<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.3" -->
<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>Comments for HttpWatch Blog</title>
	<link>http://blog.httpwatch.com</link>
	<description>News, articles and all things HttpWatch</description>
	<pubDate>Wed, 23 Jul 2008 19:15:19 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.3</generator>
		<item>
		<title>Comment on Test Drive of the Google Hosted Ajax Libraries by Httpwatch Blog</title>
		<link>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1026</link>
		<dc:creator>Httpwatch Blog</dc:creator>
		<pubDate>Fri, 11 Jul 2008 06:01:17 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1026</guid>
		<description>Glenn,

Yes, a conditional tag is probably the best way to handle this. Google have a similar problem with Google Analytics, but they did provide SSL support.

There's no problem with different SSL certs (try https://www.httpwatch.com). The page includes this conditional code to switch protocols:

&#60;script type="text/javascript"&#62;
	var gaJsHost = (("https:" == document.location.protocol) ?
	"https://ssl." : "http://www.");
	document.write(unescape("%3Cscript src='" + gaJsHost +
	"google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
&#60;/script&#62;</description>
		<content:encoded><![CDATA[<p>Glenn,</p>
<p>Yes, a conditional tag is probably the best way to handle this. Google have a similar problem with Google Analytics, but they did provide SSL support.</p>
<p>There&#8217;s no problem with different SSL certs (try <a href="https://www.httpwatch.com" rel="nofollow">https://www.httpwatch.com</a>). The page includes this conditional code to switch protocols:</p>
<p>&lt;script type=&#8221;text/javascript&#8221;&gt;<br />
	var gaJsHost = ((&#8221;https:&#8221; == document.location.protocol) ?<br />
	&#8220;https://ssl.&#8221; : &#8220;http://www.&#8221;);<br />
	document.write(unescape(&#8221;%3Cscript src=&#8217;&#8221; + gaJsHost +<br />
	&#8220;google-analytics.com/ga.js&#8217; type=&#8217;text/javascript&#8217;%3E%3C/script%3E&#8221;));<br />
&lt;/script&gt;</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Test Drive of the Google Hosted Ajax Libraries by Breck</title>
		<link>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1024</link>
		<dc:creator>Breck</dc:creator>
		<pubDate>Fri, 11 Jul 2008 03:49:08 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1024</guid>
		<description>Thanks! That's a significant improvement. Enough to convince me to use them as well.</description>
		<content:encoded><![CDATA[<p>Thanks! That&#8217;s a significant improvement. Enough to convince me to use them as well.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Test Drive of the Google Hosted Ajax Libraries by Glenn Rempe</title>
		<link>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1023</link>
		<dc:creator>Glenn Rempe</dc:creator>
		<pubDate>Fri, 11 Jul 2008 03:47:43 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1023</guid>
		<description>Couldn't you wrap the script src tag in a conditional block so:

if connection is https
  script src local copy
else
  script src google copy
end

This would allow usage of their copy for most access, and in the hopefully limited cases where you are operating over an ssl connection, you get the copy served over the local web server.

I'm not sure how it would ever be possible to make ssl work from a google host since even if they serve it over ssl won't browsers complain on retrieving ssl content from different servers with different ssl certs?

