<?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: P3PC: Collective Media</title>
	<atom:link href="http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/</link>
	<description>Essential knowledge for making your web pages faster.</description>
	<lastBuildDate>Sat, 04 Feb 2012 09:46:28 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.2.1</generator>
	<item>
		<title>By: Server Herder</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-2180</link>
		<dc:creator>Server Herder</dc:creator>
		<pubDate>Tue, 13 Jul 2010 03:41:59 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-2180</guid>
		<description>There&#039;s seems to be a paradox in the anti document.write mantra: How is this script supposed to know where its output should go? 

This script, for example, uses a common technique of finding the last script on the page and inserting before that.  This, however, fails when dynamically inserted unless it truly is the last script on the page (unlikely).  

Without being able to rely on the assumptions that only document.write can provide, is there another mechanism for finding &quot;this script element&quot; that&#039;s works when scripts are attached to the DOM in arbitrary order?</description>
		<content:encoded><![CDATA[<p>There&#8217;s seems to be a paradox in the anti document.write mantra: How is this script supposed to know where its output should go? </p>
<p>This script, for example, uses a common technique of finding the last script on the page and inserting before that.  This, however, fails when dynamically inserted unless it truly is the last script on the page (unlikely).  </p>
<p>Without being able to rely on the assumptions that only document.write can provide, is there another mechanism for finding &#8220;this script element&#8221; that&#8217;s works when scripts are attached to the DOM in arbitrary order?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Christian</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-2006</link>
		<dc:creator>Christian</dc:creator>
		<pubDate>Wed, 28 Apr 2010 22:39:48 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-2006</guid>
		<description>Using 204 for image beacons doesn&#039;t sound very clean because the onerror eventhandler is being invoked by the img tags when 204 is received.

I like to attach an onload handler to my beacons to remove them after use which wouldn&#039;t work anymore. Using an onerror handler for this purpose is ugly.</description>
		<content:encoded><![CDATA[<p>Using 204 for image beacons doesn&#8217;t sound very clean because the onerror eventhandler is being invoked by the img tags when 204 is received.</p>
<p>I like to attach an onload handler to my beacons to remove them after use which wouldn&#8217;t work anymore. Using an onerror handler for this purpose is ugly.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Thiago Cavalcanti</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1972</link>
		<dc:creator>Thiago Cavalcanti</dc:creator>
		<pubDate>Sun, 25 Apr 2010 21:46:39 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1972</guid>
		<description>I suppose it&#039;s not so bad for Collective Media since even Google AdSense doesn&#039;t fare much better on this kind of analysis.

Good to know about BuySellAds though!</description>
		<content:encoded><![CDATA[<p>I suppose it&#8217;s not so bad for Collective Media since even Google AdSense doesn&#8217;t fare much better on this kind of analysis.</p>
<p>Good to know about BuySellAds though!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sunnybear</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1967</link>
		<dc:creator>sunnybear</dc:creator>
		<pubDate>Sat, 24 Apr 2010 06:43:12 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1967</guid>
		<description>@Steve:
1) We re-define document.write only on window.onload. On this point .write can&#039;t be applied by any widget - this just breaks the page.
2) Can you please create a post with drawbacks after document.write re-defining? This will be very interesting.</description>
		<content:encoded><![CDATA[<p>@Steve:<br />
1) We re-define document.write only on window.onload. On this point .write can&#8217;t be applied by any widget &#8211; this just breaks the page.<br />
2) Can you please create a post with drawbacks after document.write re-defining? This will be very interesting.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Souders</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1964</link>
		<dc:creator>Steve Souders</dc:creator>
		<pubDate>Thu, 22 Apr 2010 02:10:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1964</guid>
		<description>@sunnybear: I&#039;ve talked to several people who override document.write. Many of the bigger sites back away from this because of compatibility bugs and constraints (eg, more than one 3rd party snippet is doing document.write). It might be an acceptable short term workaround for smaller sites, but we as an industry need a better solution (eg, &lt;a href=&quot;http://www.stevesouders.com/blog/2010/02/15/browser-performance-wishlist/&quot; rel=&quot;nofollow&quot;&gt;Frag Tag&lt;/a&gt;).</description>
		<content:encoded><![CDATA[<p>@sunnybear: I&#8217;ve talked to several people who override document.write. Many of the bigger sites back away from this because of compatibility bugs and constraints (eg, more than one 3rd party snippet is doing document.write). It might be an acceptable short term workaround for smaller sites, but we as an industry need a better solution (eg, <a href="http://www.stevesouders.com/blog/2010/02/15/browser-performance-wishlist/" rel="nofollow">Frag Tag</a>).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sunnybear</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1963</link>
		<dc:creator>sunnybear</dc:creator>
		<pubDate>Wed, 21 Apr 2010 21:06:06 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1963</guid>
		<description>Steve, we are working on delaying such kind of widgets to window.onload event (using xhtml-document-write replacement + DOM nodes juggling). I&#039;m not sure whether it&#039;s a good idea (all resources will be still loaded, but overall load time can be greater), but this lead to better UI. If idea isn&#039;t clear - I can describe it more, but we need your opinion about this method - how potentially good it can be?</description>
		<content:encoded><![CDATA[<p>Steve, we are working on delaying such kind of widgets to window.onload event (using xhtml-document-write replacement + DOM nodes juggling). I&#8217;m not sure whether it&#8217;s a good idea (all resources will be still loaded, but overall load time can be greater), but this lead to better UI. If idea isn&#8217;t clear &#8211; I can describe it more, but we need your opinion about this method &#8211; how potentially good it can be?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: kangax</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1947</link>
		<dc:creator>kangax</dc:creator>
		<pubDate>Thu, 15 Apr 2010 13:39:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1947</guid>
		<description>&gt; It’s debatable if you need to set the language (line 6) and type (line 7), but that’s minor.

