The Labville pilot project, conducted from 2005 to 2009 to help monitor antibiotic resistance (ABR), implemented an automated, daily, and secure transfer of bacteriology data from a network of 69 community laboratories (LABMs). The data were captured through optical character recognition (OCR) of the printouts of test reports. The evaluation of the system considered the number of LABMs equipped and the epidemiological indicators produced. Between 2005 and 2009, 44 LABMs transmitted their data. The proportions of resistance observed in each bacterial species were consistent with the literature: 20.9% of methicillin-resistant Staphylococcus aureus strains and 2.2% of third-generation cephalosporin-resistant Escherichia coli strains. However, the inconsistency of print formats regularly prevented data retrieval. The rules for data extraction and transcoding had to be constantly adapted, leading to the termination of the experiment in 2009. The Labville experiment confirmed that LABMs had the relevant data to monitor E. coli ESBL production in urban areas, but did not validate a sustainable technical solution for retrieving it. Surveillance of community-acquired infections remains to be strengthened, particularly in light of the spread of extended-spectrum beta-lactamase-producing Escherichia coli strains. Several alternatives can contribute to this: targeted studies (possibly repeated), the development of indicators specific to community medicine by National Reference Centers, and collaboration with existing networks. (R.A.)