<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/1.5.2" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Smart version controlling: why diff is just not good enough</title>
	<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/</link>
	<description>Lobet den Herrn</description>
	<pubDate>Tue, 07 Feb 2012 15:58:39 +0000</pubDate>
	<generator>http://wordpress.org/?v=1.5.2</generator>

	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Gideon</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-71</link>
		<pubDate>Tue, 31 May 2005 09:49:36 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-71</guid>
					<description>Just to add: &lt;a href=&quot;http://blogs.law.harvard.edu/philg/2005/05/11#a8494&quot; title=&quot;Philip Greenspun&quot; rel=&quot;nofollow&quot;&gt;Philip Greenspun&lt;/a&gt; recently made a case that we really needed versioned file systems. 

I think it will be a great advantage when versioning is part of the file system and when applications could be written to take advantage of this ability since the application knows the semantic meaning. This would then also solve once 
and for all the file type problem.

(There is however &lt;a href=&quot;http://www.cooper.com/articles/art_digital_soup.htm&quot; rel=&quot;nofollow&quot;&gt;much more room for improvement&lt;/a&gt; in data storage and retrieval then this alone)</description>
		<content:encoded><![CDATA[	<p>Just to add: <a href="http://blogs.law.harvard.edu/philg/2005/05/11#a8494" title="Philip Greenspun" rel="nofollow">Philip Greenspun</a> recently made a case that we really needed versioned file systems. </p>
	<p>I think it will be a great advantage when versioning is part of the file system and when applications could be written to take advantage of this ability since the application knows the semantic meaning. This would then also solve once<br />
and for all the file type problem.</p>
	<p>(There is however <a href="http://www.cooper.com/articles/art_digital_soup.htm" rel="nofollow">much more room for improvement</a> in data storage and retrieval then this alone)
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Gideon</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-70</link>
		<pubDate>Tue, 31 May 2005 09:38:42 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-70</guid>
					<description>Semantic diff and merge would be a huge improvement.</description>
		<content:encoded><![CDATA[	<p>Semantic diff and merge would be a huge improvement.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Gregor J. Rothfuss</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-64</link>
		<pubDate>Sun, 22 May 2005 22:49:08 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-64</guid>
					<description>eclipse has java structure compare for this purpose: http://www.awprofessional.com/content/images/chap3_0321159640/elementLinks/03fig34.gif

they might expand this to any language where they have an AST, who knows..</description>
		<content:encoded><![CDATA[	<p>eclipse has java structure compare for this purpose: <a href='http://www.awprofessional.com/content/images/chap3_0321159640/elementLinks/03fig34.gif' rel='nofollow'>http://www.awprofessional.com/content/images/chap3_0321159640/elementLinks/03fig34.gif</a></p>
	<p>they might expand this to any language where they have an AST, who knows..
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Tjaard</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-63</link>
		<pubDate>Sun, 22 May 2005 20:10:32 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-63</guid>
					<description>Hmmm, when the various office formats will become xml (which I really hope) extracting such information should be a piece of cake. And when the printers work at the university I'll see if I can print Schubert's paper, it looks too interesting to be read from a screen ;) . I heard from &lt;a href=&quot;http://www.zefhemel.com&quot; rel=&quot;nofollow&quot;&gt;Zef&lt;/a&gt; that someone at our university (the &lt;a href=&quot;http://www.rug.nl&quot; rel=&quot;nofollow&quot;&gt;R&lt;em&gt;u&lt;/em&gt;G&lt;/a&gt;) recently got his MSc degree for investigating a similar subject, I'll ask him about it when I see him.</description>
		<content:encoded><![CDATA[	<p>Hmmm, when the various office formats will become xml (which I really hope) extracting such information should be a piece of cake. And when the printers work at the university I&#8217;ll see if I can print Schubert&#8217;s paper, it looks too interesting to be read from a screen ;) . I heard from <a href="http://www.zefhemel.com" rel="nofollow">Zef</a> that someone at our university (the <a href="http://www.rug.nl" rel="nofollow">R<em>u</em>G</a>) recently got his MSc degree for investigating a similar subject, I&#8217;ll ask him about it when I see him.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Martijn Vermaat</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-62</link>
		<pubDate>Sun, 22 May 2005 19:12:12 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-62</guid>
					<description>See also &lt;a href=&quot;http://www.logilab.org/projects/xmldiff&quot; title=&quot;XmlDiff&quot; rel=&quot;nofollow&quot;&gt;Logilab's XmlDiff&lt;/a&gt;. Recent work is &lt;a href=&quot;http://blog.drinsama.de/erich&quot; rel=&quot;nofollow&quot;&gt;Erich Schubert&lt;/a&gt;'s thesis &lt;a href=&quot;http://idefix.pms.informatik.uni-muenchen.de:8080/idefix/description?key=@id&amp;#38;value=^PA_Erich.Schubert$&amp;#38;action=show&amp;#38;selectionFunction=match&amp;#38;view=noNavigation#Navigation&quot; title=&quot;Stucture-Preserving Difference Search in Semistructured Data&quot; rel=&quot;nofollow&quot;&gt;Stucture-Preserving Difference Search in Semistructured Data&lt;/a&gt; and accompanied &lt;a href=&quot;http://ssddiff.alioth.debian.org/&quot; title=&quot;SSDDiff&quot; rel=&quot;nofollow&quot;&gt;SSDDiff implementation&lt;/a&gt;.

