New York's Metropolitan Museum of Art makes available almost 400,000 images

http://github.com/jedahan/collections-api

http://scrapi.org

This is my first foray into generators, and building an API that hopefully will be used more publicly, so any suggestions are welcome.

Also if you want to chat about what its like working for the Met, more specifically the media lab there, ask away!

Also, if you are looking for a giant dump of those images on a hard drive, get in contact with me as the site is not the easiest to crawl. I only have about 260,000 of the images, and its about 260gb.

I don't understand this in two ways.

i) How can the Met claim copyright of reproductive scans of works which are in the public domain?

ii) If the Met provides "license-free" access to images to a certain group of academics, even if we accept that the Met holds a copyright over those images surely the academics are then free to redistribute those images license-free such that they can be used by anyone for any purpose?

I don't really understand what "license-free" means in this context.

What they have done recently is assert that images of artworks that are public domain are also in the public domain. Other museums and institutions have done this, but for the Met to do it is a big deal.

"The Museum's OASC initiative provides license- and cost-free access to images of artwork in the collection that the Museum believes to be in the public domain and free of other known restrictions for scholarly use in any media...

"What is scholarly content?

"Scholarly content encompasses scholarly publication in all media and is defined here as the dissemination of ideas and knowledge derived from study or research for educational/cultural purposes.... Scholarly publication is not limited to academic institutions or university publishers, since commercial publishers/entities may also produce scholarly publications. Scholarly publication is not restricted by print run. For the purposes of the Museum's license-free Open Access for Scholarly Content (OASC), users are encouraged to make their own assessment. However, all users must carefully review this website's Terms and Conditions prior to downloading or using images or any materials on this website.

"What are some examples of scholarly content?

"All school and academic work (including theses, dissertations, etc.), conference proceedings, journal articles, essays in Festschrifts, museum exhibition catalogues, non-commercially produced textbooks and educational materials, books published by university presses or the academic/scholarly imprint of commercial publishers, self-published books, and documentary films. All of these examples apply to scholarly publication in any media format (print, electronic, film, etc.)."

Note that this doesn't specifically include blogs, though "self-published books" tends to suggest they are included. However:

"What is not scholarly content?

"Commercial use, publication, or distribution in any media or format is not scholarly content. Some examples include: commercially published general-interest books in print or electronic media; all products, merchandise, (including posters, calendars, notecards, datebooks, mugs, etc.), advertisements, or promotional materials for any services or products in any media format; feature films or documentaries funded by commercial organizations."

While I can appreciate the desire to prevent untrammeled commercial use, this creates a fairly considerable area of ambiguity. Would a beaker incorporating an image be considered scholarly use? An academic group using an image in a presentation? For conference signage? For a conference shirt? Does the consideration change if the shirts are included in registration or sold separately?

How about images used in a scholarly work which is then cited, referenced, quoted, or included in a "general interest" publication?

In previous licensing developments, attempts have been made to distinguish "commercial" from "noncommercial" use. The boundaries have generally proven less than clear. In particular, they tend to discourage further use of works under such licenses.

One alternative is to distinguish use of images with other marks of the Museum in use. The Debian Project draws this distinction: the entire distribution is available free of charge, and there are a set of logos which are similarly free to be used. There is also a set of restricted logos which may be used only by the Debian Project itself (or others to whom it's specifically licensed).

A converse of this would be to require that an attribution or logo be included in all uses of the images. In effect the images then become additional advertising and publicity for the Museum. Of course, one risk becomes use of images in contexts for which the Museum may not wish to be associated (political content, sexually explicit uses, violent or other uses).

Another approach would be to limit the quality of images available. Even fairly high-quality online images tend not to be applicable to, say, high-quality hardcopy replication.

The fact that many of the works depicted in the images are themselves far outside copyright protection only further muddies the situation.