GregorHagedorn - Thu May 04 2006 - Version 1.14
Parent topic: ClosedTopicSchemaDiscussionSDD09
This topic is for (the discussion of) recommendations to developers.
-- Main.BobMorris - 08 Mar 2004
Yes. Neither sex nor development stages (is a baby a human species?) are related to the concept of a taxonomic hierarchy. Populations, metapopulations, subspecies, species, genera etc. all reflect evolutionary history - sex and stages don't. Please see the discussion under ResolvedTopicRankLevelBogosity. The problem raised in TheProblemOfSex is nevertheless valid and we don't have a solution yet! (See also the attempt to clarify it under SecondaryClassifiersWithinClasses.)
-- Gregor Hagedorn, 9. Mar 2004
-- Main.JacobAsiedu - 11 Mar 2004
This surely depends entirely on the context of the description/key. For instance, if we are dealing with a discriminatory structure (a key, whether pathway or matrix) and the contradistinction is between one group of butterlies with dorsal hindwings that are "black bordered with orange-brown" versus another group that are "brown bordered with orange-black" then it would be quite appropriate to have those two as states of a single character. Note that each "state" is actually a "mini-description", a common occurrence (particularly in pathway keys, less so in matrix keys where the level of atomization is usually greater). If however the purpose of the character is for a description, it may be desirable to atomize it into "Sub-Margin of dorsal hindwing: black" and "Border of margin of dorsal hindwing: orange-brown". The point is that the level of atomization must be left to the user who will behave according to the context - we cannot be proscriptive here, as that way danger lies.
Is this what you meant?
-- Main.KevinThiele - 11 Mar 2004
Yes that is what i meant. Thanks. I think i have a better understanding now than i did this morning :) will let you all know when i face any more problems.
-- Main.JacobAsiedu - 12 Mar 2004
How should states true and false be declared? Is it up to each Terminology to deal with this? I hope not, for integration's sake. -- Main.BobMorris - 09 Aug 2004
At the moment they are simply declared as concept states, like red and blue. Why not? I do not see an integration issue depending on this. "Yes/No", "true/false", "absent/present", "square/not so" all may express the same boolean fact, depending on how the character is formulated. If we would define standard-wide constants for true and false, what would that gain us, without also standardizing the character for it? -- Gregor Hagedorn - 9 Aug. 2004