<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:trackback="http://madskills.com/public/xml/rss/module/trackback/" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:pingback="http://madskills.com/public/xml/rss/module/pingback/" version="2.0">
  <channel>
    <title>Clemens Vasters - Talks|JAOO 2003</title>
    <link>http://vasters.com/clemensv/</link>
    <description>Cloud Development and Alien Abductions</description>
    <language>en-us</language>
    <copyright>Clemens Vasters</copyright>
    <lastBuildDate>Tue, 23 Sep 2003 08:05:35 GMT</lastBuildDate>
    <generator>newtelligence dasBlog 1.9.7067.0</generator>
    <managingEditor>cvasters@guhhome.com</managingEditor>
    <webMaster>cvasters@guhhome.com</webMaster>
    <item>
      <trackback:ping>http://vasters.com/clemensv/Trackback.aspx?guid=da00752f-3028-4c5c-bdce-178bfd14ebb3</trackback:ping>
      <pingback:server>http://vasters.com/clemensv/pingback.aspx</pingback:server>
      <pingback:target>http://vasters.com/clemensv/PermaLink,guid,da00752f-3028-4c5c-bdce-178bfd14ebb3.aspx</pingback:target>
      <dc:creator />
      <wfw:comment>http://vasters.com/clemensv/CommentView,guid,da00752f-3028-4c5c-bdce-178bfd14ebb3.aspx</wfw:comment>
      <wfw:commentRss>http://vasters.com/clemensv/SyndicationService.asmx/GetEntryCommentsRss?guid=da00752f-3028-4c5c-bdce-178bfd14ebb3</wfw:commentRss>
      <slash:comments>3</slash:comments>
      <body xmlns="http://www.w3.org/1999/xhtml">
        <p>
Here are the two PPT decks from yesterday's talks at the <a href="http://www.jaoo.dk">JAOO</a> conference
and a few notes...
</p>
        <p>
          <a href="http://staff.newtelligence.net/clemensv/content/binary/Layers-Tiers-Aspects-CV-V2.ppt">Layers-Tiers-Aspects-CV-V2.ppt
(1.24 MB)</a>:
</p>
        <p>
This deck is about layers and tiers and highlights (well, the talk that goes along
with the deck does) how I make a strict distinction between the term "layer" and "tier".
"Layer" is about organizing code in order to make it more resilient against change
in other layers and "tier" is about distributing layers across processes and machines
and defining appropriate boundaries as well as selecting technologies to cross these
boundaries. I am also advocating to generalize the "classic" 3-layer (not tier!)
model of "presentation", "business logic", and "data access" and make the underlying
idea a pervasive and recursive pattern for basically all code in a business
app.
</p>
        <p>
Any class and any module may have one or multiple "public interfaces" that
may be mapped to several incoming channels bound to different technologies. The public
interfaces themselves (this includes public methods of a plain class) don't implement
any logic, but always delegate to a strictly private internal implementation.
That implementation, in turn, will not talk to external resources and services
directly, but bind to abstract interfaces and access them via factories. <em>(I
will explain this in more detail here when I can make the time to do so)</em></p>
        <p>
At JAOO, the short AOP section of this deck drew some furious comments from an
attendee after the session, who said that I was totally wrong and the AOP worked brillantly
as a general purpose programming paradigm. However, when talking to him for a while,
he had to admit that he and the colleagues on his project are indeed carefully considering
and defining aspect dependencies and he sort of acknowledged that while their
set of aspects will work great in and by itself, but it would be hard to combine it
with an arbitrary foreign set of aspects. My main takeaway from the discussion with
him was, though, that (a) it's due time for Java (and C#) to get support for generics,
because that may be a better tool for a couple of things he pointed out and (b)
that if you give people a tool like AspectJ, they will just jump and reinvent
the wheel. The aspects he said his team implemented were (in ES terms) Transactions,
JITA, Tracing, Security, etc. All the usual suspects.
</p>
        <p>
          <a href="http://staff.newtelligence.net/clemensv/content/binary/SOA-CV-V1-final.ppt">SOA-CV-V1-final.ppt
(745.5 KB)</a>
        </p>
        <p>
This deck is an updated version of the Service Oriented Architectures deck that I've
been using for this year's Microsoft EMEA Architect's Tour. I've included a couple
of new aspects, including a stronger endorsement of UDDI, an explanation of the relevance
WS-Policy and WS-Addressing, a look at the relevance of WSDL in the presence
policy and addressing and a reference (and two borrowed slides) to my friend Arvindra
Sehmi's <a href="http://www.thearchitectexchange.com/PortalCSVS/uploads/Autonomous%20Computing%20-%20Abstract%20Queuing%20Nets%20and%20The%20AgileMachine.zip">most
excellent presentation </a> (free registration may be required) on autonomous
computing and queing networks, which has become a very important part of
the overall SOA story for me.
