<?xml version="1.0" encoding="UTF-8"?><!-- generator="bbPress" -->

<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
>

<channel>
<title>XCRI Forum Topic: Identifier - enforced use of URI</title>
<link>http://www.xcri.org/forum/</link>
<description>XCRI Forum Topic: Identifier - enforced use of URI</description>
<language>en</language>
<pubDate>Sat, 18 May 2013 15:53:10 +0000</pubDate>

<item>
<title>Alan Paull on "Identifier - enforced use of URI"</title>
<link>http://www.xcri.org/forum/topic.php?id=34#post-86</link>
<pubDate>Thu, 07 Feb 2008 08:56:11 +0000</pubDate>
<dc:creator>Alan Paull</dc:creator>
<guid isPermaLink="false">86@http://www.xcri.org/forum/</guid>
<description>&#60;p&#62;Thanks, Scott.  I agree with this entirely.&#60;/p&#62;
&#60;p&#62;Alan
&#60;/p&#62;</description>
</item>
<item>
<title>scottwilson on "Identifier - enforced use of URI"</title>
<link>http://www.xcri.org/forum/topic.php?id=34#post-84</link>
<pubDate>Mon, 04 Feb 2008 12:51:54 +0000</pubDate>
<dc:creator>scottwilson</dc:creator>
<guid isPermaLink="false">84@http://www.xcri.org/forum/</guid>
<description>&#60;p&#62;I think the intention of the rule is so that where an identifier is present in unqualified form the aggregator has some sense of it being a genuine GUID (such as a course URL) rather than a LUID. The guidance for those not ready for 'real URI's' may be to take their institution domain and append their local identifier to it to create a URI. Certainly this seemed to be the practice in some of the projects.&#60;/p&#62;
&#60;p&#62;Possibly this should be reworded as:&#60;/p&#62;
&#60;p&#62;- Identifiers MUST either be (1) A Uniform Resource Identifier (URI) conforming to the URL scheme as specified by IETF RFC 3986 or (2) defined using and explicitly-referenced encoding scheme.&#60;br /&#62;
- There MUST be one and only one instance of an identifier that is not defined using an explicitly-referenced encoding scheme (see (1), above). Aggregators should be capable of relying on this identifier to be globally unique. Where no URI scheme is available for the provider, it is suggested that providers develop a URI by combining their domain URL with local course identifiers to create course URIs, with the potential of migrating at a future stage to using persistent URIs for courses.&#60;br /&#62;
- A resource may have any number of identifiers defined using explicitly-referenced encoding schemes (see (2), above).
&#60;/p&#62;</description>
</item>
<item>
<title>Alan Paull on "Identifier - enforced use of URI"</title>
<link>http://www.xcri.org/forum/topic.php?id=34#post-83</link>
<pubDate>Mon, 04 Feb 2008 11:53:25 +0000</pubDate>
<dc:creator>Alan Paull</dc:creator>
<guid isPermaLink="false">83@http://www.xcri.org/forum/</guid>
<description>&#60;p&#62;Currrent guidance in XCRI-CAP 1.1 on Identifier is:&#60;/p&#62;
&#60;p&#62;&#34;It is recommended that identifiers likely to be used by aggregators should be URLs that resolve to human-readable content.&#60;/p&#62;
&#60;p&#62;Default identifier: Structures MUST NOT contain more than one identifier without an explicitly-defined encoding scheme, and this MUST be A Uniform Resource Identifier (URI) conforming to the URL scheme as specified by IETF RFC 3986.&#34;&#60;/p&#62;
&#60;p&#62;This rule means that a course couldn't have a UCAS course number and a learndirect course number without a URI.  I'm not sure that this is what we mean.  I don't think we are yet in a position to enforce the use of URIs on all XCRI-CAP structures, and I'm a bit worried that this might put people off.&#60;/p&#62;
&#60;p&#62;Instead could we recommend use of URIs without the restriction in the default identifier section?
&#60;/p&#62;</description>
</item>

</channel>
</rss>
