[eml-dev] [Bug 3181] - xs:string to ComplexType TextType, minOccurs=0, judiciously applied

[bugzilla-daemon at ecoinformatics.org]

http://bugzilla.ecoinformatics.org/show_bug.cgi?id=3181


mob at icess.ucsb.edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|jones at nceas.ucsb.edu        |jbrunt at lternet.edu
            Summary|xs;string to ComplexType    |xs:string to ComplexType
                   |TextType, minOccurs=0,      |TextType, minOccurs=0,
                   |judiciously applied         |judiciously applied
   Target Milestone|EML2.1.0                    |Postpone




------- Comment #3 from mob at icess.ucsb.edu  2008-10-01 09:57 -------
Summary 2008-09-30:
Matt Jones, Margaret O'Brien, James Brunt, Mark Servilla, Inigo San Gil, Chris
Jones, Corinna Gries, Ken Ramsey

Retyping from xs:string to TextType to allow formatting in certain fields is
not an ideal solution for 2 reasons (also discussed in previous comments to
this bug)
1. any embedded elements within titles and abstracts will create barriers to 
searching in both knb and other catalogs, and 
2. the practice of adding visual formatting to EML does not conform to the
basic principle that XML should encode semantics not style.  

A better solution might be to identify the terms which need additional 
formatting and create structures to accomodate them which can be transformed as
needed. 
Known terms include species binomials and chemical formulae, which might be
surrounded by tags such as <speciesBinomial> and <chemicalFormula>. 
<speciesBinomial> would have simple content that could be transformed 
(e.g., to <emphasis> in stylesheets). Authors would need to employ another
schema (e.g. ChemicalML) inside <chemicalFormula>. Other types of terms used by
EML authors will need to be identified.

A subgroup of eml-dev (James Brunt, Chris Jones) will consider possible 
olutions.  Current uses of txt:TextType in instance docs should be examined to
determine whether these are semantic or presentational, and to identify terms
to consider for additional structures. Other schema languages could also be
examined (FDGC has no provisions for formatting resource titles).