<?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: Google has Microsoft&#8217;s flaw</title>
	<atom:link href="http://www.onlineobservations.net/google-has-microsofts-flaw/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.onlineobservations.net/google-has-microsofts-flaw/</link>
	<description>Observing the Internet</description>
	<pubDate>Thu, 04 Dec 2008 04:48:41 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5</generator>
		<item>
		<title>By: The Observer</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-71</link>
		<dc:creator>The Observer</dc:creator>
		<pubDate>Fri, 23 May 2008 06:13:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-71</guid>
		<description>@Anton, yes... but a simple loading gif doesn't mean that the page is not frozen.

My personal preference would be to have some kind of text which changes (indicating various steps, even if unreal)... I remember that good old Sim City used to have status messages such as "Reticulating Spines" when it was loading up...

Usability is where it's at, right?</description>
		<content:encoded><![CDATA[<div class="KonaBody">@Anton, yes&#8230; but a simple loading gif doesn&#8217;t mean that the page is not frozen.</p>
<p>My personal preference would be to have some kind of text which changes (indicating various steps, even if unreal)&#8230; I remember that good old Sim City used to have status messages such as &#8220;Reticulating Spines&#8221; when it was loading up&#8230;</p>
<p>Usability is where it&#8217;s at, right?</p></div>
]]></content:encoded>
	</item>
	<item>
		<title>By: Anton</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-70</link>
		<dc:creator>Anton</dc:creator>
		<pubDate>Fri, 23 May 2008 00:26:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-70</guid>
		<description>Progress bars, for that very reason, are inherently flawed.  The "behind-the-scenes" computation is not something that can be easily measured.  It's not something we should want to measure either, at least not in a continuous sense.

Several progress bars have adopted the "milestone" approach which advances the bar to a pre-specified percentage at key points in the computation.  It's not a clean architecture, but it's a workable solution.

My personal preference is for a simple loading animation (i.e. http://www.smh.com.au/images/gmaps/loading.gif).  All I'm really interested in, after all, is something to keep my attention and let me know that the page isn't frozen.

...which Google's loading bar does.</description>
		<content:encoded><![CDATA[<div class="KonaBody">Progress bars, for that very reason, are inherently flawed.  The &#8220;behind-the-scenes&#8221; computation is not something that can be easily measured.  It&#8217;s not something we should want to measure either, at least not in a continuous sense.</p>
<p>Several progress bars have adopted the &#8220;milestone&#8221; approach which advances the bar to a pre-specified percentage at key points in the computation.  It&#8217;s not a clean architecture, but it&#8217;s a workable solution.</p>
<p>My personal preference is for a simple loading animation (i.e. <a href="http://www.smh.com.au/images/gmaps/loading.gif" rel="nofollow">http://www.smh.com.au/images/gmaps/loading.gif</a>).  All I&#8217;m really interested in, after all, is something to keep my attention and let me know that the page isn&#8217;t frozen.</p>
<p>&#8230;which Google&#8217;s loading bar does.</p></div>
]]></content:encoded>
	</item>
	<item>
		<title>By: The Observer</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-69</link>
		<dc:creator>The Observer</dc:creator>
		<pubDate>Thu, 22 May 2008 17:03:22 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-69</guid>
		<description>@Oli, i totally agree that it is a huge usability improvement, but I always though of Google as very classy, and I would have 'expected' them to come up with something better.... I guess, it's one of those things....

