W3C home > Mailing lists > Public > public-html-comments@w3.org > April 2010

Personally Identifiable Information

From: Gannon Dick <gannon_dick@yahoo.com>
Date: Sat, 3 Apr 2010 01:04:19 -0700 (PDT)
Message-ID: <802146.42540.qm@web112617.mail.gq1.yahoo.com>
To: public-html-comments@w3.org
Had a couple of questions.

A little background - I maintain the Personally Identifiable Information (PII) namespace: http://purl.org/pii/terms/ This is a group of tokenized URL's which can be used to "change the subject", although in RDF you actually are changing the Object, or adding a new triple. Quality is defined by having thought through the hazards of disclosure before publication, redacting if necessary, marking with a "pii:rel" attribute, validation by XSD, etc..  There is no reason to sacrifice conformance for the purely selfish reason you don't want your bank account drained :)

1. I thought I saw an <author> (DC.creator) tag in the header, but I do not see that in the RDF. There are a considerable number of meta data terms which do not identify persons, <title>, for example. Agent Classes, Languages, MIME (Content-Type), etc. are likewise exempt as groups without individual identity. Agents, though, require inspection.

2. If there is an easy way to filter the personal "profile"  in the HEAD, from the general DOM item properties the task is much easier.  Will the GRDDL/Dublin Core Syntax mechanism for linking meta data schema to meta@name still work as before ?  It would seem that <link rel="schema.DC" ... would end up in the general html vocab population.  If one needs to use a CURIE that is fine, as long as it is known that a CURIE must be used.

Gannon J Dick

Received on Saturday, 3 April 2010 08:04:52 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:26:26 UTC