RECMGMT-L Archives

Records Management

RECMGMT-L@LISTSERV.IGGURU.US

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Larry Medina <[log in to unmask]>
Reply To:
Records Management Program <[log in to unmask]>
Date:
Mon, 11 Oct 2010 13:44:14 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (66 lines)
Paula-

While it's a good idea to find out what others are doing and their success
rate, the only way you're REALLY going to find out how well this works in
your environment is to perform some 'on the ground' tests with your own
source materials.

When it comes to issues of quality and performance, and especially those
meeting the expectations of a client population, nothing equals hard testing
of the material you will be working with day in and day out.

As for the comment about 300dpi being a standard for OCR, we found with our
HR files there was NO SUCH THING as a standard.  We actually went with
150dpi, based on a random sampling of files we took- some of which dated
back 40+ years, others that were about 15 years old and some more recent. 
Because of the wide variety of materials in these files (one sided, two
sided, colored paper, NCR forms, varying weights and sizes, etc.) there was
no way we could get by at 300dpi. We even had to end up making second high
contrast copies of some materials and scanning them at 75dpi to get a decent
output on the monitors.  Lots of materials had to be hand fed as well, and
our QC rate was 1:5 images.

And yes, the resultant files were rather large in size- can't give an
average because there really wasn't one.  All source materials were
reassembled and filed in batches following scanning, and after the first
round of QC, our Legal department made the determination that the images
would be used for reference only, and the paper files would remain the legal
record.  It's fortunate we did this because we've had to go back into the
files numerous times to replace bad images over the past 6 years. Our
retention requirement is 75 years beyond separation of staff- I doubt yours
is that long, but if you have DOT/FERC/EPA requirements, yours may be rather
long as well.

One thing we found was the slow processing of files was due to too small of
a 'communications pipe' and insufficient RAM on the machines receiving the
files.  After we knew what the file sizes were going to be that the users
would be working with, an evaluation of the systems was performed and a few
upgrades were in order, including the use of a server to 'cache' files when
requested that was cleared out daily at the end of the day.  The decision
was made to leave them for the entire day because we found it was common for
a file to be re-requested for reference on the same day it was initially
requested.  And we also found that because new materials were added to
active files after the original file was scanned, there was sometimes a
delay on requests because content had to be accessed form more than one
platter to get the entire content of an HR file.

Another problem with viewing the output was the quality of the monitors
being used by the HR staff.  Many of these had to be changed out for higher
resolution monitors that had better contrast control, and their lighting
even had to be adjusted in the office spaces some to reduce glare on the
screens.

I'd suggest a consideration of a deeper analysis into what is posing the
problems, it may be a combination of items (system architecture, software,
hardware, file sizes/types) and until you determine what all needs to be
addressed, you may not want to try tackling pieces that may only exacerbate
the problem.

Larry
[log in to unmask]

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance
To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message.
mailto:[log in to unmask]

ATOM RSS1 RSS2