Printed working documents of the Conseil d'Etat
   Printed documents of the Conseil d'Etat: Gérando, the digitised collection

Presentation of the digitised collection

The specifications of the digitised collection

Heading The printed working documents of the Conseil d'Etat: the Gérando collection, Conseil d'Etat Library
Dates 1800-1814
Volume of material 3660 documents
Technical characteristics of digitisation

With the aid of sponsorship from the Fondation Napoléon the whole Gérando collection was digitised in 2000:

1. As an image: each page of the original documents was scanned and stored as a TIFF group IV, two-tone, 300 dpi image.

2. As text: the full text of the documents, with the exception of a few complex tables, was keyed in manually from the scanned images, with an accuracy of 99.99 % (one error for every 10,000 characters). Available on this site.

Search tools

1. The general inventory of the printed documents of the Conseil d'Etat: select Conseil d'Etat Library in the document location field.

2. The full-text search engine which searches through all the words in all of the documents.

Detailed description

Reason for choosing the Gérando collection

The most complete unbound collection
The original collection comprises 3,660 documents, i.e., a total of 31,420 printed pages.

Physical description of the collection of originals
This is a full collection of printed working documents compiled by a Conseiller d'État in the discharge of his duties.

The Gérando collection of the Conseil d'État was chosen for practical reasons: it consists of unbound fascicles, approximately 25 x 21 cm, with a broad margin both towards the outer edge and at the bottom of the page, leaving a text area of just 18 x 10 cm approximately. Indeed even though some of the printed documents are in a larger format of approximately 32 x 21 cm, the text usually has the same layout as in the smaller formats. The fascicles, always bifolios, vary a great deal in bulk, from a single printed sheet (giving a total of 4 pages) to several hundred pages, and the pagination is in most cases printed. The most voluminous fascicles have sewn sections; as for the others, the bifolios are simply folded one inside the other. Some very large format tables (A2 or A3) were folded inside a few of the documents.
The digitisation project, which took 12 weeks, did not therefore require the originals to be handled excessively, and has not caused damage to the collection.

The technical options for digitisation and giving access to the documents

The digitisation was performed by an external service. The Gérando collection was taken to the premises of the Berger-Levrault company in Nancy to be scanned. The documents were scanned manually page by page in A3 format, in black and white (two-tone) with a definition of 300 dpi. The double-page images thus obtained were then cut to give one page at a time and stored in TIFF format with Group IV compression. The images of the pages of each document were then stored in the form of a multipage PDF file, named according to the barcode affixed by the library to the first page of the original document.

These image files were then sent to a workshop in the French-speaking country of Madagascar and used as the source for keying the text in manually (1): using Unicode. Quality control guaranteed 99.99% accuracy, i.e., one error per 10,000 characters. Typographical errors, common in these documents, which were often put together hurriedly, have been retained. Forty-five documents containing complex tables, for which HTML entry proved particularly problematic, were not fully keyed in. These documents are available for consultation in image form only.

The site gives online access to the digitised documents in text mode for obvious reasons: 1) ease of navigation (small size of text files) 2) greater search efficiency (indexing possible). In the case of the complex tables and documents containing handwritten annotations, they have been made available in image form and can be viewed using the freeware, Acrobat Reader.

Characteristics of the digitised collection

Digitisation in text mode
Digitisation in image mode
Name of file Name of file and extension ".html" Name of file and extension ".pdf"
Format HTML, character set Unicode TIFF Group IV, stored in the form of multipage PDF files
DTD HTML, slightly enlarged  
Method of digitisation Manual keying in, 99.99% accuracy 1 bit (line / two-tone)


300 dpi
Quantity 3660 files 31,420 images taken from the 3660 files.
Number of pages / file   1 to 100
Size of each file From 1 to 400 Ko. The average file is around 30 Ko. 100 to 8000 Ko
Use Web access via
Full-text indexing.
Archival storage, printing of facsimiles on request, access to the complex tables and handwritten annotations.
Display software required Navigator version 4 and above Acrobat Reader
Examples Projet de loi sur les dépenses relatives aux Cultes – GER01448.html Projet de loi sur les dépenses relatives aux Cultes – GER01448.pdf

The use of OCR (Optical Character Recognition) was rejected after tests showed that the quality of the texts thus produced was very poor as a result of the old typface, the damp patches on some of the documents and the complex layout of certain pages. [back]