<?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"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: N:1 Web Service Applications &#8211; Getting from V1 to V2</title>
	<atom:link href="http://www.goland.org/webservicesn1versioning/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.goland.org/webservicesn1versioning/</link>
	<description>Technology, Politics, Food, Finance, etc.</description>
	<lastBuildDate>Mon, 23 Jan 2012 17:37:17 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Yaron</title>
		<link>http://www.goland.org/webservicesn1versioning/comment-page-1/#comment-827</link>
		<dc:creator>Yaron</dc:creator>
		<pubDate>Wed, 28 Jul 2004 09:47:40 +0000</pubDate>
		<guid isPermaLink="false">#comment-827</guid>
		<description>&lt;br/&gt;Please, feel free to reference it. &lt;br/&gt;&lt;br/&gt;TDD - I think TDD is interesting but I suspect it won&#039;t prove enough for backwards compatibility. The issue I point out in the article is that the test matrix for backwards compatibility is enormous. So in a sense I&#039;m assuming that the developers are doing something like TDD but the number of tests needed to capture all the edge case behaviors of the clients as they interact with the service for V1 are just too many to be useful enumerated. In most cases, where dropping V1 clients isn&#039;t an option, just leaving the V1 code alone is probably the most productive option.&lt;br/&gt;&lt;br/&gt;Facade? Proxy? Intermediary? - I used the term intermediary because it has a nice web service sound. It really was a rather arbitrary choice.</description>
		<content:encoded><![CDATA[<p>Please, feel free to reference it. </p>
<p>TDD &#8211; I think TDD is interesting but I suspect it won&#39;t prove enough for backwards compatibility. The issue I point out in the article is that the test matrix for backwards compatibility is enormous. So in a sense I&#39;m assuming that the developers are doing something like TDD but the number of tests needed to capture all the edge case behaviors of the clients as they interact with the service for V1 are just too many to be useful enumerated. In most cases, where dropping V1 clients isn&#39;t an option, just leaving the V1 code alone is probably the most productive option.</p>
<p>Facade? Proxy? Intermediary? &#8211; I used the term intermediary because it has a nice web service sound. It really was a rather arbitrary choice.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Paul Downey</title>
		<link>http://www.goland.org/webservicesn1versioning/comment-page-1/#comment-828</link>
		<dc:creator>Paul Downey</dc:creator>
		<pubDate>Wed, 28 Jul 2004 07:04:52 +0000</pubDate>
		<guid isPermaLink="false">#comment-828</guid>
		<description>&lt;br/&gt;&lt;br/&gt;Yaron, that&#039;s an interesting walk through. i&#039;d like to reference this in a paper i&#039;m writing on practical versioning if that&#039;s OK.&lt;br/&gt;&lt;br/&gt;Test Driven Development is the answer to removing the fear of rewriting - i guess you may have already seen Tim Bray&#039;s TDD sermon: &lt;br/&gt;http://www.tbray.org/ongoing/When/200x/2004/06/13/TDD2004&lt;br/&gt;&lt;br/&gt;i wonder if the term &#039;facade&#039; or &#039;proxy&#039; is better than &#039;intermediary&#039; here since you&#039;re describing an agent invisible to the sender.&lt;br/&gt;&lt;br/&gt;as you point out, branching is the killer, and there is no silver bullet here..</description>
		<content:encoded><![CDATA[<p>Yaron, that&#39;s an interesting walk through. i&#39;d like to reference this in a paper i&#39;m writing on practical versioning if that&#39;s OK.</p>
<p>Test Driven Development is the answer to removing the fear of rewriting &#8211; i guess you may have already seen Tim Bray&#39;s TDD sermon: <br /><a href="http://www.tbray.org/ongoing/When/200x/2004/06/13/TDD2004" rel="nofollow">http://www.tbray.org/ongoing/When/200x/2004/06/13/TDD2004</a></p>
<p>i wonder if the term &#39;facade&#39; or &#39;proxy&#39; is better than &#39;intermediary&#39; here since you&#39;re describing an agent invisible to the sender.</p>
<p>as you point out, branching is the killer, and there is no silver bullet here..</p>
]]></content:encoded>
	</item>
</channel>
</rss>

