There is an important and unfortunate caveat for managing citations that I bring up because this refers to IEEE. In IEEE, ACM, and many other computer science and electrical engineering venues, it is often the case that there are strict page limits and the citations are counted toward that limit. This happens mostly with conferences, but sometimes even with journals.
Because of this, I often find it necessary to maintain both a "master" reference database and an ad hoc "abbreviated" copy of the database used for a particular paper. In the master database, I keep the full everything in all its bibliographic detail to the best of my ability. In the abbreviated copy, conference and journal names get shortened as necessary to purge the bibliography of widow and orphans
and extract those last few lines necessary to get the text to fit. It's a nasty, undignified practice, and I consider it the lesser of two evils if it can preserve technical content without decreasing the ability of a reader to locate citations.
The ability to find the citation, though, is sacrosanct. Authors, title, year (and volume and issue if available), must never be trifled with. As such, the shortening targets that tend I use, in order, are:
- "Proceedings of the..." can always go, as can the location of a conference.
- "International Conference on" goes to "Int'l Conf. on" and then vanishes
- Most EE/CS conference have acronyms: an exceedingly well-known conference can be entirely replaced by its acronym.
Thus, for example, "Proceedings of the 23rd International Joint Conference on Artificial Intelligence, Beijing, China" can ablate all the way down to "IJCAI", saving 1-2 lines in typical IEEE format bibliography. It's an ugly business, and maybe I shouldn't admit to it out loud on the internet, but it's a useful practice that I still find ethical as long as the spirit of readily locatable citation is preserved.