<?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: Data Structures for Time Records</title>
	<atom:link href="http://blogs.teamb.com/craigstuntz/2009/01/29/37964/feed/" rel="self" type="application/rss+xml" />
	<link>http://blogs.teamb.com/craigstuntz/2009/01/29/37964/</link>
	<description>C# • Entity Framework • Functional Programming • MVC • Web</description>
	<pubDate>Sat, 11 Feb 2012 15:40:09 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.2</generator>
		<item>
		<title>By: Fred Z</title>
		<link>http://blogs.teamb.com/craigstuntz/2009/01/29/37964/#comment-4498</link>
		<dc:creator>Fred Z</dc:creator>
		<pubDate>Mon, 16 Feb 2009 09:47:09 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/?p=37964#comment-4498</guid>
		<description>Some time ago I wrote a Paradox based timekeeping system for a group of lawyers and their assistants. In the end multiple redundancies were the best answer, because of human inaccuracy. My pre-posting check routines demanded answers where there were conflicts, for example if time worked did not match start and stop entries.

We also had multiple instances of a timer form, one started for each task, with the timer running only when the form for the task was on top. For a time we went completely stupid and recorded all of those time intervals separately in an effort to see how often people got interrupted and how accurately they clicked on the form when they changed tasks or estimated time. That got too complicated and the results were not good. Either accuracy of time estimate or recording discipline were not working.

The result of all of it was that I have confidence only in the order of magnitude of time reported.

Your lawyer feels the same way: next time you get a time itemized legal bill, tell your lawyer you want to review the time records and he'll probably offer you a discount on the spot.</description>
		<content:encoded><![CDATA[<p>Some time ago I wrote a Paradox based timekeeping system for a group of lawyers and their assistants. In the end multiple redundancies were the best answer, because of human inaccuracy. My pre-posting check routines demanded answers where there were conflicts, for example if time worked did not match start and stop entries.</p>
<p>We also had multiple instances of a timer form, one started for each task, with the timer running only when the form for the task was on top. For a time we went completely stupid and recorded all of those time intervals separately in an effort to see how often people got interrupted and how accurately they clicked on the form when they changed tasks or estimated time. That got too complicated and the results were not good. Either accuracy of time estimate or recording discipline were not working.</p>
<p>The result of all of it was that I have confidence only in the order of magnitude of time reported.</p>
<p>Your lawyer feels the same way: next time you get a time itemized legal bill, tell your lawyer you want to review the time records and he&#8217;ll probably offer you a discount on the spot.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Craig Stuntz</title>
		<link>http://blogs.teamb.com/craigstuntz/2009/01/29/37964/#comment-4421</link>
		<dc:creator>Craig Stuntz</dc:creator>
		<pubDate>Fri, 30 Jan 2009 05:45:03 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/?p=37964#comment-4421</guid>
		<description>Heinz Z., in addition to Anthony's excellent reply, recall that not all time records will have a StartTime or a StopTime. Some of them will only have TotalHours, as with the example data shown in the post.</description>
		<content:encoded><![CDATA[<p>Heinz Z., in addition to Anthony&#8217;s excellent reply, recall that not all time records will have a StartTime or a StopTime. Some of them will only have TotalHours, as with the example data shown in the post.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Anthony</title>
		<link>http://blogs.teamb.com/craigstuntz/2009/01/29/37964/#comment-4418</link>
		<dc:creator>Anthony</dc:creator>
		<pubDate>Fri, 30 Jan 2009 05:04:07 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/?p=37964#comment-4418</guid>
		<description>Because the date that a work activity occurs on might not necessarily be the same date that the shift itself started on.  Like pretty much everyone who works a night shift starting at 11 p.m..</description>
		<content:encoded><![CDATA[<p>Because the date that a work activity occurs on might not necessarily be the same date that the shift itself started on.  Like pretty much everyone who works a night shift starting at 11 p.m..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Heinz Z.</title>
		<link>http://blogs.teamb.com/craigstuntz/2009/01/29/37964/#comment-4414</link>
		<dc:creator>Heinz Z.</dc:creator>
		<pubDate>Fri, 30 Jan 2009 04:35:09 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.teamb.com/craigstuntz/?p=37964#comment-4414</guid>
		<description>Hallo Craig,

why you need Workday at all when StartTime and StopTime are of type timestamp?</description>
		<content:encoded><![CDATA[<p>Hallo Craig,</p>
<p>why you need Workday at all when StartTime and StopTime are of type timestamp?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

