[fitsbits] updates to the FITS standard document

[Tom McGlynn (NASA/GSFC Code 660.1)]

Just to put the current changes to the FITS standard in context I'd be 
curious if there is any standard list of changes to the FITS standard to 
which these changes could be compared.

My quick recollections of updates includes:

New structures:
    ASCII tables
    Binary tables
    Random groups (deprecated for new uses)

Semantics without any structural or format changes (i.e., just defines 
meaning of some keywords)
    WCS papers I-n  (general, spatial, time, ...)
    [The definition of a correction matrix may fall more into the 
previous category but I've never
    really read this part of the standard....]
    Deprecation of EPOCH and definition of EQUINOX

New datatypes
    Support for long integers
    Multidimensional arrays using TDIM

Formatting:
    Y2K handling of dates
    Allowing characters like tab, newline in ASCII table content
    Free formats in headers for non-critical params

Raw format:
    Blocking (since deprecated)


To my eye, the tiled image compression would fit in the new structure 
category and the long strings in formatting.  The column limits stuff is 
another semantic definition of keywords.  The Green bank, inheritance 
and blank space proposals seem unprecedented. They either couple things 
previously uncoupled (in the standard), or talk about processing of FITS 
data which has generally not been part of the standard.  If this is 
really the case we may want to tread more cautiously here.

     Tom McGlynn