James Weinheimer suggested on Autocat:

>"I believe the basic problem lies in the "260 $c - Date of
>publication, distribution, etc." We are simply putting far too much
>information in this [sub]field ..."

Subfield code 260$d is available for copyright year as opposed to
publication year.  But whether separate subfield coding would be an
advantage to patrons depends of how it is applied in RDA.  The
presence of a year in any one 260 date subfield, it seems to me, would
obviate the need for an inclusion indicting its absence in any other.  

Following present RDA examples would result in: 260$c[year of
publication not identified],$g2010.  This would be a nonsense display.  

To go back to the pre AACR2 [n.d.] or phrase equivalent is
retrogressive.  To introduce labels in front of every 260 year would
take too much display space.  

SLC will not use $gyear in the absence of $fmanufacturer.  That name
would take no more room than a label (which may or may not be
accurate), and would be more informative.  SLC will not allow an
inclusion indicating absence of year in copy or original cataloguing.  
If copyright year is present, that will suffice.  If no year is in the
item, a guess will be made as in AACR2.  We would know this year that
the item was not published in 2012, which is more than the patron
would know seeing [date of publication not identified] in some future

   __       __   J. McRee (Mac) Elrod ([log in to unmask])
  {__  |   /     Special Libraries Cataloguing   HTTP://
  ___} |__ \__________________________________________________________