W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > March 2010

ISSUE-1: Default profile document

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Thu, 11 Mar 2010 13:30:01 -0500
Message-ID: <4B9936A9.9090001@digitalbazaar.com>
To: RDFa WG <public-rdfa-wg@w3.org>
"The default profile document URL could be specified in the RDFa Core
spec. This document would outline what prefixes and tokens are pre-defined."

There are three questions here:

1) Should we specify a default profile document URL in RDFa Core 1.1 or
   let the host languages define a default profile document URL?
2) How does the RDFa processor react when it sees a @profile that isn't
   the default profile document URL.
3) Do we allow the default profile mappings to be cached? Do we allow
   them to be hard-coded?

The first question deals with whether or not one default RDFa profile
document would apply to XHTML, HTML, SVG, and ODF. Defining it in one
place (RDFa Core 1.1) makes it simpler. However, HTML may want to have
FOAF as a pre-defined prefix, but SVG may not want to have FOAF as a
pre-defined prefix. Similarly, it may not make sense to have "index",
"prev", and "next" as a part of the tokens/keywords allowed in SVG.

Another concern related to the first question is whether or not SVG and
RDFa documents will always be processed in the same way. As much as we
want this to stay true going forward, there may be a compelling use case
to process something /extra/ in SVG+RDFa that we don't process in
HTML+RDFa. If this is true, one may want to use a different @profile
document, say marked up in SVG+RDFa.

The second question concerns what the RDFa processor does when it
initializes with a default RDFa profile document and meets a @profile
declaration. Does it a) Layer the new profile on top of the mappings
that already exist from the default RDFa profile b) Ignore the default
RDFa profile mappings and use the mappings from the new profile instead?

The third question is a bit easier to answer and has to do with the
amount of caching/pre-definition that we allow via the RDFa spec. We may
want to consider allowing implementations to cache and be hard coded for
vocabularies that are known to not change often.

-- manu

Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: PaySwarming Goes Open Source
Received on Thursday, 11 March 2010 18:30:29 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:05:17 UTC