<?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: Nouns and Resources</title>
	<atom:link href="http://www.subbu.org/blog/2008/04/nouns-and-resources/feed" rel="self" type="application/rss+xml" />
	<link>http://www.subbu.org/blog/2008/04/nouns-and-resources</link>
	<description></description>
	<lastBuildDate>Wed, 08 Feb 2012 05:16:45 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Subbu Allamaraju</title>
		<link>http://www.subbu.org/blog/2008/04/nouns-and-resources/comment-page-1#comment-248</link>
		<dc:creator>Subbu Allamaraju</dc:creator>
		<pubDate>Thu, 24 Apr 2008 19:15:05 +0000</pubDate>
		<guid isPermaLink="false">http://wp.subbu.org/2008/04/nouns-and-resources/#comment-248</guid>
		<description>&gt;&gt; I think the focus on MVC has been incorrectly mixed with the helpfulness of conventions. The Rails conventions for mapping the URL to controllers is very helpful in many cases, but it does not define how all applications should be designed.

Can&#039;t agree more. Action oriented frameworks and/or mindset encourage RCP-style action-oriented interfaces (http://www.subbu.org/weblogs/main/2007/12/why_is_bad_rest.html).
</description>
		<content:encoded><![CDATA[<p>>> I think the focus on MVC has been incorrectly mixed with the helpfulness of conventions. The Rails conventions for mapping the URL to controllers is very helpful in many cases, but it does not define how all applications should be designed.</p>
<p>Can&#8217;t agree more. Action oriented frameworks and/or mindset encourage RCP-style action-oriented interfaces (<a href="http://www.subbu.org/weblogs/main/2007/12/why_is_bad_rest.html" rel="nofollow">http://www.subbu.org/weblogs/main/2007/12/why_is_bad_rest.html</a>).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eric Larson</title>
		<link>http://www.subbu.org/blog/2008/04/nouns-and-resources/comment-page-1#comment-247</link>
		<dc:creator>Eric Larson</dc:creator>
		<pubDate>Wed, 23 Apr 2008 12:11:26 +0000</pubDate>
		<guid isPermaLink="false">http://wp.subbu.org/2008/04/nouns-and-resources/#comment-247</guid>
		<description>One misunderstanding of REST is the implicit association with language specific paradigms such as MVC. When you are talking about creating a RESTful application, it doesn&#039;t mean that your URLs map directly to your model objects. It means that you have constructed the interface to your application RESTfully.

A RESTful application doesn&#039;t suggest OO, functional  or any other &quot;kind&quot; of programming. You may or may not have &quot;model&quot; objects. The implementation of a RESTful service or application should not have an impact on the design of its interface.

I think the focus on MVC has been incorrectly mixed with the helpfulness of conventions. The Rails conventions for mapping the URL to controllers is very helpful in many cases, but it does not define how all applications should be designed.
</description>
		<content:encoded><![CDATA[<p>One misunderstanding of REST is the implicit association with language specific paradigms such as MVC. When you are talking about creating a RESTful application, it doesn&#8217;t mean that your URLs map directly to your model objects. It means that you have constructed the interface to your application RESTfully.</p>
<p>A RESTful application doesn&#8217;t suggest OO, functional  or any other &#8220;kind&#8221; of programming. You may or may not have &#8220;model&#8221; objects. The implementation of a RESTful service or application should not have an impact on the design of its interface.</p>
<p>I think the focus on MVC has been incorrectly mixed with the helpfulness of conventions. The Rails conventions for mapping the URL to controllers is very helpful in many cases, but it does not define how all applications should be designed.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Subbu Allamaraju</title>
		<link>http://www.subbu.org/blog/2008/04/nouns-and-resources/comment-page-1#comment-246</link>
		<dc:creator>Subbu Allamaraju</dc:creator>
		<pubDate>Sat, 05 Apr 2008 08:50:22 +0000</pubDate>
		<guid isPermaLink="false">http://wp.subbu.org/2008/04/nouns-and-resources/#comment-246</guid>
		<description>(replacing my previous reply with a more elaborate one)

Not all models are equal, but if you are building web services that are consumed over  HTTP, the chosen resource model is a key. When you have verbs in your URIs, it generally implies that the application is using HTTP verbs in ways not intended by RFC 2616. In fact, most POST-only or GET-only web services are using HTTP in incorrect ways, potentially confusing intermediaries such as caches and proxies. So,  instead of emphasizing on RPC-style methods, by focusing on resource modeling, it is not tough or time consuming to come up with a model that can take advantage of HTTP well.

