Draft draft 0.04

[Rob Seaman]

Alasdair made lots of good comments (the document is in good hands):

> As someone with no background in astronomy, I find that I do not  
> understand a lot of the labels, this is to be expected. However, one  
> of the use cases we are trying to satisfy is that an astronomer who  
> specialises in one field can understand the specific terms used in  
> another field. So, for the general terms like planet, satellite,  
> spiral galaxies, it does seem a bit pointless having a definition  
> that explains these terms, however for more specific terms it would  
> be essential for cross field understanding and therefore all terms  
> should have definitions for good practice.

Wouldn't better practice be to omit definitions the WG regards as  
pointless?  :-)

There is also a scale problem, particularly in astronomy.  A "galaxy"  
isn't the same thing to all people, for all purposes.  A stellar  
astronomer (supposedly the typical class of astronomer) can't see the  
forest for the trees.  The scale of a galaxy is many times too large  
to matter.  On the opposite end of the spectrum (barely qualifies as a  
pun), a cosmologist will treat entire clusters of galaxies as raisins  
in rising bread dough.  The scale of a galaxy is many times too  
small.  No one brief description will span down from treating galaxies  
like nutmeg swirling on eggnog in brownian motion, to the Larry Niven  
regime of tens of thousands of "Known Spaces" within our own beloved  
candy bar.

Rather than Hilbert space, we occupy Horton space.

> Both parties could publish mappings along with anyone else who  
> decides to create a set of mappings.

Copacetic.

> Of course, the mappings themselves all have inverses, so if you have  
> a mapping from vocabulary A to vocabulary B then you automatically  
> have one from vocabulary B to vocabulary A.

Um.  Must mappings be one-to-one?  They certainly aren't onto.  In any  
event, a particular mapping can assert multiple exact matches (and  
narrower-than and broader-than matches).  I would think it would be  
trivial to construct a non-reflexive mapping, or a non-transitive pair  
of mappings.  Is there some sort of SKOS validator that detects such?   
(And might not such creative mappings be an interesting feature for  
some purposes?  Rock, paper, galaxy...)

Rob