tshrinivasan
commented
5 years ago Update from Carl Malamud -
The scanner we are using at the Indian Academy of Sciences is an Internet Archive "Table Top Scribe.":
https://internetarchivebooks.zendesk.com/hc/en-usThe advantages of this device are several:
It is peddle operated. Scanners achieve speeds of 500-800 pages per hour. That is a significant improvement over hand operation.
It is integrated into the Internet Archive workflow. So, not only does one scan, but after the images are completed and you "push," they are run through post-processing: cropping, deskewing, dewarping.
They then push into the Internet Archive where (for western languages) you get OCR, plus hosting, plus all the other stuff. (You can also choose not to push a book and keep it local, but I hope that is an exception rather than the rule.)
The software on the local computer is open source and is maintained by the Internet, so things like updates happen smoothly. We can replicate some of step 2 ourselves, but it will take a bit of heavy lifting. However, there is no reason we can't reverse engineer the TT Scribe hardware.
Here are a couple of other pointers:
-
The DIY Scanner on which the TT Scribe is based:
-
Additional drawings of the TT Scribe:
I see several of you are Chennai-based. I am having discussions with Brewster Kahle in January, I return to India in February. I'd like to spend a few days in Chennai discussing what it would take to start making a compatible device in India at a significantly lower cost. The camera (a Sony A6000) is about $1000, but the frame itself should be able to manufactured much less than the full cost at the Internet Archive (a system from them costs $13,000, but it is possible they will donate a few to me). Perhaps we can all meet then and plot strategy?
My goal is that in February we have developed a plan of action and then begin to execute on it over the course of 2019. I would like to see significant progress in scanning activity in Chennai, Bangalore (where things are well underway), and Mangalore (where we have a talented team on the ground as well). Would love to come to the Punjab as well. :)
Though good scanners are available in marker, a Table Top Scanner is good.
Archive.org sells one https://archive.org/details/tabletopscribesystem
Carl Malamud, carl@media.org , USA scans all public works and shares in archive.org for many years - https://public.resource.org/
He is setting up public knowledge centers in india to scan public works. As scanner from archive.org costs around 9 lakh rs, he is looking to build such scanners in India.
Rupika, ons3112@gmail.com from Punjab, India is scanning old punjabi books for a project for wikipedia. https://meta.wikimedia.org/wiki/GLAM-Wiki_Municipal_library,_Patiala_Collaboration
Subashini, germany, ksubashini@gmail.com - Preserves old tamil heritage content via http://www.tamilheritage.org/
Natkeeran, canada - For Noolaham Foundation - scans tamil books and share here - http://www.noolaham.org/wiki/index.php
Kalyan in Adambakkam, Chennai, kalyan@greendms.in had built a DIY scanner for around 50L INR.
See his video here https://youtu.be/WFf_C3pTZ5A
he followed the guides here - https://www.diybookscanner.org/
Spoke to him on this. He Donated the scanner to a college in coimbatore. He asked me to go there and check the scanner. If it fine, he can build another one for us and punjabi team.
Now we need a better design for the DIY scanner, so that we can use latest technologies, easy assemble, easy disassemble, easy operation etc.
Natkeeran, natkeeran@gmail.com from Noolaham Foundation is checking for the designs they used to build a scanner. Carl is looking for the designs of archive.org's scanner.
Will share here, if we get better design.