<?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: More on BPMN-to-BPEL</title>
	<atom:link href="http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/</link>
	<description>Bruce Silver's blog on business process management</description>
	<pubDate>Thu, 03 Jul 2008 22:32:54 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: Derek Miers</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4828</link>
		<dc:creator>Derek Miers</dc:creator>
		<pubDate>Fri, 11 Apr 2008 09:58:55 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4828</guid>
		<description>Thanks Bruce - an excellent post (you see I do read your stuff ... just dont know where you get the time to create the postings).

Anyway - I think you nail the key point - "This is why BPEL will, in my view, remain a poor runtime partner for BPMN, at least for BPMN 1.x ... There is a chance the BPEL vendors now driving the BPMN 2.0 bus will force that standard to conform to the constraints of BPEL.  But unless and until that happens, a BPMN “engine” remains a much better solution." 

BPMN 2.0 is still a wait and see ... as of the last OMG meeting, the two groups are allegedly talking to each other to see how this will pan out. Personally, I am trying to stay out of this fight.</description>
		<content:encoded><![CDATA[<p>Thanks Bruce - an excellent post (you see I do read your stuff &#8230; just dont know where you get the time to create the postings).</p>
<p>Anyway - I think you nail the key point - &#8220;This is why BPEL will, in my view, remain a poor runtime partner for BPMN, at least for BPMN 1.x &#8230; There is a chance the BPEL vendors now driving the BPMN 2.0 bus will force that standard to conform to the constraints of BPEL.  But unless and until that happens, a BPMN “engine” remains a much better solution.&#8221; </p>
<p>BPMN 2.0 is still a wait and see &#8230; as of the last OMG meeting, the two groups are allegedly talking to each other to see how this will pan out. Personally, I am trying to stay out of this fight.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Arsalan Minhas</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4799</link>
		<dc:creator>Arsalan Minhas</dc:creator>
		<pubDate>Sat, 05 Apr 2008 20:57:16 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4799</guid>
		<description>Bruce, you are absolutely correct. Mapping BPMN to block-oriented BPEL is an issue and future revisions in the standards have to look into all such incompatibilities. Moreover you cannot force a business analyst to keep in mind all the strategies and patterns while drawing BPMN diagrams so that they are also BPEL-friendly. The user is least interested in all such executable details and wants to focus only on the business process in hand. Therefore for the moment proprietary runtime engines for e.g. from Lombardi, Appian, Cordys (BPML based) seem to make more sense as they at least allow both business users and IT to work hand in hand which was the prime objective of BPM/SOA wave. Even some of the vendors for e.g. Cordys allows importing/exporting of models in BPEL and XPDL for interoperability. This sounds more reasonable than working on immature standards and get stuck by roundtripping and business-IT handoff. Of course in the long-term one would like to see standards addressing this key issue and tools adhereing to these standards.</description>
		<content:encoded><![CDATA[<p>Bruce, you are absolutely correct. Mapping BPMN to block-oriented BPEL is an issue and future revisions in the standards have to look into all such incompatibilities. Moreover you cannot force a business analyst to keep in mind all the strategies and patterns while drawing BPMN diagrams so that they are also BPEL-friendly. The user is least interested in all such executable details and wants to focus only on the business process in hand. Therefore for the moment proprietary runtime engines for e.g. from Lombardi, Appian, Cordys (BPML based) seem to make more sense as they at least allow both business users and IT to work hand in hand which was the prime objective of BPM/SOA wave. Even some of the vendors for e.g. Cordys allows importing/exporting of models in BPEL and XPDL for interoperability. This sounds more reasonable than working on immature standards and get stuck by roundtripping and business-IT handoff. Of course in the long-term one would like to see standards addressing this key issue and tools adhereing to these standards.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: bruce</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4798</link>
		<dc:creator>bruce</dc:creator>
		<pubDate>Sat, 05 Apr 2008 20:06:51 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4798</guid>
		<description>Sebastian,
Your comment points out I forgot an important clarification.  When I say BPMN engine I am talking only about automating the orchestration.  The implementation of the tasks themselves (service and user) will and should remain technology-specific.  The process "model" does not deal with task internals, only the orchestration.  If you accept this, I think it eliminates the distinction between your horizontal and vertical.  Or at least blurs it.
-Bruce</description>
		<content:encoded><![CDATA[<p>Sebastian,<br />
Your comment points out I forgot an important clarification.  When I say BPMN engine I am talking only about automating the orchestration.  The implementation of the tasks themselves (service and user) will and should remain technology-specific.  The process &#8220;model&#8221; does not deal with task internals, only the orchestration.  If you accept this, I think it eliminates the distinction between your horizontal and vertical.  Or at least blurs it.<br />
-Bruce</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sebastian</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4796</link>
		<dc:creator>sebastian</dc:creator>
		<pubDate>Sat, 05 Apr 2008 10:22:40 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4796</guid>
		<description>There is one thing you have to consider: Are you aiming at a vertical or horizontal transformation? Horizontal transformation means your BPMN is the executable model, all details are there. If you are doing a horizontal transformation, there is no point in transforming it to BPEL. On the other hand, if you are aiming at a vertical transformation, you are actually doing a refinement of your business process into a technical one. That means, your BPMN process stays platform independent and the BPEL process is platform specific (considering BPEL to be a specific platform). A horizontal transformation approach is a short-term solution and it works as long as you don't have to deploy your process on another technology. Vertical transformations are more complicated, can't be fully automated, but are long-term oriented, because you don't pollute your business process with technology specific details.</description>
		<content:encoded><![CDATA[<p>There is one thing you have to consider: Are you aiming at a vertical or horizontal transformation? Horizontal transformation means your BPMN is the executable model, all details are there. If you are doing a horizontal transformation, there is no point in transforming it to BPEL. On the other hand, if you are aiming at a vertical transformation, you are actually doing a refinement of your business process into a technical one. That means, your BPMN process stays platform independent and the BPEL process is platform specific (considering BPEL to be a specific platform). A horizontal transformation approach is a short-term solution and it works as long as you don&#8217;t have to deploy your process on another technology. Vertical transformations are more complicated, can&#8217;t be fully automated, but are long-term oriented, because you don&#8217;t pollute your business process with technology specific details.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Antoine</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4794</link>
		<dc:creator>Antoine</dc:creator>
		<pubDate>Sat, 05 Apr 2008 07:23:26 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4794</guid>
		<description>Bruce, I would recommend that you take a look at the &lt;a href="http://www.eclipse.org/stp/bpmn/eclipsecon2007/Modeling_BPM_For_Execution_BPMN_2_BPEL.pdf" rel="nofollow"&gt;slides&lt;/a&gt; of our talk at EclipseCon last year, from slide 13 to 18. The slide 16 shows the transformation of BPMN to a BPEL tree when a partial merge occurs.

