CrossTeX is a new bibliography management tool. It comprises a new bibliographic database format that is much less prone to error and much easier to manage compared to other alternatives like BibTeX, and a new, more flexible tool for creating the citations that appear at the end of scholarly texts.
CrossTeX was born out of the author's frustration with inconsistencies and errors in the citations of scholarly papers.
Citation databases for BibTeX are written in a format that is easy for computers to parse but difficult for humans to maintain. In turn, this leads to inaccurate or inconsistent citations, difficulty in recalling citation keys which in turn can lead to mis-citations, and frequent, massive changes to the database, often to get a paper to fit under a given page limit.
A typical BibTeX file, assembled over time by cutting and pasting citations from the web, looks like a hodge-podge where some authors get full names while others only get initials, some states are spelled out while others get just two letters, and the conferences are abbreviated at random. This leads to messy citations. Certainly, no scientist would keep their experimental data in a sloppy format, or randomly insert punctuation into their code, or haphazardly abbreviate words in the main body of their paper. This standard of care, sadly, seems not to be applied to BibTeX-generated references, making otherwise excellent scientists look disorganized and unprofessional.
CrossTeX has three critical features that address these problems:
CrossTeX is backwards-compatible with BibTeX. You can use a BibTeX database with CrossTeX, and invoke crosstex where you used to invoke bibtex. It is particularly well-suited for use in Computer Science and Engineering, with support for frequently cited publication formats and an extensive library of published articles at top conferences.
CrossTeX can generate HTML as well as LaTeX, and can thus create professional citations for web pages.
Check out the feature list, frequently asked questions, and the tutorial.