tag:blogger.com,1999:blog-16081779.post2688432315388258210..comments2023-10-28T09:24:38.420+01:00Comments on iPhylo: Wiki modelling - Part 2Roderic Pagehttp://www.blogger.com/profile/00269598293846172649noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-16081779.post-55028327498814173162009-02-04T04:04:00.000+00:002009-02-04T04:04:00.000+00:00"I tend to view taxa in the same way -- results of...<I>"I tend to view taxa in the same way -- results of queries. It's been a long day, so I'm not sure if this is making sense"</I><BR/><BR/>It makes sense to me from an operational viewpoint. Taxa are sets, and what's the equivalent of a set when it comes to databases? Query results.<BR/><BR/>Not sure if it's quite what you mean, but I have some posts on SQL queries yielding information about taxa <A HREF="http://3lbmonkeybrain.blogspot.com/2009/01/using-sql-to-make-phylogenetic-queries.html" REL="nofollow">here</A> and <A HREF="http://3lbmonkeybrain.blogspot.com/2009/01/pursuant-to-my-last-post-using-sql-to.html" REL="nofollow">here</A>.<BR/><BR/>By "point location" do you mean geographical coordinates? If so, I don't see why all of those entities couldn't have coordinates. If not ... what's a point location?<BR/><BR/>Anyway, I have to admit that even in my schema, where one class represents every kind of taxon signifier, from clade to kingdom to species to specimen to character state, I do still find some need to associate them with semi-ad hoc categories (nomen, specimen, OTU, etc.). Oh well, no schema's perfect.Mike Keeseyhttps://www.blogger.com/profile/00147156174467903264noreply@blogger.comtag:blogger.com,1999:blog-16081779.post-75031096121117624662009-02-02T23:48:00.000+00:002009-02-02T23:48:00.000+00:00Mike,Thanks for the links. My own (still poorly fo...Mike,<BR/><BR/>Thanks for the links. My own (still poorly formed) view is that taxonomic names are tags, which are applied to a range of entities (such as specimens, sequences, observations, sets of taxa, etc.). Some of these entities are quite different things (e.g., it makes sense for a specimen to have a point location). Ultimately I'm more interested in the underlying data and links -- the names themselves are convenient labels.<BR/><BR/>The thing which struck me when I first started playing with relational databases some years ago is the realisation that many things I thought should be stored were actually better thought of as queries (or, more precisely, query results). I tend to view taxa in the same way -- results of queries. It's been a long day, so I'm not sure if this is making sense,Roderic Pagehttps://www.blogger.com/profile/00269598293846172649noreply@blogger.comtag:blogger.com,1999:blog-16081779.post-41980950828572870292009-02-02T19:07:00.000+00:002009-02-02T19:07:00.000+00:00I had a similar dilemma with taxa and taxonomic un...I had a similar dilemma with taxa and taxonomic units in <A HREF="http://3lbmonkeybrain.blogspot.com/search/label/Names%20on%20Nodes" REL="nofollow">my current project</A>. Recently I've decided to simply say that <A HREF="http://3lbmonkeybrain.blogspot.com/2009/02/taxa-without-organisms.html" REL="nofollow">a "taxon" is some kind of set and I don't care what the elements are</A> (organisms, species, genes, populations, whatever). Also, I'm considering taxonomic names, OTU/HTU labels, specimen identifiers, and character descriptions to all be the same thing: ways of signifying taxa. That cut out a lot of chaff in my class schema (<A HREF="http://3lbmonkeybrain.blogspot.com/2008/11/names-on-nodes-entities.html" REL="nofollow">here</A> if you're interested). Of course, what's chaff for my purposes may not be for yours.Mike Keeseyhttps://www.blogger.com/profile/00147156174467903264noreply@blogger.com