http://www.eclipse.org/stp/bpmn/eclipsecon2007/Modeling_BPM_For_Execution_BPMN_2_BPEL.pdf

Disclaimer: I work for Intalio, Inc.</description>
		<content:encoded><![CDATA[<p>Bruce, I would recommend that you take a look at the <a href="http://www.eclipse.org/stp/bpmn/eclipsecon2007/Modeling_BPM_For_Execution_BPMN_2_BPEL.pdf" rel="nofollow">slides</a> of our talk at EclipseCon last year, from slide 13 to 18. The slide 16 shows the transformation of BPMN to a BPEL tree when a partial merge occurs.</p>
<p><a href="http://www.eclipse.org/stp/bpmn/eclipsecon2007/Modeling_BPM_For_Execution_BPMN_2_BPEL.pdf" rel="nofollow">http://www.eclipse.org/stp/bpmn/eclipsecon2007/Modeling_BPM_For_Execution_BPMN_2_BPEL.pdf</a></p>
<p>Disclaimer: I work for Intalio, Inc.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: kswenson</title>
		<link>http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4793</link>
		<dc:creator>kswenson</dc:creator>
		<pubDate>Sat, 05 Apr 2008 06:01:45 +0000</pubDate>
		<guid isPermaLink="false">http://www.brsilver.com/wordpress/2008/04/04/more-on-bpmn-to-bpel/#comment-4793</guid>
		<description>Excellent example!  It really makes clear the limitations of a block structured language.  It shows well that we are not just talking about a subset of the language, nor are we talking about a lack of semantics of the individual elements, but in fact the details of the engine effect how you can compose element to form a diagram.  I think we will see more of this example.</description>
		<content:encoded><![CDATA[<p>Excellent example!  It really makes clear the limitations of a block structured language.  It shows well that we are not just talking about a subset of the language, nor are we talking about a lack of semantics of the individual elements, but in fact the details of the engine effect how you can compose element to form a diagram.  I think we will see more of this example.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
