrew Newton <a...@hxr.us> escreveu:
> On Thu, Sep 22, 2016 at 3:52 PM, John Levine <jo...@taugh.com> wrote:
>> In article <83c4fe76-52e8-48ed-b6c2-0691555d3...@icann.org> you write:
>>> I think there are two parts to your question. First, the Registry
>>> Stakeholder Group (RySG) request for reconsideration that triggered this
>>> perhaps someone else in this list (from the RySG) can speak to that.
>> I took a look at the request and I read it as saying that RDAP would
>> cost money, and if it turns out that the final RDAP spec is different,
>> they'll have to spend even more money. I didn't see any compelling
>> technical issues. Needless to say, registries deeply resent anything
>> that requires them to spend money, particularly the new gTLDs whose
>> business plans are imploding.
> Here's the thing I found skewed:
> "Furthermore, some of our members are on record stating that since the
> RDAP profile replicates the known deficiencies of WHOIS – which is
> currently being studied by a PDP WG – so it’s not commercially
> feasible to deploy it to mimic a flawed system."
> Where it seems to me they define "not commercially feasible" as any
> technology that has flaws. I guess we should all quit using the
> Internet, throw-away our cell phones, and quit driving cars.
> Snark aside, this argument about commercial feasibility falls flat
> considering all 5 RIRs and several ccTLDs have implemented it. It's
> simply not a credible argument.
There are two angles to commercial feasibility: one is whether implementing
RDAP is feasible; RDAP is a generic protocol that can be tailored very
differently by each use case, and being from one of the ccTLDs that implemented
it (see https://rdap.registro.br/domain/nic.br) and from one of NIR/RIRs that
implemented it (see https://rdap.registro.br/ip/188.8.131.52), I can tell you
that they are very different. The other angle is whether implementing RDAP with
the ICANN RDAP profile for gTLDs is commercially feasible; the later is the one
being challenged, not the former.
regext mailing list