<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.0.4" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Why desktop search will give way to personal information search</title>
	<link>http://dubroy.com/blog/2008/02/07/why-desktop-search-will-give-way-to-personal-information-search/</link>
	<description>on programming, usability, and design; by Patrick Dubroy</description>
	<pubDate>Wed, 20 Aug 2008 13:52:42 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.4</generator>

	<item>
		<title>by: Chris</title>
		<link>http://dubroy.com/blog/2008/02/07/why-desktop-search-will-give-way-to-personal-information-search/#comment-8663</link>
		<pubDate>Thu, 07 Feb 2008 21:06:43 +0000</pubDate>
		<guid>http://dubroy.com/blog/2008/02/07/why-desktop-search-will-give-way-to-personal-information-search/#comment-8663</guid>
					<description>&lt;p&gt;Do we not need some kind of metadata standard?  If all the web applications published information about the pieces of data in some commonly accepted format, then search wouldn't be restricted by the type of data or the location of the data.  &lt;/p&gt;

&lt;p&gt;I know it's not trivial - but it seems like the only way we'd ever get so many different types of information talking the same language.  Most of the time, I don't want to consider what kind of file or application I stored a piece of information.  &lt;/p&gt;

&lt;p&gt;As it stands, nothing really works for me.  I have too much of my life stored online, in different places.  I use Spotlight on occasion, and it works fine to an extent.  If I'm hunting for an mp3 I know the title of, or an application I've lost track of.&lt;/p&gt;

&lt;p&gt;Chris.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Do we not need some kind of metadata standard?  If all the web applications published information about the pieces of data in some commonly accepted format, then search wouldn&#8217;t be restricted by the type of data or the location of the data.  </p>
<p>I know it&#8217;s not trivial - but it seems like the only way we&#8217;d ever get so many different types of information talking the same language.  Most of the time, I don&#8217;t want to consider what kind of file or application I stored a piece of information.  </p>
<p>As it stands, nothing really works for me.  I have too much of my life stored online, in different places.  I use Spotlight on occasion, and it works fine to an extent.  If I&#8217;m hunting for an mp3 I know the title of, or an application I&#8217;ve lost track of.</p>
<p>Chris.</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: e</title>
		<link>http://dubroy.com/blog/2008/02/07/why-desktop-search-will-give-way-to-personal-information-search/#comment-8653</link>
		<pubDate>Thu, 07 Feb 2008 18:14:22 +0000</pubDate>
		<guid>http://dubroy.com/blog/2008/02/07/why-desktop-search-will-give-way-to-personal-information-search/#comment-8653</guid>
					<description>&lt;p&gt;I've tried desktop search on a Windows box (with Google search) and an Ubuntu box (with Beagle). Neither really helped me out. I don't want a generic search that wanders through all of my "stuff," because I usually know about the type of thing I care about. I search in IM conversations because I know I'm looking for something in IM. I grep for files because my hierarchically organized file system keeps similar things in the same part of the file tree - to find similar stuff, I just have to grep from a higher location in the tree. There's just too much noise in a generic search. &lt;/p&gt;

&lt;p&gt;Desktop search could be improved if we could add weights to each blob of information. Files that I open often should be considered more important than files I seldom open. Longer IM conversations should be considered more important than short conversations. &lt;/p&gt;

&lt;p&gt;The idea of an across-the-board search doesn't really interest me. I always know the medium of the stuff that I'm looking for (IM/email/source code/files). I'm more interested in seeing improvements in how I can search within the medium. Why can't I grep for images yet? If I want to find structured data (like a phone number, email address, or physical address), why can't I just search for all instances of that data type?&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>I&#8217;ve tried desktop search on a Windows box (with Google search) and an Ubuntu box (with Beagle). Neither really helped me out. I don&#8217;t want a generic search that wanders through all of my &#8220;stuff,&#8221; because I usually know about the type of thing I care about. I search in IM conversations because I know I&#8217;m looking for something in IM. I grep for files because my hierarchically organized file system keeps similar things in the same part of the file tree - to find similar stuff, I just have to grep from a higher location in the tree. There&#8217;s just too much noise in a generic search. </p>
<p>Desktop search could be improved if we could add weights to each blob of information. Files that I open often should be considered more important than files I seldom open. Longer IM conversations should be considered more important than short conversations. </p>
<p>The idea of an across-the-board search doesn&#8217;t really interest me. I always know the medium of the stuff that I&#8217;m looking for (IM/email/source code/files). I&#8217;m more interested in seeing improvements in how I can search within the medium. Why can&#8217;t I grep for images yet? If I want to find structured data (like a phone number, email address, or physical address), why can&#8217;t I just search for all instances of that data type?</p>
]]></content:encoded>
				</item>
</channel>
</rss>
