Re: [apps-discuss] draft-ietf-weirds-bootstrap-00 and our lawn -- feedback?

[Graham Klyne <[e-mail]>]

+1 to what Mark says.

[later, on seeing some of the responses to this...]

A question I have (not having read your specs, and not intending to) is this: 
are you defining a specific *service*, or are you defining a protocol that 
others may use to deliver a service over the web?

If the former, then maybe there's not so much harm here, but it's not clear to 
me why this is seen as a potential IETF standard.  I.e. you define a service, 
and you define the URIs that will be used to access it.  Fine, it's your lawn.

But if you are defining a protocol, that tells others how to design and allocate 
URIs in their own URI space, then that tramples over the lawns of other web 
participants, and constrains their implementation and deployment choices. 
That's bad.

#g
--

On 13/02/2014 01:34, Mark Nottingham wrote:
> It's this attitude that I (and apparently other Web folks here) find disturbing -- to paraphrase, "we're building a new protocol, so we don't have to worry about what we do to the Web."
>
> Because WEIRDS has opted into using HTTP and URIs, it's opted into the Web, and that means it shouldn't harm other uses of the Web.
>
> Squatting on URIs is bad practice on the Web. Requiring implementations to use certain URI patterns is bad practice on the Web. And so on.
>
> While there may be no existing RDAP servers, the Web is pre-existing (and doing pretty well). If WEIRDS doesn't want to honour its architectural constraints, that's fine -- RDAP can use or define something else.
>
> Cheers,
>
>
>
> On 12 Feb 2014, at 9:32 am, John Levine <[e-mail]> wrote:
>
>>> domain: example.com
>>>    rel: domainlookup    href-template: http://example.com/lookup/{domain}
>>>
>>> I'm very curious to hear what other APPS folks think about this
>>
>> RDAP is a new design.  There are no existing RDAP servers other than a
>> few prototypes run by people on the WEIRDS list.  There's nothing to
>> be backward compatible with.
>>
>> RDAP is intended as a replacement for WHOIS, to answer the same
>> questions that people ask now using WHOIS, e.g. information about
>> domain names, IP addresses, ASNs, and a few other things.  The
>> questions that people ask with WHOIS haven't changed materially in 20
>> years, and I see no reason to expect them to change in the future.
>>
>> Small RDAP servers will likely adapt the RDAP prototype being funded
>> by ICANN.  Large RDAP servers will be written by a handful of large
>> registries, all of which are represented on the WEIRDS list and can
>> speak for themselves, but I can't remember any of them showing notable
>> enthusiasm for templates.
>>
>> In this particular application, I really can't see any benefit to
>> templates other than the ability to be gratuitously different just to
>> prove that you can.
>>
>> R's,
>> John
>
> --
> Mark Nottingham   http://www.mnot.net/
>
>
>
> _______________________________________________
> apps-discuss mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/apps-discuss
>