Another example where usual diff capabilities of source version control systems fail are Microsoft Word documents which have built-in support for change and user tracking. This is not exploited in any general purpose versioning system I know. I'm sure similar things are provided by various other data formats.</description>
		<content:encoded><![CDATA[	<p>See also <a href="http://www.logilab.org/projects/xmldiff" title="XmlDiff" rel="nofollow">Logilab&#8217;s XmlDiff</a>. Recent work is <a href="http://blog.drinsama.de/erich" rel="nofollow">Erich Schubert</a>&#8217;s thesis <a href="http://idefix.pms.informatik.uni-muenchen.de:8080/idefix/description?key=@id&amp;value=^PA_Erich.Schubert$&amp;action=show&amp;selectionFunction=match&amp;view=noNavigation#Navigation" title="Stucture-Preserving Difference Search in Semistructured Data" rel="nofollow">Stucture-Preserving Difference Search in Semistructured Data</a> and accompanied <a href="http://ssddiff.alioth.debian.org/" title="SSDDiff" rel="nofollow">SSDDiff implementation</a>.</p>
	<p>Another example where usual diff capabilities of source version control systems fail are Microsoft Word documents which have built-in support for change and user tracking. This is not exploited in any general purpose versioning system I know. I&#8217;m sure similar things are provided by various other data formats.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Turnip</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-61</link>
		<pubDate>Sun, 22 May 2005 16:12:44 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-61</guid>
					<description>&lt;a href=&quot;http://gould.cx/ted/blog/Senior_Project_Time.html&quot; rel=&quot;nofollow&quot;&gt;A (non-existent) diff tool for SVGs&lt;/a&gt;</description>
		<content:encoded><![CDATA[	<p><a href="http://gould.cx/ted/blog/Senior_Project_Time.html" rel="nofollow">A (non-existent) diff tool for SVGs</a>
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Smart version controlling: why diff is just not good enough by: Anne</title>
		<link>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-60</link>
		<pubDate>Sun, 22 May 2005 13:15:33 +0000</pubDate>
		<guid>http://tjaard.nl/2005/05/21/smart-version-controlling-why-diff-is-just-not-enough/#comment-60</guid>
					<description>In XML white space is actually quite significant. Except for attribute values, that are normalized by a conforming XML parser.</description>
		<content:encoded><![CDATA[	<p>In XML white space is actually quite significant. Except for attribute values, that are normalized by a conforming XML parser.
</p>
]]></content:encoded>
				</item>
</channel>
</rss>

