tag:blogger.com,1999:blog-83221272023-07-23T13:13:34.639+01:00Discussions about Java architecture, design, development, open source tools & frameworksBenhttp://www.blogger.com/profile/14528257402298962104noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-8322127.post-1095170396896824452004-09-14T15:00:00.000+01:002004-09-16T10:17:20.016+01:00<em>GMail accounts for the first 3 helpful comments!</em> <br /> <br />I'm currently working in a company where for various reasons, they have chosen to <strong>physically separate </strong>Tier 1 (presentation stuff) from Tier 2 (business logic stuff) for new web projects. <br /><ul><li>Tier 1 is an ATG Dynamo application running on a box <em>and</em></li><li>Tier 2 is a J2EE application running on WebSphere on another box</li></ul><p>In terms of designs, the architecture team recommends that where possible, all persistence code is built in WebSphere, and exposed as a service (think SOA) so that it could be reused from various other applications. They also recommend using Hibernate.</p><p>ATG developers are questioning this strategy, and say that <strong>ATG repositories are better than Hibernate</strong>. I was surprised to hear that, but I'm not able to comment on that, not knowing ATG repositories myself... </p><p>I'd like to hear comments about this especially if you have some experience in both technologies.</p> <br />Anonymoushttp://www.blogger.com/profile/14528257402298962104noreply@blogger.com38