The Language Lab site is fascinating, Jo. I'd never came across it before...

Ronnie

Getting an EU contract? It's easy. Just make sure you're born into the correct Belgian or French family, go to one of the elite schools and then universities, make friends with all the future Eurocrats, and then just phone them up and remind them of all the good times you had at school eating frogs' legs or whatever. (You might need a time machine.) Bribery probably comes in handy too.

It's great, isn't it, Ronnie? Especially the stuff about eggcorns.

Squander,

If your estimations are correct??? this is the way of how do business...
Infotechnique has understood everything about that and I have to employ the sales guy who sold this project...

Regards,
juliette

Just a comment from Infotechnique : the project is more complex that only digitisation : books of certificates notarized are handwritten and we process all scanned pages using dual data input with encryption to optimize data confidentiality. The result of this data input is stored in XML files that are used within the new electronic Property Register. Please have a look on http://www.infotechnique.com/main_en.htm to a better understanding of the scope of the project AMALFI.

Thank you, JF.

The Kirtas machines are by no means fully automated, they require a dedicated operator for each machine. Also, the average throughput is far below the 1200 or 2400 that is theoretically possible, when you take into account the time it takes to process the images with the included software. For color output you would be lucky to get 500-600 pages per hour. Also they are unsuitable for many of the larger dimensioned volumes that Europe has. That's why Infotechnique uses four of the Swiss 4digitalbooks machines, in addition to a single Kirtas APT2400. The 4digitalbooks machines can do the big volumes, and at higher resolutions than the Kirtas.

Also, when looking at the cost of a project, you need to remember that the time/labor spent indexing/entering metadata for each scanned page can often be a larger cost than the cost of initially scanning. No point having 30 million pages scanned really cheaply, if you can't search for the content inside each page.