W3C home > Mailing lists > Public > public-egov-ig@w3.org > November 2011

Re: Bibliographic Resources and Citations

From: Chris Beer <chris@codex.net.au>
Date: Sat, 19 Nov 2011 19:24:56 +1100
Message-ID: <3m9bta1g2byxthd4mjqwmdm0.1321690600449@email.android.com>
To: public-egov-ig@w3.org
a) Miss me? ;)
b) Biblio resources are ALWAYS the provenence of the library community, for which there is RDAF metadata
c) Citations must always be as the author wants, publisher of the citing works want, or using Harvard et al in RDAF (in order of precedence)
d) Many Govs have thier own RDF ready (usually Dublin Core, Geonetwork or other stable standard based) descriptor set in use already
e) LOC is not an international thing, nor are Public Domain Marks

Any solution must be agnostic to polity, standard, schema etc, and reinventing the wheel is always ill advised.

Cheers

Chris Beer
Invited Expert who isn't time rich but remains dedicated

Sent from Samsung Mobile

 Gannon Dick <gannon_dick@yahoo.com> wrote: 

In addition to the E&O hint below, the LOC Subject Heading and Vocabulary pages are served in a blended and dispersed human readable form with links to different formats.  This is great for the LOC, but the different formats do not explicitly carry a Public Domain Mark which Public Sector Information is entitled to.  A different way of registering the Public Sector Information Provenance is a Compound Document (which uses multiple existing Standards, including validation, of course).  This would insure that the (embedded) dataset retains all relevant provenance information.

I made the necessary revisions to the XHTML 1.0 Strict schema to include a DCMI bibliographicCitation tag (container class).  The attributes' list serves as a second document <meta>+ section, specific to the contents of the container.  The container content is not validated, but the attributes list (provenance) is validated

<html> ...

<body>...

      <dct:bibliographicCitation
                   style="display:none;"                    // content not displayed 
                   type="text/plain; charset=utf-8" // XHTML thinks the content-type is text of some kind
                   media="application/json"            // the real content-type is JSON
                   dct:alternative="Massachusetts"                                               // the Nickname is Massachusetts
                   dct:identifier="[North America].[United States].[Massachusetts]"  // the full text identifier is ...
                   pii:pii="http://purl.org/pii/terms/misc"                                          // there is no Personally Identifiable Information here
                   xlink:type="linkedData"                // this is a collection which may be fragmented at some point
                   xlink:role="http://creativecommons.org/publicdomain/mark/1.0/"   // Public Domain Mark
                   xlink:title="http://www.loc.gov/mads/rdf/v1#Geographic"              // SKOS Top Concept
                   xlink:label="http://www.loc.gov/mads/rdf/v1#authoritativeLabel"    // RDFS Label
                   xmlns:dct="http://purl.org/dc/terms/"                                           // the namespaces besides XHTML apply only to this container
                   xmlns:xlink="http://www.w3.org/1999/xlink"
                   xmlns:pii="http://purl.org/pii/terms/">
      <![CDATA[    RDF/JSON/Chicago Style/Zotero/etc.         ]]>
...</body>
</html>


This should be useful to Governments, NGO's, Libraries smaller commercial organizations, etc. in tracing dispersed Public Sector Information back to the source.  The example documents validate, but I'm going to hold off posting the whole package in case I hear any suggestions for improvement.  There is not much point to making this a formal specification, but it will be very handy for certain *cough* SOPA *cough* problems. 


--Gannon  



----- Original Message -----
From: Gannon Dick <gannon_dick@yahoo.com>
To: eGov IG (Public) <public-egov-ig@w3.org>
Cc: "team-gld-chairs@w3.org" <team-gld-chairs@w3.org>
Sent: Tuesday, November 15, 2011 9:25 AM
Subject: Re: W3C eGovernment Interest Group: Today's Meeting Rescheduled

re: Education and Outreach

While not nearly as much fun as a LEGO breakout, the Community Directory presents a nice leveraging opportunity for E&O.  The Library of Congress Subject Headings and Vocabularies provide data in MADS/RDF.  These URL's in turn can be processed by the W3C RDF Validator to provide Triple Listings and detailed Graphical Visualizations for presentations.

For example: Albania

1. Go To <http://www.rustprivacy.org/2011/phase/gld/cd/>
2. Choose "[Europe].[Albania]" (Nickname:Open Data Albania) this opens <http://id.loc.gov/vocabulary/countries/aa.html>
3. Open the W3C Validator in a new window/tab <http://www.w3.org/RDF/Validator/uri>
4. Copy the link on the LOC Albania page, under Alternate Formats> MADS RDF/XML
5. Paste this link into the Validator "Check by URI" form and (there are several style options available) parse

--Gannon
Received on Saturday, 19 November 2011 08:29:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 19 November 2011 08:29:12 GMT