1

I'm writing my thesis with Latex and I use Mendeley for the citation.
When I add a book and specify some attributes, like title, author and ISBN, Mendeley automatically adds some missing attributes.
One of those attributes is the URL of a book. Usually, this is a URL from amazon.com or books.google.com.

After generating the document with Latex, the URL also appears in my bibliography. It could look like:

Peter Smith. A Long Book Title. Pearson Education, 2008, p. 530.
isbn: 0-123-54431-2. doi: 10 . 1336 / 976 - 3 - 434 - 03844 - 5. arXiv:
arXiv : 1011 . 1889v3. url: http://www.amazon.com/gp/product/012345678?ie=UTF8&tag=petersmith-20&linkCode=as2&camp=4321&creative=4567&creativeASIN=087239586.

So the URL is very long and, in my opinion, useless. This is not the "real" URL of the book (not the URL of the author or the publisher) and I don't think, it adds any value to the bibliography.

In Mendeley, when I click on Tools->Options->Document Details, I can specify which attributes I want to appear. Sadly, I can't uncheck URL in the document type Book. So I assume, this is important?

Can I just manually remove the URLs of all by books?

7

Yes you can remove the URL as you have the ISBN number for each book; which is a unique identifier, universally given and searchable.

Note On The Generated URLs by E-Commerce Websites: Also make sure the on the URL you are using. The URL you posted is a generated one from Amazon that has number of variables for marketing purposes.

  • Agree with the basic principle. That said, the URLs mentioned are pretty useless - the Amazon one even contains an affiliate referrer ID code! - but it might be worth quickly checking before deleting them all, in case some are legitimately useful URLs (eg a book which is fully available online as well as in print). – Andrew Mar 14 '16 at 16:48
  • @Andrew Good point on Amazon's links, will add it to the answer, thanks. Is there any book without an ISBN number? As long as there is an ISBN number I don't think URL is that much important as the author uniquely identifies the book. Am I wrong on this? – o-0 Mar 14 '16 at 16:53
  • For unique identification & tracking a print copy, you're right that an ISBN is usually all that's needed. But the other links may have content value as opposed to simply identification value - eg if I was citing something like ebooks.cambridge.org/ebook.jsf?bid=CBO9781316161012 I'd want to always include the DOI as well as the ISBN, simply because it makes it much more accessible. – Andrew Mar 14 '16 at 17:11
  • Too bad there isn't a ISBN resolver like dx.doi.org... – Federico Poloni Mar 15 '16 at 7:31
  • 1
    @FedericoPoloni There is WorldCat which is enough in many cases. – Yet Another Geek Mar 15 '16 at 9:59
2

A trivial web search with the author name and the book header should usually return Amazon and similar sites where the book could be purchased, as well as probably some alternative sites, and the authors website as well. Same about ISBN number.

Hence, I think, URLs are redundant. Also, it is not uncommon for them to "rot" (disappear or change over time).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.