Head Belly Root
Seeds Text Creative Commons License
NetBibTeX

Introduction

Maintaining BiBTeX databases is cumbersome. Sure, you can copy entries from somewhere else. But it would be even easier if BiBTeX could access these remote database directly, and get the records it needs from there.

User interface

As a user, you can also add URL's as bibliographic databases, e.g.
\bibliography{http://www.
acm.org/dl/jacm.bib}

Because people tend to have their own favourite citation key scheme, we allow aliasing: a locally maintained file that gives for a local citation key the corresponding citation key in the remote database.

A properly set-up remote BiBTeX database should list predefined strings and conference proceedings in separate files, one for each citation style (e.g. for long - "Journal of the ACM" - and short - "J. ACM" - styles).

Implementation

BiBTeX fetches the specified databases as if they are on the local file systems, performs any alias substitutions, and runs as normal.

Drawbacks

There are two fundamental problems with this approach. First of all, the on-line BiBTeX database you use may become unavailable, or one of the entries you used changed after the last time you processed the article. As a result, a later run over the same article produces a different output. This may influence your version management.

Secondly, the way the information is entered in the BiBTeX database may not be complete, or may not be completely what you want (e.g. you prefer to list the location of a conference in the address field of an InProceedings or Proceedings entry). This may be mitigated somewhat by implementing a local substitution table, or by using only the on-line strings and conference databases (saving already quite some trouble).