Search This Blog

Monday, April 09, 2012

DOI numbers, Web of Science, and article numbers

Two recurring complaints about bibliographies and citations for papers and proposals:
  • Most people really like DOI, a system meant to assure that reference materials like journal articles get an effectively permanent web address, something that will "always" point to that article.  It's become very very popular, and every online journal that I know provides a doi reference for each article.  It shows up in every Web of Science reference these days, too, if it exists.  So, why can't Web of Science make those doi numbers a clickable link?  That is, instead of forcing me to copy and paste the doi into a browser URL line with "http://dx.doi.org/" stuck in front, why not just make the doi itself a link to that?  I mean, why would anyone just want the doi without the link??  Is this some weird bs rule about Web of Science not wanting to have direct links?
  • How come Physical Review handles bibliographic information so badly when it comes to article numbers?   A number of years ago, Phys Rev switched from old-fashioned page numbers for articles to 6-digit article numbers.  Unfortunately, when you try to export bibliographic information for reference management software, for many Phys Rev articles, the automatic response is to stick the article number (which replaced the page number for all practical purposes) in some completely random field, and instead list the page numbers as either blank or the oh-so-useful "1-4" for a four-page article.  Can someone please fix this?  
Both of these are trivial, silly things, but I'd be willing to be that hundreds of person-hours (at least) are lost per year dealing with the latter one.

5 comments:

Anonymous said...

Try this for the DOI:
https://chrome.google.com/webstore/detail/goanbaknlbojfglcepjnankoobfakbpg

Then you don't have to wait around for WoS to update their site.

Jacques Distler said...

You are, I presume, familiar with the \href{}{} command, provided by the hyperref LaTeX package?

Or, even easier, these bibtex styles turn the journal-reference into a clickable DOI link, if you include a doi field in your .bib file.

Anonymous said...

>> I mean, why would anyone just want the doi without the link??

This is why CrossRef now insist that the DOI is always displayed with the http://dx.doi.org/ prefix. Although not all publishers adhere to the rule.

Douglas Natelson said...

Anon8:43, Jacques, thanks for the tips. My point is, as emphasized by anon2:46, I shouldn't have to work hard to convert a doi to a link, since whoever has the doi could just as easily put in a link in the first place. Also, if a journal is going to provide downloadable bibliographic information, they should format t in a way that conforms to standard use, so that it doesn't require later correction. Don't get me started on Nature Publishing Group's refusal to provide easily downloadable bibtex formatted info either....

Anonymous said...

Frankly, since they could, the publishers should be embedding the bibliographic information in the pdf they provide so that everything is in one unit that you can then pass around.

Jabref will let you do this, but the publishers ought to be doing this on their end.