The behavior is consistent in both FF and IE. And the bar loads much more consistently when I clear offline files (so I think it's got to do with some caching issues as well)...</description>
		<content:encoded><![CDATA[<div class="KonaBody">@Oli, i totally agree that it is a huge usability improvement, but I always though of Google as very classy, and I would have &#8216;expected&#8217; them to come up with something better&#8230;. I guess, it&#8217;s one of those things&#8230;.</p>
<p>The behavior is consistent in both FF and IE. And the bar loads much more consistently when I clear offline files (so I think it&#8217;s got to do with some caching issues as well)&#8230;</p></div>
]]></content:encoded>
	</item>
	<item>
		<title>By: Oli</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-68</link>
		<dc:creator>Oli</dc:creator>
		<pubDate>Thu, 22 May 2008 13:50:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-68</guid>
		<description>Though I agree that it is frustrating, one has to consider how this bar loads and what causes it to load.
In the case of Gmail, it probably loads a fraction each time a particular request has been completed. The way the requests are handled (AJAX) is both dependent on the server and the browser. If the last request is taking the longest, then that last bit of the bar will take longer. 
We should see the lenghts of progress bars as discrete jumps that announce that a particular action has been completed. 
Have you seen differences in loading between Firefox and Explorer? If that's the case then it could be the difference between the way Explorer's ActiveX Object manages the request.
Still, in my opinion, this sort of loading bar is better than no bar at all.</description>
		<content:encoded><![CDATA[<div class="KonaBody">Though I agree that it is frustrating, one has to consider how this bar loads and what causes it to load.<br />
In the case of Gmail, it probably loads a fraction each time a particular request has been completed. The way the requests are handled (AJAX) is both dependent on the server and the browser. If the last request is taking the longest, then that last bit of the bar will take longer.<br />
We should see the lenghts of progress bars as discrete jumps that announce that a particular action has been completed.<br />
Have you seen differences in loading between Firefox and Explorer? If that&#8217;s the case then it could be the difference between the way Explorer&#8217;s ActiveX Object manages the request.<br />
Still, in my opinion, this sort of loading bar is better than no bar at all.</div>
]]></content:encoded>
	</item>
	<item>
		<title>By: The Observer</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-67</link>
		<dc:creator>The Observer</dc:creator>
		<pubDate>Thu, 22 May 2008 11:19:30 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-67</guid>
		<description>@Effi, I understand this.... on Windows, the problem is not just with Loading progress bars, I think the one that gets most flak is file transfer progress bars (when you move or copy files)

And it is a usability issue really. Because a progress bar that zooms to the end and then gets stuck is immensely irritating to a user.

Maybe one that shows steps is better....</description>
		<content:encoded><![CDATA[<div class="KonaBody">@Effi, I understand this&#8230;. on Windows, the problem is not just with Loading progress bars, I think the one that gets most flak is file transfer progress bars (when you move or copy files)</p>
<p>And it is a usability issue really. Because a progress bar that zooms to the end and then gets stuck is immensely irritating to a user.</p>
<p>Maybe one that shows steps is better&#8230;.</p></div>
]]></content:encoded>
	</item>
	<item>
		<title>By: Effi</title>
		<link>http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-66</link>
		<dc:creator>Effi</dc:creator>
		<pubDate>Thu, 22 May 2008 09:40:24 +0000</pubDate>
		<guid isPermaLink="false">http://www.onlineobservations.net/google-has-microsofts-flaw/#comment-66</guid>
		<description>I am no lover of Microsoft - and I will ALWAYS jump at any chance to blame them for anything I can :) ...but this is NOT  Microsoft's flaw. A progress bar on any system (Mac, Unix, Linux, Windows, et al) always shows time remaining based on the loading speed at that very moment, if that loading speed were to remain constant. As loading speed does not remain constant, but rather constantly fluctuates due to a whole range of factors, the time remaining that shows in the progress bar fluctuates along with it.

The "flaw" is in thinking that the progress bar measures actual load time, rather than potential load time based on current speed.

Hope this helps.</description>
		<content:encoded><![CDATA[<div class="KonaBody">I am no lover of Microsoft - and I will ALWAYS jump at any chance to blame them for anything I can <img src='http://www.onlineobservations.net/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> &#8230;but this is NOT  Microsoft&#8217;s flaw. A progress bar on any system (Mac, Unix, Linux, Windows, et al) always shows time remaining based on the loading speed at that very moment, if that loading speed were to remain constant. As loading speed does not remain constant, but rather constantly fluctuates due to a whole range of factors, the time remaining that shows in the progress bar fluctuates along with it.</p>
<p>The &#8220;flaw&#8221; is in thinking that the progress bar measures actual load time, rather than potential load time based on current speed.</p>
<p>Hope this helps.</p></div>
]]></content:encoded>
	</item>
</channel>
</rss>
