Feel free to to leave a comment here and on the Talk page: http://en.wikipedia.org/wiki... - Pierre Lindenbaum

Thank you very much Neil. At least, I might use this "Talk page" in the future as a Reference. - Pierre Lindenbaum

Hum.. he's still not convinced. Any librarian/publisher to help ? - Pierre Lindenbaum

Maybe ask David Lipman for advice? - Maxine

I will also comment on Wikipedia, but please see http://www.ncbi.nlm.nih.gov/About... and note the public domain and fair-use statement. - joergkurtwegner

Joerg, doesn't that link say the the abstracts actually are *not* PD, but remain copyrighted by the respective owners, and that only Fair Use applies to those? - Egon Willighagen

@Egon: Yes, left exactly that comment on the Wiki, so happy digging into the copyright issues of journal abstracts for different publishers. Any copyright experts around, we could need one ? - joergkurtwegner

Alas I think Joerg/Egon are right - Deepak Singh

Folks, as Maxine suggested earlier, I've asked David Lipman for advice about this issue. Bear with me. - Graham Steel

Thank you Graham !. I just hope Dr Lipman's mailbox is not filled by this kind of requests every days :-) - Pierre Lindenbaum

why does it even matter? the style of writing for wikipedia is quite different to the normal style of a paper, and an abstract wouldn't make for much of a wikipedia article anyway -- especially someone's biography. wouldn't it be easier just to write the article in your own words? - Joe Dunckley

The owner of a copyrighted work controls how it can be used. Publishers own articles in most cases of scientific work. There is a fair use doctrine under copyright law but there isn't a bright line between to distinguish fair use from infringement. Copying and pasting into Wikipedia could constitute infringement. The safest path is to always ask the copyrigh owner for permission to use the work or portions of it. - Jack H. Pincus

There's no ambiguity. The license agreement for NLM data references the fact that reprinting of an abstract, as a summary of the larger work, is fair use. You don't have to re-word or excerpt an abstract, since it's already an excerpt. See http://www.nlm.nih.gov/databas... No need to contact each publisher individually. All of this is entirely beside point, because violating copyright isn't illegal. The content owner has to sue for removal, and they don't seem to care. - Mr. Gunn

Thanks for editing and signing my anon comment on the talk page, Neil. Should I not have mentioned Hubmed? - Mr. Gunn

The NML link references Appendix A which refers to data that NLM owns. Abstracts are owned by a publisher. The safest method for quoting is to summarize the abstract in a few sentences and link to it on the PubMed site. - Jack H. Pincus

So my question is, what about sites like GoPubMed, Hubmed, etc? They publish abstracts as is, and I prefer using them. - Deepak Singh

Jack, You don't have to summarize an abstract. It's already a summary, and the NLM license policy specifically says that if you license NLM data, you get the abstracts, which although they are contributed by a third party, are provided under fair use. Sites like GoPubmed and Hubmed exist precisely because abstracts fall under fair use. If they weren't, anyone could excerpt a paper to equivalent end. It's not like abstracts are valuable on their literary merits. - Mr. Gunn

Even if it weren't quite clear that abstracts are fair use summaries(see the Google cache decision), there would be no downside from an author's point of view to acting as if they were, so I don't know why they're giving you such a hard time, Pierre. My limited experience with wikipedia editing suggests that sometimes these arguments are won only through attrition. - Mr. Gunn

We have many opinions here. But copyright owenrhsip and rights are legal matters. Arguments here do not matter. The case law does. The case law does not really draw a bright on these matters. Why risk litigation when there is a safer path to the goal? - Jack H. Pincus

To Deepak's point, one can publish anything verbatim with the copyright owner's permission. Journals will give permission, sometimes without charge, under certain circumstances. - Jack H. Pincus

Jack, the only relevant case law I'm aware of is that regarding the legality of Google caching a copy of a page, and that's a pretty clear cut decision in favor of using abstracts according to fair use. It's not even worth worrying about. Even if you did violate someone's copyright, it's not like that's a crime. They just ask you to take it down. The benefit to society of having free and open access to citation information is significant, the risk that a publisher could make you stop is insignificant. - Mr. Gunn

Mr. Gunn: are you sure you understand what fair use is? i'm not sure i really understand it myself, but your characterisation of it doesn't look like the fair use that i'm familiar with. fair use does not mean public domain. it means some limited circumstances where limited portions of a copyright work may be reused. whether something is fair use depends on the circumstance, so fair use of an abstract in pubmed does not automatically mean you can claim fair use of the same material in an encyclopedia - Joe Dunckley