</p>
        <p>
 
</p>
        <img width="0" height="0" src="http://vasters.com/clemensv/aggbug.ashx?id=da00752f-3028-4c5c-bdce-178bfd14ebb3" />
      </body>
      <title>JAOO conference notes &amp; PPTs</title>
      <guid isPermaLink="false">http://vasters.com/clemensv/PermaLink,guid,da00752f-3028-4c5c-bdce-178bfd14ebb3.aspx</guid>
      <link>http://vasters.com/clemensv/2003/09/23/JAOO+Conference+Notes+PPTs.aspx</link>
      <pubDate>Tue, 23 Sep 2003 08:05:35 GMT</pubDate>
      <description>&lt;p&gt;
Here are the two PPT decks from&amp;nbsp;yesterday's talks at the &lt;a href="http://www.jaoo.dk"&gt;JAOO&lt;/a&gt; conference
and a few notes...
&lt;/p&gt;
&lt;p&gt;
&lt;a href="http://staff.newtelligence.net/clemensv/content/binary/Layers-Tiers-Aspects-CV-V2.ppt"&gt;Layers-Tiers-Aspects-CV-V2.ppt
(1.24 MB)&lt;/a&gt;:
&lt;/p&gt;
&lt;p&gt;
This deck is about layers and tiers and highlights (well, the talk that goes along
with the deck does) how I make a strict distinction between the term "layer" and "tier".
"Layer"&amp;nbsp;is about organizing code in order to make it more resilient against change
in other layers and "tier" is about distributing layers across processes and machines
and defining appropriate boundaries as well as selecting technologies to cross these
boundaries. I am also advocating to&amp;nbsp;generalize the "classic" 3-layer (not tier!)
model of "presentation", "business logic", and "data access" and make&amp;nbsp;the underlying
idea&amp;nbsp;a pervasive and&amp;nbsp;recursive pattern for basically all code in a business
app.
&lt;/p&gt;
&lt;p&gt;
Any class and any module&amp;nbsp;may have&amp;nbsp;one or multiple "public interfaces" that
may be mapped to several incoming channels bound to different technologies. The public
interfaces themselves (this includes public methods of a plain class) don't implement
any logic, but always delegate to a strictly private&amp;nbsp;internal implementation.
That implementation, in turn, will not talk to external&amp;nbsp;resources and services
directly, but&amp;nbsp;bind&amp;nbsp;to abstract&amp;nbsp;interfaces and access them via&amp;nbsp;factories. &lt;em&gt;(I
will explain this in more detail&amp;nbsp;here&amp;nbsp;when I can make the time to do so)&lt;/em&gt;
&lt;/p&gt;
&lt;p&gt;
At JAOO, the short AOP section of this deck&amp;nbsp;drew some furious comments from an
attendee after the session, who said that I was totally wrong and the AOP worked brillantly
as a general purpose programming paradigm. However, when talking to him for a while,
he had to admit that he and the colleagues on his project are indeed carefully&amp;nbsp;considering
and defining&amp;nbsp;aspect dependencies and he sort of acknowledged that while their
set of aspects will work great in and by itself, but it would be hard to combine it
with an arbitrary foreign set of aspects. My main takeaway from the discussion with
him was, though, that (a) it's due time for Java (and C#) to get support for generics,
because that may&amp;nbsp;be a better tool for a couple of things he pointed out and (b)
that if you give people a tool like AspectJ, they will just&amp;nbsp;jump and reinvent
the wheel. The aspects he said his team implemented were (in&amp;nbsp;ES terms) Transactions,
JITA, Tracing, Security, etc. All the usual suspects.
&lt;/p&gt;
&lt;p&gt;
&lt;a href="http://staff.newtelligence.net/clemensv/content/binary/SOA-CV-V1-final.ppt"&gt;SOA-CV-V1-final.ppt
(745.5 KB)&lt;/a&gt;
&lt;/p&gt;
&lt;p&gt;
This deck is an updated version of the Service Oriented Architectures deck that I've
been using for this year's Microsoft EMEA Architect's Tour. I've included a couple
of new aspects, including a stronger endorsement of UDDI, an explanation of the relevance
WS-Policy and WS-Addressing, a&amp;nbsp;look at&amp;nbsp;the relevance of WSDL in the presence
policy and addressing and a reference (and two borrowed slides) to my&amp;nbsp;friend&amp;nbsp;Arvindra
Sehmi's &lt;a href="http://www.thearchitectexchange.com/PortalCSVS/uploads/Autonomous%20Computing%20-%20Abstract%20Queuing%20Nets%20and%20The%20AgileMachine.zip"&gt;most
excellent presentation &lt;/a&gt;&amp;nbsp;(free registration may be required) on autonomous
computing and queing networks, which&amp;nbsp;has become a&amp;nbsp;very important part of
the overall SOA story for me.
&lt;/p&gt;
&lt;p&gt;
&amp;nbsp;
&lt;/p&gt;
&lt;img width="0" height="0" src="http://vasters.com/clemensv/aggbug.ashx?id=da00752f-3028-4c5c-bdce-178bfd14ebb3" /&gt;</description>
      <comments>http://vasters.com/clemensv/CommentView,guid,da00752f-3028-4c5c-bdce-178bfd14ebb3.aspx</comments>
      <category>Talks</category>
      <category>Talks/JAOO 2003</category>
    </item>
  </channel>
</rss>