Closed joernwilms closed 4 years ago
It's not impossible in principle, but consider this scenario:
smith:2011
), and by Jones from the same year (gets key jones:2011
). You cite bothsmith2011a
If at step 1. I would have given them citekeys smith:2011a and jones:2011a that looks weird (and ugly, if you ask me). There's nothing to disambiguate, so why add the postfix?
If I'd do business as usual in step 1, but in step 2 change smith:2011
to smith:2011a
before handing out smith:2011b
, you have a broken reference in one of your papers. It could even be a different paper.
One of the major premises of BBT is that keys are dependable, so 2. is not an option. But to implement 1., keys would always get a postfix, even where not required. If that's what you actually want (always), I can think about how that would be achieved without disrupting the existing key generation, but it won't be the default.
Hi,
thanks for the fast reply. Yes, I always want the postfix added. I know that at first this sounds weird, but the reason is that I would estimate that 95% of all of my bibkeys need to be disambiguated - I'm talking about a database with about 3000 entries here and in my field it is fairly typical that people have more than one first author publication per year. This means that in the end almost all bibkeys end up having a postfix. So it makes more sense to always have a postfix rather than having one random article per year for each author not to have a postfix.
I guess I am biased by the citation culture in my field where you would cite two papers of the same author which were published in the same year as Smith et al. (2011a) and Smith et al. (2011b), rather than Smith et al. (2011) and Smith et al. (2011a).
That is fairly common I think, but not that you'd also get Jones (2011a)
absent any other papers from Jones.
In any case, you can give https://s3.eu-central-1.amazonaws.com/zotplus-builds/zotero-better-bibtex-1.6.34-travis-2427.xpi a spin if you want. After installation, you will have to go into about:config
, set citekeyAlwaysPostfix
to true
, and restart Firefox. OSX is particularly stubborn about keeping apps running even when you've closed the last window. This setting is only read out on startup for performance reasons (the key generation gets called often) so any time you change it you will have to restart.
(btw, if you are looking for consistency between citation key and citation, if you cite only smith:2001a and smith:2001c, you're going to get Smith (2001a) and Smith (2001b) in your paper -- no way around that)
(or even: if you cite only smith:2001c and smith:2001d, you're going to get Smith (2001a) and Smith (2001b) in your paper)
I can merge if this does what you want.
assuming fixed to satisfaction
Coming back to this issue. I also encountered this problem, but rather because of not liking the BBT format, it is because I just migrated from BibDesk, where the format of my keys was Author:Yearaa, always with two postfixes, e.g.: Smith:1990aa, Smith:1993aa, Smith:1993ab, etc.
Is there a way to force this format in BBT in order not to change all my .tex files?
Thanks!
Jeez, how many papers by that one author do you cite? 😆
Can you join #1554?
This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Hi,
thanks a lot for better bibtex, which makes zotero useful ;-). I have a small little feature request rather than an error: I have a slight (psychological) problem with the bibtex keys generated by better bibtex, in the sense that the disambiguation results in somewhat illogical bibtex keys. What I mean is that I have set my key format to
This results in keys
smith:2011 smith:2011a smith:2011b
etc. Is there a way to start with the character appended to the year already for the first paper of an author for that year? I.e. I would like the above sequence to be
smith:2011a smith:2011b smith:2011c
This would be more logical to me, since I read the labels in my publications to mean that smith:2011a is the first publication by that person in the year, smith:2011b the second, etc. (the current approach confuses the hell out of me for my own publications and the ones I cite often, since I know the sequence in which they were submitted).
Thanks a lot again for this great piece of software!