Just over a week ago I came across http://springerimages.com and stated publicly that some of my material had been recopyrighted by Springer and my legal and moral rights had been violated. Discussions ensued on this blog and on https://plus.google.com/101714021929763578604/posts/5d9Q5BJgzL7 (Springer’s GooglePlus). The discussion was heated in places. I used the word “theft” which I have retracted; Springer have referred to me in denigrating terms (“Mr Peter Murray-Rust, A Blogger”) which they have not further commented on. Yesterday an account appeared in Wikipedia’s Signpost http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2012-06-11/Special_report which is probably a reasonably complete and balanced account of the affair up to now (Wikipedia authors had their moral and legal rights infringed by Springer Images).
Yesterday Springer put out a statement http://www.springer.com/about+springer/media/pressreleases?SGWID=0-11002-6-1380826-0 and this post is a formal reply and summary of part of my position. I shall try to deal only with facts. Since the metadata on Springer Images is very limited I amy make errors and assumptions which are incorrect and I will correct them if so.
First Springer images. I will deal only with facts and legalities, not the moral and ethical aspects of the site and its business model which I shall hope to address later when facts are established. The metadata is very brief (http://www.springerimages.com/about.aspx ) and does not give details of where the images come from. My current belief is that almost all comes from content submitted by authors to Springer-owned journals and that all this content is , be default, in scope for inclusion in Springer Images. Over 90% appears to be published in closed access journals where authors transfer their copyright to Springer. The transfer probably allows Springer legally to re-use these images on SpringerImages in the way that is done. I shall comment on the ethics and morality in a separate post.
My interpretation of SpringerImages is that the public face is a catalog. To view most of the content a subscription is required (595 USD/year). Metadata and content are often not shown in the catalogue. It is unclear what the rights (if any) of a subscriber are as I have not paid the 600 USD. The re-use is given as “non-commercial” but it is unclear if there is an explicit licence. In many items the actual content is not given (even with a thumbnail – this is particularly true of tables).
Over 90% of the site is non-free and individuals have to pay approximately 63 USD for the re-use of a single image even if they are the author (I will show this later). Many authors have had their moral rights removed from the catalogue. I will illustrate this is a later post.
Biomed Central. When I discovered the site all of my BMC content had been rebadged as BMC’s copyright and my author’s moral rights (the right to be known as the author of the work) had been removed. An inappropriate licence (“non-commercial”, unspecified) had been used. BMC (but not SpringerImages) have apologized to me and the attribution and rights are now acceptable on the sample I have inspected. (BMC asked me to check that all my content on SpringerImages was acceptably labelled – I declined as that is their job, not mine).
Other rights holders (Wikipedia, other publishers, other rights holders). Much content had been systematically mislabelled (Wikipedia estimate over a thousand items and my own investigations suggest many thousand more for other rights holders). Since I do not know how SpringerImages assign metadata I cannot comment precisely, but it appears to have been done by automatic software which has made mistakes. Whatever the ethics of using software, the fact is that rights have been and continue to be violated.
Comments on Springer’s statement. I reproduce this in full and interleave comments:
Springer responds to issues around Springer Images
We would like to thank the research community for drawing our attention to several defects in our Springer Images product. Although we try our very best to deliver high-quality products and services that benefit the scientific community, sometimes we make mistakes, and we would like to apologize for those we have made in this case.
Thank you for this apology.
On 5 June 2012, we announced that we would be correcting the mistakes as quickly as possible, and we are now able to inform you that we have made a number of corrections to the product, but we must also admit that not all of them are solved yet.
This we agree on.
Our dedicated team will continue to work as fast as it can to eliminate the remaining glitches.
What we have done so far:
The new release of Springer Images went live last night at around 1800 hrs CET, and we are pleased to say that we have made the following changes which apply to all open access images BioMed Central and Springer Open (based on DOI):
• We have removed all copyright lines that contain the words BioMed Central and not the word licensee.
• We have replaced the copyright policy paragraph near the bottom of the image details page with “This image is published with open access under the Creative Commons Attribution license.”
• We have removed the line “If you would like to obtain permissions for the re-use or re-print of this image, please click here”. (NOTE, we have removed this line from *all* open access images (regardless of DOI).
This appears to correct the BMC-related problem and so SpringerImages are not, by default, violating the rights of BMC authors.
In addition, we have manually stopped display of *all* images with MediaWiki or Wikipedia in the caption. These images will not be displayed again until we can reliably differentiate among those that have non-commercial restrictions.
I note this. It is impossible to say how complete this is.
What we still must do:
We must still continue to closely monitor copyrightable adaptations of open access images, including those that build on previous work.
The words “continue to closely monitor” suggest that there has been a high-quality approach to metadata and rights. Since this process is not public I cannot assess this. None the less until I pointed it out over 1000 violations had occurred.
In these cases, it may often be unclear whether a figure is genuinely adapted, or simply reprinted.
This is a major, and well-known problem with metadata and rights. The SpringerImages pages give no indication that it had been addressed in a high-quality manner.
We will also be reaching out to Creative Commons and Wikipedia to investigate whether working together might help us to find the best approach to these challenges.
I do not see that it is the responsibility of Creative Commons and Wikipedia to give voluntary assistance to a commercial company whose business is re-selling IPR.
We will need to quickly address the fact that now several open access images do not list any copyright at all. This is a stopgap measure and our team is working on correcting it.
This is another major problem and if tackled responsibly takes a great deal of effort.
We will also continue to listen to our authors, researchers and the wider scientific community not only to ensure that our high quality products remain so, but also to correct any bugs that they may have. Should you wish to point out any further technical/copyright difficulties with Springer Images, please contact springerimages@springer.com.
We would like to stress that the (non-exclusive) inclusion of open access images in Springer Images is an example of the greater visibility and reuse possibilities that open access enables, and we would like to explicitly state that our intention is to ensure that all open access images in Springer Images are searchable and freely available in full, with no subscription required to access them.
Again, we would like to apologize to the authors whose images were affected as well as to the scientific community for the problems with this product. Furthermore, we would like to thank those of you that provided us with constructive criticism which has allowed us to correct the major issues quickly.
I do not know whether I am included in this list
Wim van der Stelt
Executive Vice President
Corporate Strategy
Springer Science+Business Media
I give Wim van der Stelt his full title here. We have corresponded before so he knows very well who I am. I would ask him, publicly, to confirm that “Dr Peter Murray-Rust, University of Cambridge” is the appropriate means of address. We may differ in many things but we can at least be courteous.
When Wikimedia got wind of this, some of us decided it would be best to reach out to Springer and ask if we could help them not completely shoot their foot off, which would be what their paragraph is referring to. And both Wikimedia and CC are very open to people asking “oh God how do I not mess this up?” particularly when they just have 🙂
That’s a very fair point and appreciated and Wikip/media should be congratulated for it.
I will address the ethics and morality of Springer Images in following posts.
Really? “Courteous”? If I am not mistaken, you, Dr. Murray-Rust, University of Cambridge, have used fantastically discourteous language in several of your posts. For example, I fail to see how referring to the people that work in publishing as “thieves” is courteous. (Have you apologized for this, btw, or just removed the reference?) You are lucky, Dr. Murray-Rust, University of Cambridge, that your employer obviously allows you a great deal of freedom to say what you like to whomever you like (occasionally without regard to the truth). If only those of us that are lower on the academic totem pole who would quite like to work in publishing (as their partner does) could shoot our mouthes of so… At the very least, you can only expect courtesy if you return it.
A Poor Postgrad in France
Pingback: Unilever Centre for Molecular Informatics, Cambridge - Making images Open can and should be routine « petermr's blog