[Geopriv] review of draft-google-self-published-geofeeds

[Andrew Newton <[e-mail]>]

Alissa, et al,

I have reviewed the geofeeds document. Here's my feedback (which has
already been given to the document authors):

1) Section 3.2 is about using the RIR Whois system as a bootstrap for
finding the CSV file. It might be useful to mention that the IETF
WEIRDS
working group is in the process of specifying RDAP, a replacement for
Whois. Section 5.2 of draft-ietf-weirds-json-response-06 discusses the
linking system for RDAP, which simply re-uses the RFC 5988 IANA
registry. Therefore it would be useful for RDAP and other protocols if
there is an RFC 5988 link relationship registered for geofeeds.

2) Similar to the above, registration of a media type would be useful
for referencing the geofeeds. Something along the lines of
application/geofeed+csv.

3) We might want to consider offering the ability for specifying sets
of postal codes for situations where a single postal code might be too
specific for certain contexts. This might address some privacy
concerns, and may offer the ability to reduce redundant lines in the
CSV file (e.g. Where a network prefix can be found in 2 zip codes).

-andy