<?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"
	>
<channel>
	<title>Comments on: Reinventing RSS Readers</title>
	<atom:link href="http://www.igvita.com/2006/12/21/reinventing-rss-readers/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/</link>
	<description>A goal is a dream with a deadline.</description>
	<pubDate>Sat, 30 Aug 2008 00:27:19 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: Shoob &#187; Blog Archive &#187; links for 2007-01-22</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-13359</link>
		<dc:creator>Shoob &#187; Blog Archive &#187; links for 2007-01-22</dc:creator>
		<pubDate>Mon, 22 Jan 2007 16:37:49 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-13359</guid>
		<description>[...] Reinventing RSS Readers Analyse sur la lecture des flux RSS (sur base des chiffres d&#8217;une enquête Microsoft) (tags: rss reader feed statistics blogging microsoft design) [...]</description>
		<content:encoded><![CDATA[<p>[...] Reinventing RSS Readers Analyse sur la lecture des flux RSS (sur base des chiffres d&#8217;une enquête Microsoft) (tags: rss reader feed statistics blogging microsoft design) [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Blog of Leonid Mamchenkov &#187; More on RSS : PostRank</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-12199</link>
		<dc:creator>Blog of Leonid Mamchenkov &#187; More on RSS : PostRank</dc:creator>
		<pubDate>Thu, 04 Jan 2007 12:12:03 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-12199</guid>
		<description>[...] Today I came across some excellent thinking at Ilya Grigorik&#8217;s blog - part 1 and part 2. There was also a link to RSS habbits study by Microsoft. [...]</description>
		<content:encoded><![CDATA[<p>[...] Today I came across some excellent thinking at Ilya Grigorik&#8217;s blog - part 1 and part 2. There was also a link to RSS habbits study by Microsoft. [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Garçon aka Martin Kopta &#187; Blog Archive &#187; links for 2007-01-02</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-12145</link>
		<dc:creator>Garçon aka Martin Kopta &#187; Blog Archive &#187; links for 2007-01-02</dc:creator>
		<pubDate>Tue, 02 Jan 2007 23:27:16 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-12145</guid>
		<description>[...] igvita.com » Blog Archive » Reinventing RSS Readers (tags: rss survey) [...]</description>
		<content:encoded><![CDATA[<p>[...] igvita.com » Blog Archive » Reinventing RSS Readers (tags: rss survey) [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ilya Grigorik</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-12060</link>
		<dc:creator>Ilya Grigorik</dc:creator>
		<pubDate>Mon, 01 Jan 2007 15:37:03 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-12060</guid>
		<description>@Kevin: My apologies! I corrected the quote.

@P-Air: You're right, it's definitely not an exclusive relationship. Most collaborative filtering systems adopt the exact scenario you described. My point was, even with this adaptation it is still extremely hard to develop a truly 'personalized' system that works. Even more importantly, I'm not convinced that we need to, and that is why I pointed to Google's PageRank.</description>
		<content:encoded><![CDATA[<p>@Kevin: My apologies! I corrected the quote.</p>
<p>@P-Air: You&#8217;re right, it&#8217;s definitely not an exclusive relationship. Most collaborative filtering systems adopt the exact scenario you described. My point was, even with this adaptation it is still extremely hard to develop a truly &#8216;personalized&#8217; system that works. Even more importantly, I&#8217;m not convinced that we need to, and that is why I pointed to Google&#8217;s PageRank.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: P-Air</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11931</link>
		<dc:creator>P-Air</dc:creator>
		<pubDate>Sat, 30 Dec 2006 01:38:49 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11931</guid>
		<description>In your conclusion in the last section "Rank the posts not the feeds", not sure why you feel that there's an either/or need between subjective and objective interestingness.  Perhaps simply using objective interestingness for the cold start makes a good way to get started until enough information is captured to move into the more personalized subjective interestingness model.</description>
		<content:encoded><![CDATA[<p>In your conclusion in the last section &#8220;Rank the posts not the feeds&#8221;, not sure why you feel that there&#8217;s an either/or need between subjective and objective interestingness.  Perhaps simply using objective interestingness for the cold start makes a good way to get started until enough information is captured to move into the more personalized subjective interestingness model.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kevin Burton</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11925</link>
		<dc:creator>Kevin Burton</dc:creator>
		<pubDate>Fri, 29 Dec 2006 22:54:34 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11925</guid>
		<description>One note... that quote from wired was mine but you extract is misleading and make it sound as if it's was Ryan Singel's writing...

No big deal of course.. just wanted to point that out.