'murky' (hey, I just learned a new english word ! :-) ). Neil, I don't claim to be an expert about all the articles I've created, and, most of the time, I copy the abstract (99% are biographies) to wikipedia and I add some wiki-stuff. See http://www.ncbi.nlm.nih.gov/pubmed... and http://tinyurl.com/byf9nt . Here I aim to learn new things, I hope that the article will be completed and I 'save' the abstract to wikipedia. The article is then read by DBPedia or Freebase http://tinyurl.com/aqdvhn - Pierre Lindenbaum

If there is something in the copyrights of WP and/or pubmed that's forbid me to copy and structure this information, I just found it too bad. - Pierre Lindenbaum

I find it interesting that people commenting here are rendering legal opinions about what constitues fair use or infringement yet none of us (including me) are lawyers. But I routineley deal with intellectual property in my work (patents, copyrights and trademarks) and understand a few basic principles. Copyright law gives the copyright holder certain basic rights including the right to display. The Fair Use doctirne is part of cpyright law but doesn't draw a bright line between what is fair use and infringement. Case law covering copyrights is complicated. It is not correct to assume one can linearly exrapolate from one case to another situation. Quoting or copying verbatim places one at risk of being sued for infringement. The saftest way to proceed is to ask for permission. Many jornals will grant permission without charge for certain purposes inlcuding posting to sites such as Wikepedia. The process for getting permisson is not onerous. Why risk litigation or harassment? - Jack H. Pincus

Another important point is that copyright attaches to a work the instant it is created. The only exception is works created by U.S. government employees as part of their duties. One does not have to register a copyright to obtain its benefits. Placing a work on the internet or omitting a copyright symbol or other designation does not place a work in the public domain. - Jack H. Pincus

Jack, I think most of us here continue to be very frustrated at how science is locked in using artificial mechanisms that don't make any logical sense. I am not even going to try and figure out what the legal aspects are, but it's pretty clear (to me) that if an abstract on pubmed is subject to copyright, especially when we have an API available, then perhaps we need to talk to the publishers, some of whom are active here, and change that. I have a feeling they'll listen. - Deepak Singh

If not, more drastic action might be required :) - Deepak Singh

several publishers already have listened, and license their whole articles CC-BY. some other publishers would likely be happy to do that for the abstract alone. but there will always be some publishers -- the sort who like to run the arms trade on the side, for instance? -- who would surely refuse even to do that. some publishers love the arbitrary rules of the copyright system... - Joe Dunckley

I could be misinterpreting what's being suggested above, but I don't think a publisher can grant permission to use copyrighted material *only* in Wikipedia, since Wikipedia itself is licensed under GFDL (which in turn grants very specific usage permissions to the general public). - Andrew Su

Weighing in with my non-lawyer-ly 2 cents: Pierre - in answer to your original question, articles on PubMed are not public domain. Works authored by US Government employees are, but while articles on PubMed are hosted on a gov website, they are typically not authored by gov employees. Being licensed under a CC- license is not equivalent to being in the public domain. You can likely use the abstracts under fair use, but this means that the author (or publisher) still retains copyright. - Hilary

(cont.) As Joe said, fair use is not equivalent to being in the public domain. Adding to the confusion is that since "facts" are not eligible for copyright protection (only the expression of them is), parts of a scientific abstract may not be under copyright to begin with. IANAL, but the best thing to do would be to reword the abstract so it is not a verbatim copy, including the important facts, and to cite the original. - Hilary

Re Deepak's comments, I agree we need better ways to disseminate scientific information and data. Creative Commons offers a means to easily disseminate information while protecting the copyright holders intellectual property rights. - Jack H. Pincus

Thanks Hilary. re-arranging an abstract sounds like 'Brazil !' ( http://tinyurl.com/j45jg ) but I do understand your clarification. - Pierre Lindenbaum

Is this still the internet, or did I take a wrong turn somewhere? It doesn't matter that none of us are lawyers. We are authors who have nothing to gain by dithering on about this. Abstracts have been used in this way for years now with no one who has standing to complain caring. Sure, it would be nice to hear an official policy, but why interpret things in the most restrictive way possible? Scared you might get some sternly worded letter? - Mr. Gunn

Mr Gunn, I understand your point of view but I wouldn't like all the things I've written be deleted later just because someone IS an enthusiastic. lawyer :-) The first version of http://en.wikiquote.org as been deleted just because of this point. - Pierre Lindenbaum

