<?xml version="1.0" encoding="utf-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	>
<channel>
	<title>Comments on: Metadata</title>
	<atom:link href="http://blog.verselogic.net/archives/2004/metadata/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.verselogic.net/archives/2004/metadata/</link>
	<description>The personal blog of Alan J Castonguay.</description>
	<pubDate>Mon, 06 Oct 2008 16:11:24 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.7-hemorrhage</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: _Quinn</title>
		<link>http://blog.verselogic.net/archives/2004/metadata/#comment-10</link>
		<dc:creator>_Quinn</dc:creator>
		<pubDate>Tue, 21 Sep 2004 04:53:31 +0000</pubDate>
		<guid isPermaLink="false">http://blog.b4k4.ath.cx/archives/2004/09/06/metadata/#comment-10</guid>
		<description>You do get a bunch of metadata for free; time and date is a big one.  Title(s) are free (or almost so); even an untitled piece could profitably be referred to by its first line.  One aspect of metadata you don't normally think about in computing but most hand-writing authors use when flipping through archives is _shape_.   The combination of shape and date can be really powerful for browsing [through large search result sets]: I can dig up the reference for you later, but displaying timelines/calendars of documents with thumbnails of them (their first page) has proven to be very effective for finding documents.  (You may want to make them selectable for a 'search selected documents' search widget.)

One idea I'd like to look into more is deriving metadata from searches and search results.  Something along the lines of 'an unsuccesful search on "Kimiko" followed by a succesful search on "brightly cute" might mean that &lt;i&gt;Insert Title Here&lt;/i&gt; might have something to do with Kimiko.</description>
		<content:encoded><![CDATA[<p>You do get a bunch of metadata for free; time and date is a big one.  Title(s) are free (or almost so); even an untitled piece could profitably be referred to by its first line.  One aspect of metadata you don&#8217;t normally think about in computing but most hand-writing authors use when flipping through archives is _shape_.   The combination of shape and date can be really powerful for browsing [through large search result sets]: I can dig up the reference for you later, but displaying timelines/calendars of documents with thumbnails of them (their first page) has proven to be very effective for finding documents.  (You may want to make them selectable for a &#8217;search selected documents&#8217; search widget.)</p>
<p>One idea I&#8217;d like to look into more is deriving metadata from searches and search results.  Something along the lines of &#8216;an unsuccesful search on &#8220;Kimiko&#8221; followed by a succesful search on &#8220;brightly cute&#8221; might mean that <i>Insert Title Here</i> might have something to do with Kimiko.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
