Re: [weirds] AD Evaluation of draft-ietf-weirds-rdap-query-14

["Hollenbeck, Scott" <[e-mail]>]

> -----Original Message-----
> From: weirds [mailto:[e-mail]] On Behalf Of Pete Resnick
> Sent: Tuesday, October 07, 2014 12:59 AM
> To: [e-mail]
> Subject: [weirds] AD Evaluation of draft-ietf-weirds-rdap-query-14
> 
> Summary: Ready for Last Call, needs a few small changes, and I have a
> few questions.
> 
> I don't think any of these issues are controversial. Addressing them
> should be quite quick, and then I'm happy to Last Call. (If the timing
> is such that you can't do a rev in short order, I can Last Call it
> as-is, but would prefer to get the changes.)

I'd much rather go to last call with all known issues addressed. I'll rev the doc.

> 2 - A reference to using-http should probably appear pretty close to
> the
> top of this section.

OK.

>     The patterns described in this document purposefully do not
> encompass
>     all of the methods employed in the WHOIS and RESTful web services
> of
>     all of the RIRs and DNRs.
> 
> and
> 
>     It is also envisioned that each
>     registry will continue to maintain WHOIS and/or RESTful web
> services
> 
> Don't you mean "other web services", not "RESTful web services", in
> each
> of these places? RDAP-over-HTTP is the RESTful web service in this
> context, isn't it?

I think that's true, but I vaguely recall Andy saying that ARIN had something else going on, too. Maybe he can chime in to confirm. "other web services" seems safe no matter what.

> Suggest changing "JSON [RFC7159] result formatting and processing" to
> "The specification of these entities". The current wording is a bit
> awkward.

OK.

> 3 - Suggest changing "For entities: Retrieve a base URL for the
> service"
> to "For entities, a base URL is retrieved for the service" and changing
> "For help: Retrieve a base URL for any service" to "For help, a base
> URL
> for any service".

OK.

> 3.1.1 - I think you can s/[RFC1166]/dotted-decimal here.

OK.

> 3.1.3/3.1.4 -
> 
>     Internationalized domain names represented in
>     either A-label or U-label format [RFC5890] are also valid domain
>     names.
> 
> Please add on to the end, "See section 4 for information on character
> encoding for the U-label format."

OK.

> 4 - I think you should swap sections 4.1 and 4.2. The reference to
> section 4 from 3.1.3, 3.1.4, 3.2.1, and 3.2.2 should lead the reader to
> the encoding stuff right away. I think it's more useful to have 4.2
> appear first. Or you could move all (or most) of 4.2 into section 6 and
> then refer to it there.

I'll move 4.2 to Section 6 (where it will become 6.1) and refer to it appropriately.

> 4.2 - You may want to add to the second paragraph:
> 
>     UTF-8 encoded strings are then appropriately percent-encoded
>     [RFC3986] in the query URL.
> 
> and then maybe start the third paragraph with "After parsing any
> percent-encoding, ". Maybe check with some URLy people about whether I
> got the wording right.

OK.

> 6. I don't really understand the normative reference to 4290. Why is
> this normative?

It shouldn't be since we're only referencing it for descriptive purposes. I'll make it an Informative reference.

Scott