Am I wrong in my understanding that the fair use doctrine means you don't have to ask for prior permission from the copyright holder to display a summary of a work on your site? Whether or not displaying an abstract is or isn't fair use would only be an issue if a publisher filed a lawsuit(which they don't seem interested in doing), so why assume they all would file and we'd lose? There are precedents which seems to indicate it would go our way, and there's no risk even if it doesn't. - Mr. Gunn

I guess I just fundamentally don't understand why anyone who's not an official representative of a publisher would have objected to Pierre's use of abstracts on Wikipedia in the first place. - Mr. Gunn

Mr. Gunn- from what I understand, fair use is a defense against acknowledged copyright infringement. The moderator on Wikipedia is trying to protect the site from possible future litigation (or takedown notices) which is what Wikipedia's policies are also designed to do. It's also interesting to note that all text contained in Wikipedia is copyrighted by Wikipedia contributors and licensed to the public under the GDFL (http://en.wikipedia.org/wiki...). So Wikipedia itself isn't in the public domain, even though it's content is open, freely available, and can be re-used/re-distributed. - Hilary

I understand that abstracts aren't generally licensable under GFDL, as the publisher maintains copyright. That's a problem of Wikipedia's design. The presumption, on the parts of the internet with which I'm familiar, is that it's OK to use something unless specific objections have been raised by the copyright owner in the past, not that all content is off limits if there's even a distant possibility of a challenge. There's certainly no need for content policemen acting on behalf of copyright holders. - Mr. Gunn

I think you need to be aware of the terms on the source site (any site) before copying and pasting, Mr G. I don't think the world works on "presumptions"??? - Maxine

I remember my dad asking almost 10 years ago, "Who owns the internet?" I didn't really know what to tell him, then, but later I realized that such ambiguity is part of everything leading edge. Progress on the internet is so much faster than laws can keep up with that we're perpetually in the realm of unclear legal guidance. If developers waited for laws to catch up, the internet itself wouldn't be here. - Mr. Gunn

The creative, inventive, innovative people who develop the technologies we all love don't sit around and wait for a legal ruling on everything they want to do. Google didn't wait to be told whether linking to pages within a site was OK, 23andme didn't wait for a ruling on whether DTC genetic testing was legally umambiguous, they just did it, and if they would have tried to wait, someone else would have gone ahead and beat them to market. Innovators just do stuff, and let the rules catch up. Life's too short - Mr. Gunn

Perhaps this lack of respect for authority and tradition is a typically American attitude(in California, you can't get through a normal working day without breaking a few minor laws), but it's as much a source of strength as it is a potential failing. So....that's it for this week's threadjack, I'll get my own blog now. Sorry if I put anyone off, it's just that this is an issue that I feel really strongly about, because it goes to the very heart of what makes the internet the great place it is. - Mr. Gunn

@Mr. Gunn: Companies such as Google and 23andMe also have this thing called lawyers :-) For the rest of us it's best to make sure that we're not only following the rules, but following the rules the way the respective copyright owners interpret them. Meh! - Eric Jain

i hope this helps: http://www.cotch.net/blog... . i wish copyright were much more relaxed than it is, but it's not, and wikipedia ain't really in a position to ignore copyright. - Joe Dunckley

The argument is "you can't because Wikipedia decided you can't at some point and they're not going to change their minds." Did I get it right? That's a valid argument. The fact that wikipedia decided at some point to not allow it doesn't mean that they wouldn't be covered under fair use if they were to change, or that anything is settled about the use of abstracts under fair use. The fact that changes, once released to the net, are hard to roll back argues for the practice, IMO. - Mr. Gunn

I'll be right back after I upload this pubmed XML dump to Pirate Bay. Shiver me timbers!! I ain't afreared of no scurvy publisher scallywags! Arr!!! - Mr. Gunn

that was one of several points made. another was that basing a wikipedia article on an abstract almost certainly would NOT be considered fair use of that abstract. fair use only grants you very limited rights, which a Wikipedia article would go far beyond. - Joe Dunckley

I've now heard back from David Lipman. He said, "As a general rule, the abstracts in PubMed are copyrighted, a general exception being, abstracts authored by a government employee as part of his/her official duties. In this regard, you may wish to note the Disclaimer text on the PubMed page which pertains to copyrighted materials and the use thereof…" - Graham Steel