For a while I’ve had a blog post about BibTeX on the back burner, but now I don’t need to write it because Dan Wallach has done so. It’s a great post, but for emphasis I’ll repeat a handful of points that I tell students (and would like to tell authors of some papers I review) over and over. BibTeX users must:
- Repeatedly proofread the BibTeX output. (For important papers I usually proofread to fixpoint: keep reading and editing the paper until I can read the entire document without wanting to make a single change. This is incredibly time-consuming.)
- Never assume a bib entry found online is accurate.
- Strive for consistency in fields listed, abbreviations used, etc. across all entries used in any single paper.
- Proofread yet again every time the bibliography style is changed.
- Include fields that help people identify and track down the cited paper, and (mostly) leave out fields that do not.
Ah, it feels better getting that off my chest. Thanks, Dan, for spelling out all the details!
4 responses to “Life With BibTeX”
I’ve rarely paid attention to any bibliography I’ve read except to look up the title and list of authors. I’ve simply never had a scientific need to know, for example, whether a paper was presented in Las Vegas or San Francisco. I admit that has led me to be lazy in my own bibtex entries. Some people are habitually late for meetings or habitually unresponsive to email. Although I am neither, I am (as I suspect lots of other authors are) habitually lazy about the bibtex fields beyond author list, title, and name/year (and maybe volume/pages) of conference/journal. It might be better for your sanity to learn to tolerate us than to think you can change us. 🙂
I could understand the position, “include everything”, especially for the purpose of making automated citation indices more accurate. But Dan advocates leaving out lots of data, such as the LNCS volume and page numbers. I’ve also been directed by referees to include the phrase “to appear”, while Dan directs us to leave it out. I’m not saying he’s wrong and the other referees are correct, just that these are really personal preferences, rather than fundamental Laws of Science, that are probably not worth losing sleep over or chastising graduate students about.
Much more important than conference addresses or title capitalization, for instance, is the admonition never to use “et al.”, since this robs the other authors of credit and has a small but real effect on scientific careers. Inconsistent capitalization of titles has never affected a single career. Similarly, however much time is spent formatting the *existing* entries, at least 50 times as much effort should go to ensuring that the list of entries is *complete*, i.e., that relevant prior work is cited. There are a few pointers in Dan’s list, such as making sure to include all authors and distrusting auto-generated bibtex entries, that seem to make a real difference in a bibliography’s influence on the world, but they are sort of scattered randomly throughout the pet peeves.
Hi Dave, I just looked at the bibs for the top two pubs on your web page and they are not bad, probably around 80th percentile of overall goodness for all bibliographies I look at. So basically you are not the target of Dan’s or my rants.
The reason I’m anal about my own (and my students’) bibliographies is that when I’m reading someone else’s paper, if one aspect is badly done, I assume that other aspects probably were not done carefully either. There are probably many reviewers like this. Same goes for notation used in proofs, formatting of graphs, spelling of words, etc. A paper is supposed to take a research result and package it up in an attractive and understandable form and doing at sloppy job at the paper detracts from its overall appeal.
… if one aspect is badly done, I assume that other aspects probably were not done carefully either.
This is a good point. I guess I just don’t spend time reading bibliographies (once I have the paper’s title), but I do spend time reading proofs and getting annoyed by minor notational things that don’t affect the science. To this day I’ve never understood how a self-respecting mathematician can use teensy-tiny little parentheses on each side of a really tall formula with fractions and summations.
I think we all have our pet peeves, Dave, I’m not sure how bibliographies became one of mine. But once a pet peeve is acquired it’s difficult to exorcise.