Cheers.</description>
		<content:encoded><![CDATA[<p>Couldn&#8217;t you wrap the script src tag in a conditional block so:</p>
<p>if connection is https<br />
  script src local copy<br />
else<br />
  script src google copy<br />
end</p>
<p>This would allow usage of their copy for most access, and in the hopefully limited cases where you are operating over an ssl connection, you get the copy served over the local web server.</p>
<p>I&#8217;m not sure how it would ever be possible to make ssl work from a google host since even if they serve it over ssl won&#8217;t browsers complain on retrieving ssl content from different servers with different ssl certs?</p>
<p>Cheers.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Test Drive of the Google Hosted Ajax Libraries by Son Nguyen</title>
		<link>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1002</link>
		<dc:creator>Son Nguyen</dc:creator>
		<pubDate>Wed, 09 Jul 2008 22:57:33 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/07/08/test-drive-of-the-google-hosted-ajax-libraries/#comment-1002</guid>
		<description>Excellent analysis! With this result, we are more confident in the advertised improvement as we have just switched to the hosted jquery library for our ad server, AdSpeed.</description>
		<content:encoded><![CDATA[<p>Excellent analysis! With this result, we are more confident in the advertised improvement as we have just switched to the hosted jquery library for our ad server, AdSpeed.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on What does (Aborted) mean in HttpWatch? by lnielsen</title>
		<link>http://blog.httpwatch.com/2008/01/28/what-does-aborted-mean-in-httpwatch/#comment-1000</link>
		<dc:creator>lnielsen</dc:creator>
		<pubDate>Wed, 09 Jul 2008 19:05:22 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/01/28/what-does-aborted-mean-in-httpwatch/#comment-1000</guid>
		<description>I have also seen this when the same gif is used multiple times on a single page (like a grid of green_circle.gif). IE may download the gif several times before it realizes that it is now in the cache. All of the later requests will show (Aborted)</description>
		<content:encoded><![CDATA[<p>I have also seen this when the same gif is used multiple times on a single page (like a grid of green_circle.gif). IE may download the gif several times before it realizes that it is now in the cache. All of the later requests will show (Aborted)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on ERROR_INTERNET_INVALID_URL &#038; HttpWatch by Detlev Beutner</title>
		<link>http://blog.httpwatch.com/2007/11/20/error_internet_invalid_url-httpwatch/#comment-550</link>
		<dc:creator>Detlev Beutner</dc:creator>
		<pubDate>Wed, 25 Jun 2008 09:40:20 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2007/11/20/error_internet_invalid_url-httpwatch/#comment-550</guid>
		<description>At least on IE7 the replacement of "//:" by "about:blank" should do the trick (as far as I remember, on IE6 "about:blank" led to security warnings if called in an HTTP context). So - such JS lib providers maybe should change their implementation - the "about:blank" not only leads to no error message in httpWatch, but also generates to internal request at all, saving 1-3 millisecs.</description>
		<content:encoded><![CDATA[<p>At least on IE7 the replacement of &#8220;//:&#8221; by &#8220;about:blank&#8221; should do the trick (as far as I remember, on IE6 &#8220;about:blank&#8221; led to security warnings if called in an HTTP context). So - such JS lib providers maybe should change their implementation - the &#8220;about:blank&#8221; not only leads to no error message in httpWatch, but also generates to internal request at all, saving 1-3 millisecs.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Using HttpWatch to Measure Page Load Times for New and Existing Users by Httpwatch Blog</title>
		<link>http://blog.httpwatch.com/2008/05/14/using-httpwatch-to-measure-page-load-times-for-new-and-existing-users/#comment-383</link>
		<dc:creator>Httpwatch Blog</dc:creator>
		<pubDate>Fri, 23 May 2008 19:01:15 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/05/14/using-httpwatch-to-measure-page-load-times-for-new-and-existing-users/#comment-383</guid>
		<description>Tad,

That's a good point about having to re-authenticate or re-submit data to get to certain pages. We tried to simplify the post by just showing how you could test something like a home page.

The reason for having to restart IE is the image cache (see http://blog.httpwatch.com/2008/02/27/image-caching-in-internet-explorer/ ).

You can cause a new browser session to be created by using 'Clear All Cookies' or 'Clear Session Cookies' in HttpWatch. This has the side effect of creating a new session. If you then navigate to another location e.g. about:blank before clearing the session you should see the normal automatic cache validation when you re-visit the page.

