W3C home > Mailing lists > Public > public-egov-ig@w3.org > November 2010

Re: Administrative Geograph (was Re: Censorship?)

From: Gannon Dick <gannon_dick@yahoo.com>
Date: Tue, 9 Nov 2010 09:04:55 -0800 (PST)
Message-ID: <228897.80606.qm@web112616.mail.gq1.yahoo.com>
To: Leigh Dodds <leigh.dodds@talis.com>
Cc: Mike Norton <xsideofparadise@yahoo.com>, W3C e-Gov IG <public-egov-ig@w3.org>
Thanks.  I agree with the approach, and I like the result.

My version is more text centric and considerably less refined, but the basic architecture for a standard is there.
(Ireland) ./ireland.zip is an example of the framework.

I've said many times that Government is a Mature Industry.  This means there is no point in starting over from first principals with every application of Open Data.  IMHO, this is not the way to move the start line forward.  Instead, I advocate the three steps forward, one step back approach to avoid the re-invention of "Government" every time there is a personnel change (for whatever reason).  This is wasteful of resources (places do not vanish because a politician does not know they are there), dangerous (problems do not vanish when they cease to be news), and in general, willful ignorance is unworthy of encouragement.

GIS data, or Michael Norton's functional collection scheme, or RDF extensions will fill the void at the "admin3" folder level very nicely.  There is nothing deficient or subservient about the information, it is simply out of place.  If you try to dig down, you find yourself in Personally Identifiable Information Territory - but you can use the PII namespace (http://purl.org/pii/terms/) to back out gracefully.


--- On Tue, 11/9/10, Leigh Dodds <leigh.dodds@talis.com> wrote:

From: Leigh Dodds <leigh.dodds@talis.com>
Subject: Administrative Geograph (was Re: Censorship?)
To: "Gannon Dick" <gannon_dick@yahoo.com>
Cc: "Mike Norton" <xsideofparadise@yahoo.com>, "W3C e-Gov IG" <public-egov-ig@w3.org>
Date: Tuesday, November 9, 2010, 6:09 AM


Switched subject line...

On 8 November 2010 23:28, Gannon Dick <gannon_dick@yahoo.com> wrote:
> Your other point ...
> Latitude and Longitude are a complete coordinate system - the ordering is a continuous function.  Entity Names and Vocabulary
> Encoding Schemes form a complete set, something a bit different.  Every satellite picture is aimed at someone's back yard, but to
> conclude there is something is special (significant) about *that* back yard is a logical fallacy. When Governments report statistical
> information, they report by region, not by point.  You can enumerate a set of points of interest, but there is not much value in
> enumerating the whole coordinate system and backfilling.

My point was that there's nothing about RDF -- as a data model -- that
stops it being used to publish data about a co-ordinate system like
lat/long or another other continuous function: if I can define a URI
scheme for it, then I can make statements about it.

One value in that kind of dynamic or "algorithmic" URI scheme is that
it can provides relationships to other useful concepts. One of which
might be references to an administrative geography. This might allow
me to find, e.g. crime figures for a particular area. I can jump from
lat/long to administrative geography to government statistics.

A related example from a different domain are the time based URIs
created for data.gov.uk which are described here:




Leigh Dodds
Programme Manager, Talis Platform

Received on Tuesday, 9 November 2010 17:05:30 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:00:44 UTC