tag:blogger.com,1999:blog-3602612821706176004.post2163318778120356673..comments2023-08-28T02:35:43.352-07:00Comments on Semantics and Data Modeling: RDF and its role in Logical/Canonical ModelingAnonymoushttp://www.blogger.com/profile/11746604103523406806noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-3602612821706176004.post-9812290223573664512012-12-07T05:39:04.169-08:002012-12-07T05:39:04.169-08:00Thanks for the detailed response and for sharing y...Thanks for the detailed response and for sharing your modeling insights on this blog. <br /><br />[Preface micro-rant: what's with the rudimentary Blogger comment edit control? Have the folks at Google never heard of embedded links or images?... I just lost an earlier draft of this comment, after I foolishly tried to drag in an image from Skitch; argh...]<br /><br />A few quick responses, for now:<br /><br />Sorry if I confused conceptual modeling, logical modeling, and UML in my response to your previous post. I do not think UML is great for either conceptual or logical modeling, but there's a perplexing shortage of useful (and affordable) modeling tools for conceptual/logical modeling (Open ModelSphere [http://www.modelsphere.org/] is one example) and also a shortage of useful resources for learning about conceptual modeling (the Carlis/Maguire book [http://www.amazon.com/gp/product/020170045X] is still the best resource I've run across), so many people default to UML tools for modeling.<br /><br />I definitely agree it's useful to explore both instances and types/categories/classes when learning about and modeling a domain -- that's a best practice in the Carlis/Maguire technique as well.<br /><br />At the outset, I think it's useful to distinguish between resources and relations -- see slide 18 in a recent Gilbane presentation [http://www.slideshare.net/pbokelly/gilbane-boston-2012-xml-and-sql-not-dead-yet] on XML, SQL, and related topics for more details. In that framework, RDF gets a bit intertwingled, as it's a resource-based model that's often used for modeling relations...<br /><br />Anyways, more to follow -- thanks again for sharing your insights in this blog; I look forward to continuing the discussion.pbokellyhttps://www.blogger.com/profile/18207982160283918003noreply@blogger.com