3

I am a researcher, and I would like to scrape a list of scientific libraries sites such as Springerlink.com, ScienceDirect.com, IEEExplore digital library and so on, to analyse them for some purpose. In this task I'm interested in abstracts, and full text of publicly accessible papers, I would like to know the legal framework of this, since that I read before (but I don't remember where and how to find it again) that this is illegal, and if I do that any new request from my IP address to those sites will be rejected. It means that this task will be considered as a Denial of service attack.

To what extent this is true?

Nafaa Boutefer
  • 461
  • 2
  • 9
  • 4
    EnergyNumbers has given the right answer in general. As for specifics, the sites you mention do not allow scraping or text mining, except under strictly limited circumstances (which rule out almost all cases). You should treat this very carefully. If you try to get permission, you will probably waste quite a bit of time and fail in the end to get the permission you seek. If you try it without permission, you could very well find yourself in court (like Aaron Swartz). If you can carry out your study using a smaller but freely accessible data set, I'd highly recommend trying that instead. – Anonymous Mathematician Dec 26 '12 at 02:38
  • 1
    Voting to close: this is a question about law, not academia. – Nate Eldredge Dec 26 '12 at 07:05
  • 2
    I see it as having both legal and academic aspects (but I don't have a strong opinion regarding whether it is a good fit for this site). The basic issue is what the terms of service for these sites allow, and that's partly a question about what sort of terms are customary for scholarly websites. – Anonymous Mathematician Dec 27 '12 at 02:16
  • @NateEldredge I don't see any vote to close cast, was there a technical problem? –  Jan 13 '13 at 18:19
  • @CharlesMorisset: Hmm, I definitely voted to close, but I don't see a number by "close" either, yet if I click it, it says "You have already voted to close this question." – Nate Eldredge Jan 14 '13 at 03:31
  • @CharlesMorisset I think it got closed and reopened, which erases previous close votes – 410 gone Jan 14 '13 at 08:13
  • @EnergyNumbers: Yes, that's what I thought too, but I couldn't find where to check that it has been indeed closed/reopened, and who did it. Anyway, I'm just glad to see it's not a bug. –  Jan 14 '13 at 10:16
  • Keep in mind that you can get in trouble with Google Scholar. There is a related discussion: API, EULA, and scraping for Google Scholar – Artem Kaznatcheev Jan 16 '13 at 03:24

1 Answers1

16

  1. I am not a lawyer. This is a legal question. Get proper legal advice.

  2. The legal framework is contract law, and intellectual property law. There may also be implications for other parts of the legal code if your scraping behaves like, or could be interpreted as, a denial-of-service attack or hacking.

  3. Read each site's terms and conditions. Understand the contract of use for each site. Interpret each contract within the legal context of the contract's jurisdiction.

  4. Take into account intellectual property law within your own country, and the publisher's country. You might also need to consider intellectual property law within the country where the web-server is located, too.

  5. Do bear in mind that there have been recent attempts across several jurisdictions to bring intellectual property within criminal, rather than civil law / contracts.

  6. I am not a lawyer. This is a legal question. Get proper legal advice.

410 gone
  • 25,792
  • 6
  • 79
  • 135
  • It is my opinion that the Government should interpret the law for its Citizens... not Lawyers. It would prevent a lot of senseless litigation if permission were granted before instead of second guessed afterwards. – Dale Aug 22 '13 at 01:02