<?xml version="1.0" encoding="UTF-8" standalone="no" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
        <title>idkfa rss feed</title>
        <link>http://www.idkfa.com/v3</link>
        <atom:link href="http://idkfa.com/v3/rss.php" rel="self" type="application/rss+xml" />
        <description>idkfa: syndicated</description>
        <item>
           <title>Governator: LOL.  Cribbage.</title>
           <link>http://www.idkfa.com/v3/v_thread.php?thread_id=2417&amp;msg_id=2439</link>
           <description>LOL.  Cribbage.</description>
           <author>Governator@idkfa.com (Governator)</author>
           <category>idkfa</category>
           <pubDate>Wed, 06 Apr 2011 15:13:03 -0800</pubDate>
           <guid>http://idkfa.com/v3/v_thread.php?thread_id=2417&amp;msg_id=2439</guid>
       </item>
            <item>
           <title>kaiden: Well, if everyone else at work takes 2 hour</title>
           <link>http://www.idkfa.com/v3/v_thread.php?thread_id=2417&amp;msg_id=2417</link>
           <description>Well, if everyone else at work takes 2 hour lunches to play cribbage, I&#39;m entitled to work on idkfa during my lunch hour... and a half.     Implemented a caching system, sort of similar to what we had on idkfa v2, but with improvements due to some hard lessons learned previously.         On v2, I was caching HTML output, such that whenever you generated something like the &quot;Discussion Item&quot; widget, it saved the structural HTML code for that, and would return that based on a cached file. However, any time I needed to make changed to that HTML code, I would have to find a way to manually clear the cache. This meant that for every change I made, I either had to wait 5 minutes for the cache to clear, or would have to delete a file manually. By the time I was finished with v2, this was a maddening process.       Saving the HTML output of a function meant that I never had to spend time processing data again, I could just spit a file back to the user and be done with it. This, however, meant that based on whatever I could be processing, I could be caching a ton of data in order to save the full output. For idkfa v2, that meant all of my super-inefficient HTML was being cached, and sent back and forth, no matter the size of the data set that the output had been generated from.       Every function I wrote that I wanted cached I had to do on a case-by-case basis. This means I would have to copy code from one function to another each time I identified a place that could benefit from caching. Not only did this increase the length of the code written, but made it severely complex for which portions should/should not have been cached, and parameters for how to cache it (where to cache, how long to cache, etc.)       Now I&#39;m caching data sets. That is, rather than caching HTML output, I&#39;m caching the data structures returned from the database. And I have two standardized functions to do so (encache, and decache), that support all of the parameters for</description>
           <author>kaiden@idkfa.com (kaiden)</author>
           <category>idkfa</category>
           <pubDate>Fri, 01 Apr 2011 17:48:48 -0800</pubDate>
           <guid>http://idkfa.com/v3/v_thread.php?thread_id=2417&amp;msg_id=2417</guid>
       </item>
            
    </channel>
</rss> 