&quot;language&quot; attribute is deprecated for more than 10 years now, so I wouldn&#039;t say it&#039;s debatable ;) See http://perfectionkills.com/optimizing-html/#7_script_language_javascript

`if (b == null) { }` on line 4 can be easily replaced with `if (!b)` and `i.setAttribute(&#039;type&#039;, ...)` — with `i.type = ...`. 

In line 9, there&#039;s no need to break script start tag into &quot;scr&quot; and &quot;ipt&quot; (it&#039;s end tag that matters). Similarly, language attribute serves no purpose there. And speaking of script end tag, it would be cleaner to escape &quot;&lt;/&quot; into &quot;&lt;\/&quot; instead of breaking &quot;&lt;/script&quot; into &quot;&lt;/scr&quot; + &quot;ipt&quot;.</description>
		<content:encoded><![CDATA[<p>&gt; It’s debatable if you need to set the language (line 6) and type (line 7), but that’s minor.</p>
<p>&#8220;language&#8221; attribute is deprecated for more than 10 years now, so I wouldn&#8217;t say it&#8217;s debatable ;) See <a href="http://perfectionkills.com/optimizing-html/#7_script_language_javascript" rel="nofollow">http://perfectionkills.com/optimizing-html/#7_script_language_javascript</a></p>
<p>`if (b == null) { }` on line 4 can be easily replaced with `if (!b)` and `i.setAttribute(&#8216;type&#8217;, &#8230;)` — with `i.type = &#8230;`. </p>
<p>In line 9, there&#8217;s no need to break script start tag into &#8220;scr&#8221; and &#8220;ipt&#8221; (it&#8217;s end tag that matters). Similarly, language attribute serves no purpose there. And speaking of script end tag, it would be cleaner to escape &#8220;&lt;/&quot; into &quot;&lt;\/&quot; instead of breaking &quot;&lt;/script&quot; into &quot;&lt;/scr&quot; + &quot;ipt&quot;.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gregor</title>
		<link>http://www.stevesouders.com/blog/2010/04/14/p3pc-collective-media/#comment-1945</link>
		<dc:creator>Gregor</dc:creator>
		<pubDate>Wed, 14 Apr 2010 23:33:05 +0000</pubDate>
		<guid isPermaLink="false">http://www.stevesouders.com/blog/?p=1181#comment-1945</guid>
		<description>Hey Steve, I see this post for the 5th time now in my Google reader. What&#039;s happening?

And what is for times 6 = ?</description>
		<content:encoded><![CDATA[<p>Hey Steve, I see this post for the 5th time now in my Google reader. What&#8217;s happening?</p>
<p>And what is for times 6 = ?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

