The quality and reintegration of produced data
The issue of quality and vandalism is a central argument of opponents of outsourcing data production to amateur Internet users. There are, however, many proven ways to guarantee one and prevent the other using robots, selfregulation, inspection by professionals or volunteers, evaluations, votes, aptitude tests, or even double-keying.
Whatever the case may be, there are numerous studies that show that data produced by those who consult a digitized document online can be of the highest quality, as the person consulting the document is generally someone who knows the subject covered in the consulted document well. Furthermore, as expected by followers of the “wisdom of crowds”, the diversity of profiles constituted by a crowd of Internet users and the “law of large numbers” would have the effect of neutralizing individual errors in the mass of accurate data [BOE 12] and sometimes even providing better results than those obtained by experts. This seems to be confirmed by several comparative studies [ROR 10, OOM 11].
However, the data produced by Internet users are not always reintegrated by institutions sometimes aiming more towards democratization of heritage or even institutional communication on a trendy subject than real Internet user participation. When they are reused, these data born of free, volunteer contributions should always be distributed under legal conditions authorizing the largest possible reuse.