Comments and questions to:
eptcs@eptcs.org
|
|
For website issues:
webmaster@eptcs.org
|
For this reason, and also because the response page contains information that is often not in the paper itself, and because in the future other formats may replace pdf (just like pdf has largely replaced postscript), it is preferable to refer and link to the response page of a paper, rather than to an electronic manifestation of the paper in a particular format.
For that reason a reference ought to contain the DOI of a paper, with a life link to corresponding URI, rather than a direct reference or link to the current URL of publisher's response page.
Background: These days, in the scientific community, double publication of the same material is not accepted. This applies to open access publications just as well as to publication venues that require subscription or payment in exchange for access. However, technical reports and papers on author's web page do not count as publications in this sense: they do not inhibit publication of the same material elsewhere. The same applies to deposition of papers in open access repositories such as arXiv.org. Hence it makes sense to distinguish archival from official publication, where the latter is the kind that comes with a prohibition on official publication somewhere else. There is a large correlation between this notion of official publication and that of a refereed publication.
Assigning a DOI to a publication makes it into an official publication in the above sense. In fact, as virtually all the defining characteristics of the concept of publication from the past have been superseded in the electronic age, assigning a DOI is a candidate for a formal definition of what it means to publish something.
The biggest graph appears to be build by CrossRef. The nodes in this graph are simply DOIs.
Note: A DOI always starts with "10." and hence never with "http".
For those authors who do not use bibtex, or object to the EPTCS style, an alternative is to add the text \doi{xyz} somewhere in the reference; best at the end, just before the notes, and before the listing of any alternative URL (e.g. at arXiv.org). Here xyz stands for the DOI of the reference (starting with "10."). Then, at the beginning of the bbl-file, add:
The DOI of a paper is usually listed on Publisher's response page for that paper. Often, they can also be found in DBLP, or at CrossRef.
Once you obtained the DOI of a publication, you can also perform a "DOI query", at the bottom of the same form, to find the data for that publication that CrossRef has obtained directly from the publisher. (Here "xml-xsd" yields an abbreviated output, and "unixref" one that contains all data CrossRef has obtained.)
Additionally, EPTCS provides below an interface that takes as input a bibtex file, and converts it into a series XML-queries to CrossRef, one for each reference with a missing DOI. CrossRef will then answer with the bibliographic data it has obtained directly from the publishers for the references it could find, including their DOIs. We extract the DOIs from that and present them to you.
Instead of using the interface above, EPTCS authors can just upload a paper at https://login.eptcs.org/ that is bibtexed with eptcs.bst but does not yet contain (all) the DOIs. Our software then automatically looks up all DOIs that can be found in the database of CrossRef, and our webpage shows them to authors. This way authors only have to look up the (hopefully few) remaining references in DBLP or the publisher's website.
|
Comments and questions to:
eptcs@eptcs.org
|
|
For website issues:
webmaster@eptcs.org
|