At 10:08 AM 7/17/97 -0500, John Pickering wrote:
>John
>
>Sorry for my misinterpretation. I initially thought that the ksam file
>was part of the actual printing process rather than a repository for yet
>to be printed labels. [Although it is hard to imagine a label complicated
>enough to require an indexed work file :-]

It's not the label that's complicated, it's the circumstances surrounding
the label.
<grin>

>If I was faced with making changes to the existing app then I would try
>to step way back and look at how it really should be done. How about
>printing labels one by one as the information is entered? If the address

Let's see if I can fill-in some pieces I originally thought were not
important.  The work (essentially it's customer correspondence) arrives in
the mail.  It's divided into stacks (geographic regions) and given to staff
members to work on.

Until six months ago, the mail arrived by noon and the work was done by 4
pm and the labels for outgoing correspondence were printed before going
home.  The volume has increased and staffing has increased. Yet certain
times of the month (first and last weeks) they can not get it all processed
in one day so there is a carry-over.

The indexed file is keyed with/to the customer data in the main database.
It also holds certain other pieces of information such as the code(s) for
the inserts to go in the envelope.  The only reason it's indexed is to
allow the staff person to re-find the entry and modify it if necessary or
even mark the record to be ignored.  I'd rather not get into why an entry
in the file might need to be modified and why the staff person can not use
Quick's Scan mode rather than Find mode.  I should also comment that this
is not a heads-down data entry situation.

In most cases, two labels are generated: a mailing label to stick on the
form letter (shows through envelope window) and a second label to stick on
one of the pages inside. We use a HP2934 (???) dot matrix to print labels
and it's dedicated to that function.  The last I heard (3 months ago), they
were printing 10-12,000 labels per month.

The way things go right now, work on the day's correspondence stops about 4
pm so the label printing job can run.  Anything not processed by that
cutoff gets set aside for tomorrow.  If there is carry-over, there will be
another label printing run the next morning as soon as the carry-over work
is completed.  This batch printing is considered to be a bottleneck to
getting work out the door.  It also takes longer to get all the labels
printed when they're run as one big batch.

So the idea is to have a separate indexed file for each staff person.  When
that person is finished, the labels are generated.  No need to wait for the
cutoff and the label printing is spread out, too.  The staff people
sometimes have interruptions (meetings, visitors, etc.) that require them
to logoff the system.  So the files have to be permanent (not temporary).

I haven't decided what to do.  I'm hoping for some inspiration from this
group. <grinning>

Thanks for the response and I hope my explanation above helps you
understand what's happening.


------------------------------------------------------------------
John Pearce  <[log in to unmask]>       | Bethesda Management Company
Speaking for only myself             | Colorado Springs, CO  USA