Unfortunately, the IE image cache is not reset when a new browser session is created. So you are forced to restart IE if you want to replicate a primed cache visit and see which images are accessed by IE.</description>
		<content:encoded><![CDATA[<p>Tad,</p>
<p>That&#8217;s a good point about having to re-authenticate or re-submit data to get to certain pages. We tried to simplify the post by just showing how you could test something like a home page.</p>
<p>The reason for having to restart IE is the image cache (see <a href="http://blog.httpwatch.com/2008/02/27/image-caching-in-internet-explorer/" rel="nofollow">http://blog.httpwatch.com/2008/02/27/image-caching-in-internet-explorer/</a> ).</p>
<p>You can cause a new browser session to be created by using &#8216;Clear All Cookies&#8217; or &#8216;Clear Session Cookies&#8217; in HttpWatch. This has the side effect of creating a new session. If you then navigate to another location e.g. about:blank before clearing the session you should see the normal automatic cache validation when you re-visit the page.</p>
<p>Unfortunately, the IE image cache is not reset when a new browser session is created. So you are forced to restart IE if you want to replicate a primed cache visit and see which images are accessed by IE.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Using HttpWatch to Measure Page Load Times for New and Existing Users by tad</title>
		<link>http://blog.httpwatch.com/2008/05/14/using-httpwatch-to-measure-page-load-times-for-new-and-existing-users/#comment-381</link>
		<dc:creator>tad</dc:creator>
		<pubDate>Fri, 23 May 2008 17:29:44 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/05/14/using-httpwatch-to-measure-page-load-times-for-new-and-existing-users/#comment-381</guid>
		<description>The issue with starting a new instance of IE is that the session gets killed. You'd have to re-login, or in the case of a form, possibly resubmit data to get back to the page you want to check.
 
The reason why you need to restart the session (which is missing from the article) is due to the way IE handles its cache. If you hit refresh, or navigate to the page without closing the instance, you'd get inconsistent results. For details, check out: http://msdn2.microsoft.com/en-us/library/bb250442(VS.85).aspx

For these tests, you'd also want to make sure your cache setting is set to "Automatically." A different setting will also bring different results.</description>
		<content:encoded><![CDATA[<p>The issue with starting a new instance of IE is that the session gets killed. You&#8217;d have to re-login, or in the case of a form, possibly resubmit data to get back to the page you want to check.</p>
<p>The reason why you need to restart the session (which is missing from the article) is due to the way IE handles its cache. If you hit refresh, or navigate to the page without closing the instance, you&#8217;d get inconsistent results. For details, check out: <a href="http://msdn2.microsoft.com/en-us/library/bb250442" rel="nofollow">http://msdn2.microsoft.com/en-us/library/bb250442</a>(VS.85).aspx</p>
<p>For these tests, you&#8217;d also want to make sure your cache setting is set to &#8220;Automatically.&#8221; A different setting will also bring different results.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on What are people saying about HttpWatch? by Diego Carrera</title>
		<link>http://blog.httpwatch.com/2008/04/23/what-are-people-saying-about-httpwatch/#comment-380</link>
		<dc:creator>Diego Carrera</dc:creator>
		<pubDate>Thu, 22 May 2008 19:05:45 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/04/23/what-are-people-saying-about-httpwatch/#comment-380</guid>
		<description>This is one of the best tools out there for IE, it is nice to have some of the abilities that firebug users had been able to enjoy for some time now; in my humble opinion it is a good start, but it would be nice to be able to do much more things, like dom inspection, on the fly css properties edition and so on.

Kudos for the httpwatch team! keep the good work</description>
		<content:encoded><![CDATA[<p>This is one of the best tools out there for IE, it is nice to have some of the abilities that firebug users had been able to enjoy for some time now; in my humble opinion it is a good start, but it would be nice to be able to do much more things, like dom inspection, on the fly css properties edition and so on.</p>
<p>Kudos for the httpwatch team! keep the good work</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on What does (Aborted) mean in HttpWatch? by Crash1hd</title>
		<link>http://blog.httpwatch.com/2008/01/28/what-does-aborted-mean-in-httpwatch/#comment-375</link>
		<dc:creator>Crash1hd</dc:creator>
		<pubDate>Thu, 22 May 2008 02:52:41 +0000</pubDate>
		<guid>http://blog.httpwatch.com/2008/01/28/what-does-aborted-mean-in-httpwatch/#comment-375</guid>
		<description>I have the aborted comming up on a js file that I am trying to get to work in IE6 with a jQuery getScript call so its not all user related</description>
		<content:encoded><![CDATA[<p>I have the aborted comming up on a js file that I am trying to get to work in IE6 with a jQuery getScript call so its not all user related</p>
]]></content:encoded>
	</item>
</channel>
</rss>
