>I have a spreadsheet that calculates the cost of an imaging project, how
>much storage space will be required on the server for digital images,
>and how long it will take.  It will also show how many hours it will
>take to prep the documents, scan them, index them, and verify/QC them.
>The spreadsheet also includes the cost of buying scanners, licenses, and
>other setup costs.  It also calculates all of this for imaging backfile
>records.  Our departments that use our services find this spreadsheet of
>great value, and so do I.

Seriously, I'm NOT ATTEMPTING to throw water on this, BUT....

Having been involved in countless scanning/conversion/imaging projects over
the past umpty-ump years I can guarantee you there is no silver bullet to
estimating the cost or effort required to complete one of these projects
unless your source materials are all once printed, never touched, loose
pages, printed from a common source.

The threshold at which pages will be captured depends on the initial quality
and other factors, such as type size, font, text density, manual
annotations, paper color, type, weight, color of text, presence of images,
and other items.  Some pages may be able to be sufficiently captured at 75
or 150 dpi with the use of an automated feeder, others may require capture
at anywhere between 300 and 1200 dpi, and even require the production of an
intermediate copy if the content is extremely weak.

In addition, prep depends on some of the same factors mention plus whether
there are staples, two or three hole punches, comb binding, creases, forms,
two sided content, NCR paper, carbon, odd sized originals, sort sheets or
dividers, a need to return originals to the same condition (post scan) or
process as batches, etc.

All of the above will impact QC and imaging time, and depending on what
needs to be indexed on which records, indexing time.  To a degree, these
factors will influence the volume of storage space for the images, as files
scanned at a higher dpi rate require more storage space.  

Other factors to consider include the retention time for these images and
the usage pattern for the initial collection, along with the growth rate. 
If the images of the files will be subject to destruction, and they are all
burned sequentially onto the common platters but the entire 'library'
composes multiple platters, accessing newer items (which presumably would be
scanned on a day forward basis) that may be part of a collection of items
that reside on multiple platters in the initial capture would require a
jukebox or other setup that allows random access.  

There may also be a need to copy portions of the content to newer media to
allow for deletion of existing media as retention periods are reached
because most WORM (and some other) media only allow for the deletion of
indexing information or 'pointers' to the images, not the images themselves.
In an e-discovery case or a data forensics investigation, any images that
remained on the platters/media would be potentially subject to production.

The best way to estimate the cost/effort required for these types of
projects is to identify a representative sample of records from across the
collection and physically process them- prep, scan, index, 1-to-1 QC images
and indexing data, post-scan prep source materials, validate storage space
required, and do a lessons learned. 

Give some consideration to what you do with the physical source documents
(use patterns, dispositioning/destruction, charge-outs, etc.) and work with
your user population to see how well they will be able to work with the
resultant electronic images- some users may require better or bigger
monitors, faster network connection, better printers, etc. once these
changes are made and all of these costs should be taken into account as a
part of the project to convert the content.

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]