W3C home > Mailing lists > Public > www-html-editor@w3.org > April to June 1997

Shrink-wrapped frames

From: Daniel LaLiberte <liberte@ncsa.uiuc.edu>
Date: Tue, 13 May 1997 12:49:09 -0500
Message-Id: <3378A994.6EB9@ncsa.uiuc.edu>
To: www-html-editor@w3.org
Cc: liberte@ncsa.uiuc.edu, maloney@sq.com, lehors@w3.org
This is a request for a specific extension to the HTML
frames draft.

There is a need to specify the dimensions of a frame relative
to the contents.  This could be a very general capability, but
all that is really needed is to be able to create a frame that
shows *all* (and nothing but) the contents of the document 
displayed in it. I call these shrink-wrapped frames since they 
shrink (or expand) to just the size needed for the contents.

Shrink-wrapped frames are needed to properly display banners
completely when the exact rendering of the contents is not
known by the author/server, and therefore the fixed size value is 
not sufficient.  The precise contents itself may not even be 
known in advance.  If a frame ends up being too big, there is
wasted white space.  If it is too small, the user may be forced
to scroll the frame to access controls or valuable info.

Rendering such a frame is much like that of tables where we must 
wait until the contents of the cells are known before laying out 
the table.  But the problem is simpler here because the contents 
of a shrink-wrapped frame will typically be very small.

The syntax for specifying shrink-wrapping should be similar to
the syntax for other dimenion values.  I suggest using the '=' 
character, but something else might be better.

= 
  A single '=' character indicates a "shrink-wrapped" frame and is
  interpreted as a request to give the frame no more space than
  required to completely display its contents with no scroll bar. 
  Multiple shrink-wrapped frames may be specified. 

Another syntax allows more generality and is perhaps more useful.
But I haven't thought through the possible complications.

size1-size2
  The sizes are optional and each size is a fixed, proportional, 
  or relative value.  This syntax indicates a range of possible sizes
  for the frame where size1 is the minimum and size2 is the maximum.
  If size1 is missing, then the frame may grow up to size2.
  If size2 is missing, then the frame may shrink down to size1.

Similar extensions could be applied to other size attributes, for
images, tables, etc.

(Please CC liberte@ncsa.uiuc.edu in replies.)

--
Daniel LaLiberte (liberte@ncsa.uiuc.edu)
National Center for Supercomputing Applications
http://union.ncsa.uiuc.edu/~liberte/
Received on Tuesday, 13 May 1997 13:45:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:42 GMT