"I want to solve the question of ‘I don’t have any time and I subscribe to 500 feeds. I just got off the plane. What do I need to read?’"</description>
		<content:encoded><![CDATA[<p>One note&#8230; that quote from wired was mine but you extract is misleading and make it sound as if it&#8217;s was Ryan Singel&#8217;s writing&#8230;</p>
<p>No big deal of course.. just wanted to point that out.</p>
<p>&#8220;I want to solve the question of ‘I don’t have any time and I subscribe to 500 feeds. I just got off the plane. What do I need to read?’&#8221;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: &#187; Microsoft RSS Usage Survey Results&#160;&#187; &#160;InsideMicrosoft - part of the Blog News Channel</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11848</link>
		<dc:creator>&#187; Microsoft RSS Usage Survey Results&#160;&#187; &#160;InsideMicrosoft - part of the Blog News Channel</dc:creator>
		<pubDate>Wed, 27 Dec 2006 18:54:37 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11848</guid>
		<description>[...] Igvita has made graphs of most of the other data, including &#8220;How many feeds are you subscribed to in your primary RSS reader?&#8221;, &#8220;On average, how frequently do you open your RSS reader (e.g. to read unread articles, to find an old article)?&#8221; and &#8220;What time of day do you usually use your RSS reader (check all that apply)?&#8221;. [...]</description>
		<content:encoded><![CDATA[<p>[...] Igvita has made graphs of most of the other data, including &#8220;How many feeds are you subscribed to in your primary RSS reader?&#8221;, &#8220;On average, how frequently do you open your RSS reader (e.g. to read unread articles, to find an old article)?&#8221; and &#8220;What time of day do you usually use your RSS reader (check all that apply)?&#8221;. [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Chris Saad</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11780</link>
		<dc:creator>Chris Saad</dc:creator>
		<pubDate>Tue, 26 Dec 2006 00:35:41 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11780</guid>
		<description>Manual Trackback! :)
http://www.touchstonelive.com/blog/2006/12/pagerank-of-personal-relevancy.html</description>
		<content:encoded><![CDATA[<p>Manual Trackback! :)<br />
<a href="http://www.touchstonelive.com/blog/2006/12/pagerank-of-personal-relevancy.html" rel="nofollow">http://www.touchstonelive.com/blog/2006/12/pagerank-of-personal-relevancy.html</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ilya Grigorik</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11717</link>
		<dc:creator>Ilya Grigorik</dc:creator>
		<pubDate>Sat, 23 Dec 2006 18:04:47 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11717</guid>
		<description>@Chris: Touchstone looks very interesting. I'm a big fan of the architecture you are pushing (input/output adapters). Not to mention, a couple of very interesting posts you’ve got there! You're right, our thinking is along very similar lines. I'll wait for your post before I make further comments, but expect an email from me!

@Costas: Memigo and Findory adopt a slightly different approach to this problem. Adam Kalsey's answer (&lt;a href="http://kalsey.com/2006/12/solving_feed_overload/" rel="nofollow"&gt;Solving Feed Overload&lt;/a&gt;) to my article brings this out best. I rely heavily on human-filtering. Memigo and Findory try to predict important stories over the entire 'news item' space. My approach is to let the bloggers provide the digests first. It's subtle but different. More food for thought for me...

@Adam: Really interesting article. Like I said in my comments on your blog - I'll have to follow up with a part two to adjust and clarify my stand on RSS.

@Tony: Good point. However, there is an inherent tradeoff here - either I read absolutely everything and make sure I don't miss anything, or I rely on the 'popular' vote and miss some stuff. Long story short, there is no right answer to this dilemma! But, it is possible to develop reasonable schemes to factor this in. For example, instead of trying to define an absolute ranking of each post, we can consider context specific rankings (ex: within a blog itself). This way, the ratings are normalized with respect to each source, it’s a common approach in collaborative filtering systems.

@Jess: Thanks! Not sure about freelance just yet, but I’ll keep it in mind in case all else fails. ;)

And last but not least… Happy Holidays to everyone!</description>
		<content:encoded><![CDATA[<p>@Chris: Touchstone looks very interesting. I&#8217;m a big fan of the architecture you are pushing (input/output adapters). Not to mention, a couple of very interesting posts you’ve got there! You&#8217;re right, our thinking is along very similar lines. I&#8217;ll wait for your post before I make further comments, but expect an email from me!</p>
<p>@Costas: Memigo and Findory adopt a slightly different approach to this problem. Adam Kalsey&#8217;s answer (<a href="http://kalsey.com/2006/12/solving_feed_overload/" rel="nofollow">Solving Feed Overload</a>) to my article brings this out best. I rely heavily on human-filtering. Memigo and Findory try to predict important stories over the entire &#8216;news item&#8217; space. My approach is to let the bloggers provide the digests first. It&#8217;s subtle but different. More food for thought for me&#8230;</p>
<p>@Adam: Really interesting article. Like I said in my comments on your blog - I&#8217;ll have to follow up with a part two to adjust and clarify my stand on RSS.</p>
<p>@Tony: Good point. However, there is an inherent tradeoff here - either I read absolutely everything and make sure I don&#8217;t miss anything, or I rely on the &#8216;popular&#8217; vote and miss some stuff. Long story short, there is no right answer to this dilemma! But, it is possible to develop reasonable schemes to factor this in. For example, instead of trying to define an absolute ranking of each post, we can consider context specific rankings (ex: within a blog itself). This way, the ratings are normalized with respect to each source, it’s a common approach in collaborative filtering systems.</p>
<p>@Jess: Thanks! Not sure about freelance just yet, but I’ll keep it in mind in case all else fails. ;)</p>
<p>And last but not least… Happy Holidays to everyone!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jess</title>
		<link>http://www.igvita.com/2006/12/21/reinventing-rss-readers/#comment-11715</link>
		<dc:creator>jess</dc:creator>
		<pubDate>Sat, 23 Dec 2006 15:56:58 +0000</pubDate>
		<guid isPermaLink="false">http://www.igvita.com/blog/2006/12/21/reinventing-rss-readers/#comment-11715</guid>
		<description>Ilya... you really should think about writing freelance. Everytime I read a post, it's so thoughtfully laid out like a research article!

Happy holidays buddy! And hopefully I'll see you in the New Year!</description>
		<content:encoded><![CDATA[<p>Ilya&#8230; you really should think about writing freelance. Everytime I read a post, it&#8217;s so thoughtfully laid out like a research article!</p>
<p>Happy holidays buddy! And hopefully I&#8217;ll see you in the New Year!</p>
]]></content:encoded>
	</item>
</channel>
</rss>