If you think this is insane, and that we should all let go, please give examples or use cases.
</description>
		<content:encoded><![CDATA[<p>(replacing my previous reply with a more elaborate one)</p>
<p>Not all models are equal, but if you are building web services that are consumed over  HTTP, the chosen resource model is a key. When you have verbs in your URIs, it generally implies that the application is using HTTP verbs in ways not intended by RFC 2616. In fact, most POST-only or GET-only web services are using HTTP in incorrect ways, potentially confusing intermediaries such as caches and proxies. So,  instead of emphasizing on RPC-style methods, by focusing on resource modeling, it is not tough or time consuming to come up with a model that can take advantage of HTTP well.</p>
<p>If you think this is insane, and that we should all let go, please give examples or use cases.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Slim Amamou</title>
		<link>http://www.subbu.org/blog/2008/04/nouns-and-resources/comment-page-1#comment-245</link>
		<dc:creator>Slim Amamou</dc:creator>
		<pubDate>Sat, 05 Apr 2008 01:37:15 +0000</pubDate>
		<guid isPermaLink="false">http://wp.subbu.org/2008/04/nouns-and-resources/#comment-245</guid>
		<description>&quot;In this example, the resource is incorrectly identified&quot;

this would mean that there is a right model and wrong model. which is right, models are not equal. the problem here is that the wrong model won&#039;t work. which is wrong and means REST imposes too much constraints.

a typical real life case involves modeling fields you don&#039;t know first hand and relying on stake holder&#039;s vision and mental models. this is how almost all projects begin. in this case, having the wrong model work is a must. nobody takes 6 months to make sure the model is right before beginning to write code anymore.

I think REST will benefit from people getting it wrong and writing verbs in URLs. I think this is sane, and that you should let go.
</description>
		<content:encoded><![CDATA[<p>&#8220;In this example, the resource is incorrectly identified&#8221;</p>
<p>this would mean that there is a right model and wrong model. which is right, models are not equal. the problem here is that the wrong model won&#8217;t work. which is wrong and means REST imposes too much constraints.</p>
<p>a typical real life case involves modeling fields you don&#8217;t know first hand and relying on stake holder&#8217;s vision and mental models. this is how almost all projects begin. in this case, having the wrong model work is a must. nobody takes 6 months to make sure the model is right before beginning to write code anymore.</p>
<p>I think REST will benefit from people getting it wrong and writing verbs in URLs. I think this is sane, and that you should let go.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Srihari</title>
		<link>http://www.subbu.org/blog/2008/04/nouns-and-resources/comment-page-1#comment-244</link>
		<dc:creator>Srihari</dc:creator>
		<pubDate>Sat, 05 Apr 2008 00:57:59 +0000</pubDate>
		<guid isPermaLink="false">http://wp.subbu.org/2008/04/nouns-and-resources/#comment-244</guid>
		<description>Subbu,
Agree to your point on how you could end up finding the http verbs too limiting. We have been building some restful services over the last 3 months with Rails and Active Resource and the first complaint I got was this.
One interesting case was for user authentication. Initially login was a method on the user class and seemed like it was not restfully invokable. On some more thinking we figured out that the missing resource was actually a Session. A session gets crud-ed whenever someone uses the application. So login now became create a session, storing something in the session updating it and logout as destroying a session.
One thing that still bothers me is when you need to deal with plural resources. The pain is not in conceiving such resource but applying the verbs to them. Typically I need such resource when I am trying a batch CRUD. I read yours and others thoughts on this and would love to know how in the meantime you are handling such use cases
</description>
		<content:encoded><![CDATA[<p>Subbu,<br />
Agree to your point on how you could end up finding the http verbs too limiting. We have been building some restful services over the last 3 months with Rails and Active Resource and the first complaint I got was this.<br />
One interesting case was for user authentication. Initially login was a method on the user class and seemed like it was not restfully invokable. On some more thinking we figured out that the missing resource was actually a Session. A session gets crud-ed whenever someone uses the application. So login now became create a session, storing something in the session updating it and logout as destroying a session.<br />
One thing that still bothers me is when you need to deal with plural resources. The pain is not in conceiving such resource but applying the verbs to them. Typically I need such resource when I am trying a batch CRUD. I read yours and others thoughts on this and would love to know how in the meantime you are handling such use cases</p>
]]></content:encoded>
	</item>
</channel>
</rss>

