[tei-council] valDescs

[Lou Burnard]

As first mooted in, er January 2013, I have now finished checking (i.e. 
removing) all valDesc elements. A few contradicted the associated 
datatype, but most of them added nothing to what was already stated by 
it. A small number provided additional information which I have 
transferred to either  the <desc> or a <remarks> element. I took the 
opportunity also of making the phrasing of descs consistent where these 
were being modified and noted a few potential attribute classes in 
passing. I did not find any valDescs which would be better expressed as 
additional schematron rules (largely because I think Sebastian got all 
those some time ago).

Onward and upward! The equinox is upon us.