Why non-scaling solutions are bad for public access to reources

Google Books has an enormous amount of material. This is good. However, they paint copyright restrictions with a wide brush and err on the side of protecting copyright holders. So, most content on Google Books that has been published post-1923 are restricted (possibly all, but definitely most). This may or may not be good for most people, but it’s certainly bad in some specific instances, like with government documents. These are in the public domain and yet you can only see “snippets” on Google Books. Rick Prelinger described this phenomenon last year. The problem still exists. The concern, apparently is that cop[yrighted material may appear within these documents — hearings especially — and since Google can’t spare the humans to do the due diligence, we all suffer with restricted access. [freegovinfo]

google book search, what about govdocs?

One of the things I really enjoyed about the Internet Archive Open Library project was the software they used to attempting to determine whether works they were scanning were or were not under copyright. It was an elaborate set of questions and answers with access to some copyright databases. In contrast, unless I’m mistaken, Google Books just draws a line at 1923 and assumes everything after that date is in copyright. This includes government information which as you know is made with tax dollars and generally in the public domain. So why does Google Book Search treat all post-1923 books as under copyright? Just over-cautious?