<html><body bgcolor="#FFFFFF"><div>Hi Rich - some comments inline below.</div><div><br>On Jul 17, 2011, at 8:43 AM, Richard Pyle <<a href="mailto:deepreef@bishopmuseum.org">deepreef@bishopmuseum.org</a>> wrote:<br><br></div><div></div><blockquote type="cite"><div><blockquote type="cite"><font class="Apple-style-span" color="#0023A3"><font class="Apple-style-span" color="#005001"><br></font></font></blockquote><span>- A well-defined and enumerated set of whole organisms (pod of whales, pack</span><br><span>of wolves, etc.) documented in nature</span><br></div></blockquote><div><br></div><div>What distinguishes well-defined from not well-defined? It sounds like this is relatively clear to you. It's not to me, and to see whether there is broad agreement, these things need to be spelled out. (One benefit from thinking about vocabularies in terms of ontologies is that this forces one to be specific and precise, which can help a lot in surfacing and pinpointing disagreements that need to be resolved.)</div><br><blockquote type="cite"><div><span></span><span>The preceding set requires some elaboration. For example, a herbarium</span><br><span>specimen is usually a clipping or other small part of a larger whole plant.</span><br><span>Often multiple clippings from the same individual plant are taken and</span><br><span>preserved as separate herbarium specimens. Should there only be one</span><br><span>instance of this class representing the whole plant? Or should there be</span><br><span>multiple distinct instances of this class, one for each herbarium specimen?</span><br></div></blockquote><div><br></div><div>Yes, depending on what roles they play. As you say, a tissue sample can serve as an evidence (as can presumably non-biological entities, too), or may be the specimen being observed.</div><br><blockquote type="cite"><div><span>Similarly, suppose the mammal skeleton and skin in the above list are from</span><br><span>the same individual mammal. Would there be one instance of this class (for</span><br><span>the whole organism), and the Skeleton and Skin be treated as instances of</span><br><span>the Evidence class? Or, should there be two instances (one for the skin,</span><br><span>one for the skeleton)? Or should there be three instances (one for the</span><br><span>whole mammal organism, one for the preserved skin derived from it, and one</span><br><span>for the preserved skeleton derived from it)?</span><br></div></blockquote><div><br></div><div>My take is it would depend on what observation you are trying to represent, and hence which instance is the one being observed.</div><div><br></div><blockquote type="cite"><div><span></span><span>I would like to believe that all instances of Occurrence can be rooted into</span><br><span>one instance of this proposed new class. </span></div></blockquote><div><br></div><div>You may have meant something different, but just in case, instances aren't rooted, in the subsumption sense, in other instances. Only classes are.</div><span class="Apple-style-span" style="-webkit-tap-highlight-color: rgba(26, 26, 26, 0.296875); -webkit-composition-fill-color: rgba(175, 192, 227, 0.230469); -webkit-composition-frame-color: rgba(77, 128, 180, 0.230469); "><br>-hilmar<div><span class="Apple-style-span" style="-webkit-tap-highlight-color: rgba(26, 26, 26, 0.292969); -webkit-composition-fill-color: rgba(175, 192, 227, 0.222656); -webkit-composition-frame-color: rgba(77, 128, 180, 0.222656); "><br></span></div><div><span class="Apple-style-span" style="-webkit-tap-highlight-color: rgba(26, 26, 26, 0.292969); -webkit-composition-fill-color: rgba(175, 192, 227, 0.230469); -webkit-composition-frame-color: rgba(77, 128, 180, 0.230469); ">Sent with a tap.</span></div></span></body></html>