From: Dan Kohn (firstname.lastname@example.org)
Date: Sun Mar 12 2000 - 22:36:55 PST
Unfortunately, Uganda is landlocked, which tends to make shipping by sea
much more difficult. ;-(
Double-keying is arguably easier for me than OCR, but I find it hard to
believe that I'd get better accuracy. How about double-keying +
human-assisted OCR, then rectified by a 4th person.
Well, I guess I'll just have to try out different scenarios and see.
-- Daniel Kohn <mailto:email@example.com> tel:+1-425-602-6222 fax:+1-425-602-6223 http://www.dankohn.com
-----Original Message----- From: Gavin Thomas Nicol [mailto:firstname.lastname@example.org] Sent: Sunday, 2000-03-12 22:29 To: 'Fork (E-mail)' Subject: RE: Critical Decisions
> I'm a business in Seattle and would like to get 400 pages > entered with 99.9% accuracy within 48 hours. Any transfer > of paper to Uganda makes things basically infeasible, > because you have a 5-day ONE WAY FedEx trip, and the > cost probably overwhelms Uganda's low labor costs.
Depends on what you're doing... 400 pages and a premium for quick turnaround probably justify using (potentially expensive) network connections in Uganda. If I had 14,000 volumes of government records where the margin downstream was low, shipping via sea would probably be more cost effective.
FWIW. I have never seen 99.9% accuracy in large-scale conversion work. With regards to markup... the question of exactly what *is* 99.9% accurate also raises it's head... markup evolves.
> Don't you think human assisted OCR diffed with hand-keying > will be more reliable than either approach alone (or either > approach done twice)?
I'd still go with double-keying with (depending on the cost structure) reviewers.
I'm just trying to imagine the kind of people that would pay a premium for fast turnaround... or actually, what kind of data would justify it.
This archive was generated by hypermail 2b29 : Sun Mar 12 2000 - 22:49:56 PST