At 12:17 PM 1/28/00 -0500, you wrote:
>Maybe the question should be: Can a user make any sense out of a table
>with 1,500 rows in it? What if the result set contains 10,000 rows? My
>point is that maybe the result set should be broken up into smaller,
>fixed-size pieces, allowing the user to scroll around.
The question is still probably valid, of course! In the first place,
sometimes the reader is not a human being, but a process. In the second
place, some experiences with printers suggest to me that the amount of
memory consumed by a message is not always proportional to its
length. Sometimes it's reflected in its complexity. Admittedly, some of
the examples I recall involved really bad PostScript that didn't make use
of area fill operations.
Maybe the right approach in the context, though, assuming human readers, is
something like what you get with most Web searches now:
"The first [m] of approximately [O(n)] responses follow. Would you care to
refine your search criteria while I fill out a PO for more memory?"
Failing that, some attempt to reduce the flow to a structured series of
pages, might be in order, as suggested. Perhaps sub-pages keyed by the
first letter of the principal retrieval key.
John E. Koontz
--
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Archives and Other: <http://java.apache.org/main/mail.html>
Problems?: [EMAIL PROTECTED]
